Changing Health Care in Canada: The Romanow Papers, Volume 2 9781442672833

Citation preview

CHANGING HEALTH CARE IN CANADA The Romanow Papers, Volume 2 Edited by Pierre-Gerlier Forest, Gregory P. Marchildon, and Tom Mclntosh

The second in a series of three volumes presenting a selection of studies prepared for the Romanow Commission, this volume focuses on the process of change in health care and health systems. Bringing together the perspectives of health policy specialists and experienced researchers, the papers in Changing Health Care in Canada discuss such issues as technological innovation, complexity in health systems, management of human resources, organizational control and regulation, and public engagement. In keeping with the mandate of the Romanow Commission, the studies in this volume address urgent public policy problems, offering practical approaches to managing change. At the same time, the studies provide comprehensive analysis and appraisal of the most recent developments and knowledge in the field. PIERRE-GERLIER FOREST is professor of public policy and management with the department of political science at Universite Laval. He currently holds the G.D.W. Cameron Chair with Health Canada. GREGORY p. MARCHILDON holds a Canada research chair in public policy and economic history, and is professor of public administration at the University of Regina. TOM MCINTOSH is assistant professor of political science and a member of the research faculty at the Saskatchewan Population Health and Evaluation Research Unit at the University of Regina.

This page intentionally left blank

THE ROMANOW PAPERS: VOLUME II

Changing Health Care in Canada Edited by Pierre-Gerlier Forest, Gregory P. Marchildon, and Tom Mclntosh

U N I V E R S I T Y OF T O R O N T O PRESS Toronto Buffalo London

www.utppublishing.com University of Toronto Press Incorporated 2004 Toronto Buffalo London Printed in Canada ISBN 0-8020-8618-7

Printed on acid-free paper

National Library of Canada Cataloguing in Publication Romanow papers / edited by Gregory P. Marchildon, Tom Mclntosh and Pierre-Gerlier Forest. Contents: v. 1. The fiscal sustainability of health care in Canada v. 2. Changing health care in Canada - v. 3. The governance of health care in Canada. ISBN 0-8020-8626-8 (set). ISBN 0-8020-8617-9 (v. 1). ISBN 0-8020-8618-7 (v. 2). ISBN 0-8020-8619-5 (v. 3) 1. Medical care-Canada. 2. Medical policy - Canada. 3. Public healthCanada. I. Marchildon, Gregory P., 1956- . II. Mclntosh, Thomas A. (Thomas Allan), 1964- . III. Forest, Pierre-Gerlier IV. Title: The fiscal sustainability of health care in Canada. V. Title: Changing health care in Canada. VI. Title: The governance of health care in Canada. RA449.R65 2003

362.1'0971

C2003-904073-9

University of Toronto Press acknowledges the financial assistance to its publishing program of the Canada Council for the Arts and the Ontario Arts Council. University of Toronto Press acknowledges the financial support for its publishing activities of the Government of Canada through the Book Publishing Industry Development Program (BPIDP).

Contents

ACKNOWLEDGMENTS CONTRIBUTORS ix

Vll

Introduction: Changing Health Care in Canada PIERRE-GERLIER FOREST

3

Fart One: System Change, Organization Change 1 Complicated and Complex Systems: What Would Successful Reform of Medicare Look Like? SHOLOM GLOUBERMAN AND BRENDA ZIMMERMAN

21

2 Financial Rules As a Catalyst for Change in the Canadian Health Care System IAN MCKILLOP

54

3 Governance and Management of Change in Canada's Health System JEAN-LOUIS DENIS

82

Part Two: Change and Providers 4 Planning for Care: Approaches to Human Resources Policy and Planning in Health Care PAT ARMSTRONG AND HUGH ARMSTRONG

117

5 How Do Human Resource Policies and Practices Inhibit Change in Health Care? A Plan for the Future GAIL TOMBLIN MURPHY AND LINDA O'BRIEN-PALLAS

150

vi Contents

6 Changes and a Few Paradoxes: Some Thoughts on Health System Personnel CLEMENCE DALLAIRE AND SONIA NORMAND

183

7 The Influence of Physician-Payment Methods on the Efficiency of the Health Care System MICHEL GRIGNON, VALERIE PARIS, AND DOMINIQUE POLTON, WITH THE COOPERATION OF AGNES COUFFINHAL AND BERTRAND PIERRARD 207

Part Three: Change and the Public 8 Public Involvement in the Development of a Health Care Vision HARLEY D. DICKINSON

243

9 Public Participation and Citizen Governance in the Canadian Health System JULIA ABELSON AND JOHN EYLES

279

10 Practical Strategies for Facilitating Meaningful Citizen Involvement in Health Planning JAYNE RENEE PIVIK

312

11 The Responsiveness of the Canadian Health Care System towards Newcomers ANITA J. GAGNON

349

Acknowledgments

The papers in this volume, and its two companion volumes, were originally commissioned as part of the research activities of the Commission on the Future of Health Care in Canada headed by Roy J. Romanow and were first issued on the commission's Web site during the summer and fall of 2002. From the outset of the commission's work, Mr Romanow made it clear to us that we had a mandate to ensure that the Report's recommendations were based on the best available evidence and analysis. His support for our contention that there were still important gaps in our knowledge about health care in Canada that needed to be filled was unwavering. As, respectively, the commission's Executive Director, Director of Research and Research Coordinator, we were given an opportunity that few academics ever receive, and the commissioner's trust in, and support of, our work on behalf of the commission is deeply appreciated. We want also to acknowledge Steven Lewis's contribution to the development of the terms of reference for the authors and in helping us match the right author with the right research assignment. Morris Barer, Diane Watson, Donna Shields-Poe, and Rob Courchene of the Institute of Health Policy and Services Research of the Canadian Institutes of Health Research managed the often time-consuming independent peer-review process of the papers during the commission's work. The comments provided by the anonymous reviewers contributed greatly to the quality of the papers and their efforts on behalf of the commission are appreciated. Louise Seguin-Guenette provided invaluable services in coordinating the translation and publication of the papers on the commission's Web site and did so without ever losing her sense of humour. Don Moggridge, who reviewed the papers for the

viii Acknowledgments

papers on the Commission's website and did so without ever losing her sense of humour. Don Moggridge, who reviewed the papers for the University of Toronto Press, was especially helpful in assisting us in turning a set of separate studies into three thematic volumes. We also want to thank the staffs of the University of Toronto Press and the University of Ottawa Press, especially Virgil Duff (who acted as our experienced sherpa from the beginning), Bill Harnum, and Lynn Mackay, for their assistance in creating a more permanent legacy for these papers and for the work of the commission. Finally, we want to thank our contributors, who approached their respective papers with enthusiasm and who so clearly appreciated the import of the overall project in which we were engaged. GREGORY P. MARCHILDON PIERRE-GERLIER FOREST TOM MCINTOSH

March 2003

Contributors

Julia Abelson is an assistant professor in the Department of Clinical Epidemiology and Biostatistics and a member of the Centre for Health Economics and Policy Analysis at McMaster University. Hugh Armstrong is a professor in the School of Social Work at Carleton University. Pat Armstrong is a professor of sociology at York University. Agnes Couf finhal is a research associate at the Research and Information Center for Health Economics (CREDES). Clemence Dallaire is an assistant professor in the Faculty of Nursing at Universite Laval. Jean-Louis Denis is a professor of health administration and a member of the Interdisciplinary Research Group in Health at Universite de Montreal. Harley D. Dickinson is a professor of sociology at the University of Saskatchewan. John Eyles is a professor of geography at McMaster University. Pierre-Gerlier Forest is a professor of public policy and management

x

Contributors

in the department of political science at Universite Laval. He currently holds the G.D.W. Cameron Chair with Health Canada. Anita J. Gagnon is an assistant professor at the School of Nursing and Department of Obstetrics and Gynecology at McGill University. Sholom Glouberman is philosopher in residence at Baycrest Centre for Geriatric Care. Michel Grignon is research director at the Research and Information Center for Health Economics (CREDES). Tom Mclntosh is assistant professor of political science and a member of the research faculty at the Saskatchewan Population Health and Evaluation Research Unit at the University of Regina. Ian McKillop is an associate professor in the School of Business and Economics at Wilfrid Laurier University. Gregory P. Marchildon holds a Canada research chair in public policy and economic history, and is professor of public administration at the University of Regina. Sonia Normand is a research associate in the Faculty of Nursing at Universite Laval. Linda O'Brien-Pallas is a professor of nursing at the University of Toronto. Valerie Paris is senior research associate at the Research and Information Center for Health Economics (CREDES). Bertrand Pierrard is a research assistant at the Research and Information Center for Health Economics (CREDES). Jayne Renee Pivik is a doctoral candidate in the School of Psychology at the University of Ottawa. Dominique Polton is Scientific Director of the Research and Information Center for Health Economics (CREDES).

Contributors xi

Gail Tomblin Murphy is an associate professor in the School of Nursing and Department of Community Health and Epidemiology at Dalhousie University. Brenda Zimmerman is an associate professor in the Schulich School of Business, York University.

This page intentionally left blank

CHANGING HEALTH CARE IN CANADA

This page intentionally left blank

Introduction: Changing Health Care in Canada PIERRE-GERLIER FOREST

Of all the major public services, the health care system is without doubt the one that has been subject to the most powerful, if not to say the most brutal, forces for change. Every day, there are new ways of doing things, new ideas, new medications, new machines and instruments, new problems, and new organizations. It could even be said that there are new users, since the patients themselves have changed over the years and are more educated, better informed, more autonomous, and more demanding. The stability and continuity of the institutions responsible for the funding and governance of our health care systems do not seem to be guaranteed, and as a result create anxiety about the future. Of course, as Louis Omnes puts it so well, 'what is new is not always modern' (2001, 2). There is a danger that we will not know how to deal with change, but it is also possible to exaggerate the problems. Many innovations can be easily incorporated into current practices and may simply improve them: an instrument that facilitates the work of the surgeon will not necessarily change the roles of the professionals or the hierarchy of personnel present in the operating room, and it may even make life easier for them. In fact, the innovations that are disturbing are those that go beyond equipment and technology, to disrupt how we organize work, if not our entire understanding of health and illness. One only has to think of the recent explosion in home care, which has forced all the actors in the health care system, including the patients and their friends and families, to redefine their responsibilities and readjust their relationships with each other. Indeed, when we

4 Pierre-Gerlier Forest

expect a family member to monitor a patient's vital signs or a friend to administer medication intravenously, the boundaries between health professionals and laypersons may have to be renegotiated. Clearly, to put things simply, certain 'non-material' changes carry much more weight than 'hard' technologies. Sometimes the pace at which innovations are adopted is a greater source of tension than the innovations themselves. In almost all cases, furthermore, it is clear that there is not enough adjustment time available to inform or train adequately the people involved, whether these be professionals, patients, or even administrators responsible for implementing the changes. We therefore we see strong cognitive resistance, in addition to the problems inherent in putting the actual changes into effect. In their chapter on the health professions, Pat Armstrong and Hugh Armstrong aptly observe that an ageing labour force does not always easily accept changes to its habits, even when these seem to be minor, such as the introduction of more advanced equipment or new medicines. What is more, the structures of the organizations and systems are now changing faster than the equipment, bringing in their wake a complete renewal of knowledge and skills. This is quite different from the introduction of a new method for administering an analgesic, and it affects everyone, irrespective of age or experience. As for changes in governance - the new ways of sharing authority and responsibility - the fluidity of political processes could undo today what we built only yesterday with a great deal of effort. This is clearly shown by the history of regionalization over the last decade. Those who believed in and invested in the very 'local' structures of the early years must now, in several provinces, adapt to large organizations in which community participation is much more difficult. Mastering change is, of course a constant problem for human societies, and it was a concern long before the appearance of scientific medicine or universal health insurance. We know that cultural and social arrangements in 'traditional' or 'primitive' societies studied by anthropologists are often intended precisely to protect their members from change. Everything from religion to political structures is constructed to ensure that institutions and practices continue, even if this means using the most extreme forms of coercion (Clastres 1974; Lapierre 1969). Such arrangements are based on profound pessimism, since these societies see any change as a loss of identity, capability, or security, something that in the long term threatens the very existence of the community.

Introduction 5

We may think that we are far away from those realities, but during the consultations of the Commission on the Future of Health Care in Canada, many who made submissions asked the commissioner, Roy Romanow, not to touch the Canada Health Act or any other aspect of our public medicare plan, saying that any modification would lead inevitably to the destruction of the Canadian health care system (Commission 2002a). From a broader point of view, as pointed out recently by Candace Johnson Redden (2002), the emerging concept of a right to health that makes access to hospital care and medical treatment the sole raison d'etre of the health care system also represents a formidable obstacle to any reform. This point of view says that the Canadian health insurance plan must remain the same as it was in the 1950s and 1960s, that basic services must be protected at all costs. When it prevails, this conservative position also locks in the relationships among the health professions, as if the most important thing was to preserve the stratification of the professions, whether in the hierarchy of prestige, distribution of income, or participation in power (Senate Standing Committee 2002). The same position would lead to the exclusion from the domain of health services practices that are today central and necessary, such as drugs and community action. Fortunately, the fact remains that our societies usually welcome change as an integral part of the human experience. In many established approaches to economics or politics, development, expansion, and adaptation are even considered to be 'natural' processes (Jacobs 2000). There is therefore no question of avoiding change; rather, it is to be welcomed as a positive principle, an impetus for creation rather than a destructive force, as Schumpeter predicted (1939). Nonetheless, we wish to control it, to channel it, and to use its power the way we use other natural forces, in the service of progress, prosperity, and well-being. This latter conception is usually dominant in the health care system, where it takes many forms. In its most common form, it is an active and enterprising mental attitude, which considers any change, good or bad, as potentially auspicious, as an opportunity. The 'four strong winds/ to borrow Michael Decter's (2000) image, are purifying winds that blow down the old structures and sweep away the obstacles of the past. Then we can start over again in their wake. When the commission made the management of change one of the main themes of its research, it declared itself to be in favour of this approach. It seemed important to reach a better understand of how to control change, since it is inevitable. But there was an additional moti-

6 Pierre-Gerlier Forest

vation: to learn how to instigate change that would serve objectives such as justice, solidarity, and health. Too many previous commissions and task forces had neglected this aspect in their recommendations, as if the solutions would be adopted through sheer force of reason, through the mere fact that they were obviously correct; as if it were not necessary to persuade, to mobilize, to act; as if there were no professionals to be convinced, no organizations to be created, no policies to be steered; and as if there were no culture and no society. The reform of primary health care, which is covered in chapter 5 of the final report of the commission, is a good example of how discussion of change influenced the recommendations of Roy Romanow and his team (Commission 2002b). It should be remembered that there is probably no undertaking in the field of health policy that enjoys a broader consensus than the transformation of traditional approaches to the delivery of primary care in order to establish a true primary health care system, one that is characterized by a global vision of health needs and continuity of health care services, close to patients and the communities in which they live. However, in spite of the sincere efforts of many governments across Canada and in spite of many convincing initiatives, no one seems capable of making the transition from model to reality - at least not on the scale of the health care system as a whole. We could go even further and say that everyone seems to be aware of the obstacles that need to be removed, from the rigidity of professional roles and the lack of information to inflexibility in funding and the false expectations of users, but that no one really knows what to do to direct efforts towards a common goal. To speak in the terms of organizational theorists, the various decisions necessary for change have no 'inter-dependencies' among them, and all the players feel justified in pursuing their own objectives, being concerned first and foremost with their own interests (Lorino 1989). Given this situation, it is common to talk about resistance to change and call on public authorities to take the necessary measures to break through the opposition. But in fact, this thesis is not credible. How could the problems be the result of the slowness of change in a field in which everything is in movement (Hutchison, Abelson, and Lavis 2001)? If that were the case, all that would be needed would be a few incentives to 'speed things up,' which has often been tried, without much success. It even seems as if everything has been attempted: changing the remuneration of doctors, consolidation of health organizations, preferred funding for certain models of care, all more or less in

Introduction 7

vain. But the problem lies rather in the absence of a common strategy, and beyond this, in the absence of a shared vision of the action to be taken, which is an essential condition for social change. In fact, the consensus on primary health care is perhaps only apparent. It masks differences with respect to situation, methods, and even orientations. Therefore, the worst approach to try to bring about changes under such conditions would certainly be to look for common denominators, since they will inevitably be too limited to have any effect on the health care system. On the other hand, as long as the overall dynamic guarantees success for everyone, pluralism is a better option, since it makes it possible for each model to be made attractive to professionals and patients: family medicine groups, health care networks, integrated systems, and health management programs. But this is not enough. We also need an image of the ideal towards which the system should be moving, which could be considered a basic plan, something that will give meaning to action and help us persevere in our efforts (Williams 1981). The final report of the commission tried to lay the foundations of such a plan, as well as suggesting concrete means to develop it. This approach nonetheless has its limits. As pointed out by several of the authors whose studies are collected in this book, change is first and foremost a social movement. It may be that it was possible for Mr Romanow to adopt an open, 'plural' attitude towards primary care because the health care system is already moving in this direction. What should be done when this condition is not present? When should we go against the current? Should we do this when the interplay of politics and various interests threatens not only the balance of the system in the short term but also its development? Of all the texts that appear here, the chapter by Glouberman and Zimmerman is no doubt the one that answers these questions from the broadest perspective, that of the paradigm of complexity. They distinguish, according to the degree of complexity of the system being examined, among three approaches to change and social action. A 'recipe' or a program will be enough for the transformation of simple systems change in this case occurs in response to a brief sequence of decisions and instructions, like a prescription written by a doctor or a computer program. Complicated systems, in contrast, change under the effect of a succession of many sequences, always consisting of decisions and instructions but ordered according to a more or less ambitious plan, which corresponds to an extended horizon for action. The authors give

8 Pierre-Gerlier Forest

the example of acute care practices that follow precise algorithms, such as certain surgeries or emergency care. Complex systems also have an extended horizon, a long timeline, but they cannot be changed from the outside. The players in such systems must themselves give meaning to change by allocating their resources and by seeking themselves to negotiate or renegotiate their relationships. Home care or primary care, as we have already mentioned, has these characteristics. But this reasoning can be extended to the health care system as a whole. The integration of efforts and perspectives favourable to change cannot be done without 'complex thinking,' enriched by history and culture as much as by science and reason. This being said, health organizations and systems are not organizations and systems quite like others, even from the point of view of complexity. They are professional organizations and systems, and in most developed countries, including Canada, they are also public organizations and systems. In other words, health organizations and systems are distinctive, but this is based not so much on their intrinsic characteristics - their degree of complexity, for example - as on the constraints imposed on them by broader structures, those of professional relationships and those of administrative and political standards. In the two cases, the dynamics of change are different. Autonomy is one condition for change, for example, but it does not receive the same level of attention in the professional world, where it is among the prime values, and in the administrative world, where there is a tendency to impose reforms from above. In the same way, information is a key element in any reform effort, but it is not as high among the priorities of managers, who are concerned about the development of their 'tool kit/ or of health professionals, who often neglect it in their decisions (Saskatchewan 2000). The blending of these two realities is difficult and it requires not only well-designed policy, which is a minimum requirement and appropriate instruments, but also particular vigilance with regard to the conditions of implementation. These three requirements - design, instrumentation, implementation - are therefore at the heart of McKillop's presentation, in which he examines the effects of financial rules. One of the most fascinating aspects of this discussion is his dissociation of the mechanisms by which health organizations or systems are managed from the other characteristics of these organizations or systems. For McKillop, the transformation of management control mechanisms is a necessary precondition to any real change in the long term. The examples he pro-

Introduction 9

vides focus on the budget processes and their effects. The commission has learned a lot from this work, first, because it makes it possible to give a precise, operational meaning to the concept of predictability the capacity to commit to changes with long-term effects - stemming from well-defined budgetary certainty; and second, because it clearly raises the question of initiative: a health care organization or system changes because the conditions in which it operates have been modified at the outset to encourage that change. In other words, there is no change without direction. For co-existence and solidarity, the various actors in the health care system require a certain stability, a relative regularity, in their relationships. In his summary of his research on organizational change, Denis shows that it is essential to take into account this need for stability and regularity when introducing profound changes. This view is, in a way, the extension of McKillop's, but from a more inclusive sociological perspective. The three examples discussed by Denis - regionalization, experimentation, and organizational leadership - also correspond abstractly to the classical division made in sociology between social relationships, social behaviour, and institutions, and Denis strongly highlights the differences in rhythm and preparation at each of these levels. Denis's thoughts on governance, while they may not provide a better understanding of the nature of these differences, since the scope of the experimentation is limited, at least demonstrate the essential dual function of maintaining order and stimulating innovation: progress depends necessarily on a power capable of maintaining the rules of the social game while at the same time overseeing the building of a renewed system. This complex of problems is familiar to everyone who is concerned about the health professions. Two of the commission's four papers on human resources (Tomblin Murphy and O'Brien-Pallas, Armstrong and Armstrong) call for the creation of a national governing body responsible for directing the reform efforts. A third study (Dallaire and Normand) points out errors in judgment by senior managers who placed on the professionals the burden of initiating and sustaining the process of transformation of the health care system, when, in fact, it was their own fundamental responsibility. Basically, all these papers examine the role of the institutions that are supposed to oversee and orient the health professions in the interests of the system as a whole, for example, by promoting a better integration of skills or a better allocation of resources. The connections between training, regulation, and planning

10 Pierre-Gerlier Forest

were formerly under the direct responsibility of each of the professions; the fragmentation of responsibilities and profound changes in the organization of services put an end to this model, and there is no point in trying to return to it (Fooks et al. 2002). The conclusion therefore is, on the one hand, that a new model needs to be invented and implemented, and on the other, that the principle of organization that governs this new model cannot be applied without leadership that is capable of providing cohesive common action in favour of change. Of course, there is no lack of voices asserting that the best form of coordination is the one that the actors adopt for themselves - this is the idea that lies behind the market model and, more broadly, behind all the approaches that target the individual and his or her action rather than the institutions. The synthesis produced by Grignon and his colleagues at the Centre de recherche et de documentation en economic de la sante (CREDES) makes it possible to do an evaluation of these concepts through a systematic examination of conclusive data on medical compensation. Their conclusions reinforce the views of professionals, who are more sensitive to 'local' interactions with their colleagues than to incentives of an individual nature, in particular monetary benefits. They also highlight the structuring role of institutions, which alone can achieve and maintain a plurality of coordination methods training, remuneration, regulation, and so on. Attention to the role of health professionals is certainly not a new factor in health policy. On the other hand, one of the big 'innovations' of the Commission was to accord the same importance to citizens, patients, clients, and other taxpayers in its deliberations on strategies for change. The Citizens' Dialogue, a broad consultation process set up during the first month of the commission, showed clearly that Canadians strongly support initiatives to change the health care system, both clinically and organizationally, and that they see themselves as full participants in this process (Maxwell, Rosell, and Forest 2003). The subsequent studies, of which there are three prime examples here, have made it possible to further this discussion and focus on the conditions for public participation that are required to channel this force in the public interest. What is particularly interesting in Dickinson's study is the author's insistence on associating the question of the future of health care organizations and the health care system with the question, just as vital in our societies, of the future of democratic institutions. Most experts in the health sector pose the problem of public participation in a frame-

Introduction 11

work that is rather narrow, that is, assistance in decision making and the circulation of information within the health care system (Abelson et al. 2003). But in reality, just as health goes far beyond the health care system, public participation is much broader in scope than the management of a hospital or regional health authority. It expresses citizenship values such as a sense of belonging to a community and a feeling of collective responsibility towards common institutions. Basically, when Canadians make a strong link between citizenship and health services - a link that Mr Romanow chose to highlight in his final recommendations - they are not so much indicating their concern for a particular orientation in the short term as expressing a profound conviction that the public system belongs to them and that it reflects who they are or want to be. Political decision makers and even health professionals are, in short, only agents. Pivik's chapter on public involvement is in keeping with this point of view. In many respects, the author even seems less concerned with the effects of involvement than with its intrinsic moral value: it is seen as the only way to associate disadvantaged or marginalized groups and individuals - not to mention all the strata of the population that are still excluded from administrative processes controlled by the elites - with the decisions that concern them. One of the obvious interests of this pragmatic approach is that it creates clear conditions for the establishment of participatory structures that are open to the public. But Pivik, unlike Abelson and Eyles, fails to examine the results of these processes, and, unlike Dickinson, she neglects as well the ultimate goal of these processes, namely, the renewal of the health care system. Consequently, she ends up overemphasizing the effect of every consultation and even the impact of the public on decisions. We are left with the impression that the symbolic aspects of involvement are underrated here, in favour of a position of principle promoting the control of health care organizations by their users. Abelson and Eyles avoid this pitfall with a great deal of finesse in their own paper, suggesting that the tree should always be judged by its fruits. It is true, in certain cases - which have as much to do with the approaches chosen as the expectations - that experiences of public involvement and consultation contribute to favouring 'good' changes, in the sense that they are more appropriate to the needs of the population and more respectful of its values. But there are also cases where involvement and consultation, far from changing things in the public interest, lead to the reinforcement of narrow interests and place con-

12 Pierre-Gerlier Forest

straints on positive changes, because suspicion or ignorance may be expressed as readily as generosity or a willingness to change. This problem points to how important it is to emphasize the socialization of citizens or users. Standards and values do not necessarily exist prior to participation; they are exposed, revealed, and sometimes internalized during it. Abelson and Eyles's study is one of the rare contributions to the work of the commission that does not deal with the Canadian population as an undifferentiated mass, or, conversely in terms of principles but with the same results, as a huge collection of individuals without any affinities or common interests. In reality, Canadian society is marked by divisions (of language, culture, geography, and generation) that give rise to and maintain multiple and sometimes conflicting identities. A given individual's relationship to health and illness is profoundly marked by traditional traits such as gender, ethnic group, social class, and age. Without people being socialized to common values of citizenship - including solidarity, equity, justice, and respect for public institutions - a system as complex and as essential as the health care system would no doubt be unable to resist the threat that permanently hangs over it, the temptation to 'go it alone,' at the family, community, or provincial level. In other words, it is not possible to change a large-scale social system such as the health care system without first getting as many people as possible on board. It is not clear to what extent the capacity to win support is directly associated with the capacity to provide adequate services, but we can assume that there is a connection. We see this tested every day, for example, in cultural communities with large numbers of recent immigrants, which are the subject of the last study in the book. In this paper, Gagnon brings out clearly the subtle connections that exist between the socialization of immigrants, the coping skills of professionals and health care organizations, and what may be called the 'political development' of the health care system. What is taking place is much more than a simple demographic change, much more than the addition of new languages and ethnocultural traditions to a list that is already long. In fact, a health care system that is capable of learning new ways of doing things from the very people to whom it provides services is a living system, one that remains directly engaged with the realities on the ground. The big questions that emerge from all these analyses are the following. What is the place of knowledge and learning among the determi-

Introduction 13 nants of change, which are at the heart of building a better health care system? How can we sort out the good changes from the bad changes? How can we acclimatize ourselves to the necessary changes? How can we learn to do things better, to do things differently? Most of the papers in this volume emphasize the role of institutions in the transformation of the health care system, except for the one by Glouberman and Zimmerman, which accords a decisive role to major social and cultural change, and the one by Grignon and his colleagues from CREDES, which places interests on an equal footing. In the other cases, the creation and imposition of the standards that are indispensable for the adaptation of the health care system and health organizations is less a matter of expectations or values than of the institutional creation of new rules. For authors such as Denis, Dickinson, and Abelson and Eyles, one of the essential roles of institutions is to promote learning. When the appropriate conditions are met, health organizations can learn to adapt and citizens can familiarize themselves with the problems of allocation of resources. McKillop's preference for systems that steer change on the basis of information and Pivik's confidence in the mechanisms of citizen involvement are also related to this approach. Finally, as we have seen clearly, one of the key messages from human-resources experts is precisely that we must affirm the importance of collective organizations of planning and management. As I have already said, there is no change without direction. Even given the relative unanimity among these authors, it should not be forgotten that other explanations were possible. Many of the classic studies on the history of the Canadian health care system, from Malcolm Taylor to Carolyn Tuohy, have highlighted the role of ideas and images. In fact, instead of the emphasis being placed on the imposition of a new public policy regime, and, at the same time, on the imposition of the changes necessary for modernization, it could have been placed on the construction of the system itself, given the complex interplay of relationships and power among health professions, political parties, and senior managers. The apprehension expressed by Maioni (2003), who is concerned about the failure of the Romanow report to provide precise indications on how to proceed, is based on this more traditional vision of public policy, in which the uncertainty associated with negotiation between actors is synonymous with rigidity and even regression. However, the authors whose work is compiled in this volume seem to have confidence in a relatively open collective-learning process, as long as the institutions responsible for governance are well designed:

14 Pierre-Gerlier Forest

decentralized, flexible, adaptable to local social realities, and open to the involvement of actors in resolving their differences. There are examples that support this point of view, including systems in which pluralism and bitter disputes among interest groups are more obvious than in Canada, as Hackey (1998) has shown so well for the United States. It is difficult to believe that a system that is as regulated as ours would necessarily be more resistant to the emergence of new standards than a system that is left to itself. The only thing that is important is that the regulation mechanisms provide sufficient means to overcome inflexibility and habits, including the financial and symbolic resources that are indispensable to the development of legitimacy. But that is not the crux of the matter. Although this is not always said openly, institutions tend to favour a slow, gradual approach to social change. A long way from the initial 'Big Bang' - the creation of the public hospital insurance plan - they are concerned with preserving the health care system and defending what has been achieved, through the continuity of policies and stable social relationships. This reformist bias explains why some observers are tempted by more authoritarian approaches, in which the changes that are considered necessary are simply imposed by the political authorities. On the one hand, there are those who claim that persuasion and negotiation will overcome resistance; on the othe, there are those who look back to the great leaps forward of the 1950s, 1960s, and 1980s and call for quick, authoritative action - Monique Begin's fight to get the Canada Health Act passed has, from this point of view, acquired the status of a veritable paradigm. The issue of primary care once again provides a good illustration of the deep divide that still exists. From an institutional perspective, which is largely that of the Romanow Commission, the development of primary care will come through contractual management among the actors, all of whom are full partners: patients, health professionals, local communities, various associations, and groups and individuals in first-line organizations. This promotion of pluralism is pragmatic. While it must be acknowledged that the partners will change the initial project, it is hoped that the process will in turn have an effect on them and that they will also be changed by taking part in it, gradually abandoning their initial positions based on their own interests and the desire to maintain their privileges. From a more 'ideological' perspective, on the other hand, calls are heard for more coercive measures: universality of salaries for doctors, mandatory registration of patients,

Introduction 15

penalties for holdouts. Some recommendations even go so far as to demand a radical rethinking of the division of constitutional jurisdictions in the area of health, usually in favour of the federal government, in the hope that a champion will appear to force the naysayers to accept the logic of progress (Lewis 2002; Rachlis 2003). Ironically, these debates highlight the fact that there is a substantial deficit in our knowledge of policies and, in particular, in our understanding of the instruments of change. The big successes of the past are not sufficient proof of the validity of radical approaches. It is too easy to ignore the failures, the false victories, and the setbacks, and to focus on a few decisions that turned out well. On the other hand, the growing consensus in favour of reforms does not mean that the gradual, careful strategies have any better chance of success. If the choice was not theoretical and the consequences far in the future, would the consensus be as broad? To be honest, we have to admit that the question of choice between the two approaches to change has not been resolved, even though Mr Romanow's 'wager/ in the name of realism and feasibility, has been to lean towards the institutional solution - what Denis calls progressive radical change. If we consider the new forms of organizations and delivery of care that could emerge from deep changes in the health care system, the studies in this book barely go beyond the horizon of the reforms announced during the last two decades: greater integration, participation, and accountability. But they will help us look closer at the difficult, fundamental questions. Attention to the dynamics of social change, for example, may make it possible to promote structures of governance and approaches to leadership that reduce conflict and encourage cooperation. The strictly political dimension is not absent, but it is far from the optimistic voluntarism that imagines that any progress in knowledge will logically lead to equivalent progress in approaches and institutions. Such a positive result is certainly possible because the studies presented here bring together, in a thoughtful and systematic way, the best knowledge available. But beyond the image they provide of the current situation, their interest is in the application of knowledge to real problems of public policy, in real life and in real time. It is clear when we reread the mandate of the Commission on the Future of Health Care that these are the very problems Roy Romanow was asked to solve, on behalf of each and every one of us. Thank you to everyone who contributed to this effort.

16 Pierre-Gerlier Forest REFERENCES Abelson, Julia, et al. 2003. Deliberations about deliberative methods: issues in the design and evaluation of public participation processes. Social Science & Medicine 57(2): 239-51. Clastres, Pierre. 1974. La Societe centre I'Etat: Recherches d'anthropologie politique. Paris: Editions de Minuit. Commission on the Future of Health Care in Canada. 2002a. Shape the future of health care. Interim Report. Ottawa: Commission on the Future of Health Care in Canada. - 2002b. Building on values: The future of health care in Canada. Final Report. Ottawa: Commission on the Future of Health Care in Canada. Decter, Michael B. 2000. Four strong winds: Understanding the growing challenges to health care. Toronto: Stoddart. Fooks, Cathy, et al. 2002. Health human resource planning in Canada. Summary Report. Ottawa: Commission on the Future of Health Care in Canada. Hackey, Robert B. 1998.RRethinking health care policy: The new politics of state regulation. Washington, D.C.: Georgetown University Press. Hutchison, Brian, Julia Abelson, and John Lavis. 2001. 'Primary care in Canada: So much innovation, so little change.' Health Affairs s 20(3): 116-31. Jacobs, Jane. 2000. The nature of economics. Toronto: Random House. Lapierre, Jean-William. 1969.LLe pouvoir politique, 2nd ed. Paris: P.U.F. Lewis, Steven. 2002. The bog, the fog, the future: 5 strategies for renewing federalism in health care, CMAJj 166(11): 1421-22 Lorino, Philippe, 1989. LL'economiste et le manageur: Elements de micro-economie pour une nouvelle gestion. Paris: Editions la Decouverte. Maioni, Antonia. 2003. Romanow: A defence of public health care, but is there a map for the road ahead? Policy Options/Options political 24(2): 50-3. Maxwell, Judith, Steven Resell, and Stone-Gerlier Forest. 2003. The citizen's dialogue on the future of health care in Canada: Giving citizens voice in health care policy. BMJ 326 (7397): 1031-3. Omnes, Louis. 2001. Crise et management, atouts et revers de la modernite. Intervention au Club Perspectives Public Prive. Paris: Service de presse de la FIEHP. Available at http://www.clinique-privee.com/fiehp/presse/C3P/ omnes.pdf. Rachlis, Michael M. 2003. The federal government can and should lead the renewal of Canada's health policy. Ottawa: Caledon Institute of Social Policy. Redden, Candace Johnson. 2002. Health care, entitlement, and citizenship. Toronto: University of Toronto Press. Saskatchewan. Commission on Medicare. 2001. Caring for medicare: Sustaining a quality system. Regina: Commission on Medicare.

Introduction

17

Schumpeter, Josef. 1939. Business cycles: A theoretical, historical and statistical analysis of the capitalist process. New York: McGraw-Hill. Standing Senate Committee on Social Affairs, Science and Technology. 2002. The health of Canadians - The federal role. Final Report, Volume Six: Recommendations for reform. . Ottawa: Standing Senate Committee on Social Affairs, Science and Technology. Williams, Bernard. 1981. Moral luck: Philosophical papers, 1973-1980. Cambridge: Cambridge University Press.

This page intentionally left blank

PART ONE SYSTEM CHANGE, ORGANIZATION CHANGE

This page intentionally left blank

1 Complicated and Complex Systems: What Would Successful Reform of Medicare Look Like? SHOLOM GLOUBERMAN AND BRENDA ZIMMERMAN

While virtually everyone agrees that there is trouble in the Canadian health care system there is little agreement on the nature of the trouble. Pollsters and health-related surveys indicate that something is wrong, but, depending on the survey orientation and the kinds of questions asked, many different sources are identified. Solutions seem to be polarized along ideological lines when left- and right-wing think tanks offer their preferred solutions. They become polarized along professional lines when doctors, nurses, and other health care providers offer their advice. Even academics have some trouble maintaining independent scholarly views in the face of what many of them see as assaults on their positions. Some argue that the problems are the polls themselves rather than the health care system. Added to this melange of confusion are periods of mass hysteria fanned by economic projections of doom: if things continue on their current path, we will be bankrupt or die impoverished or both. Moreover, this situation is not unique to Canada. Other countries in the English-speaking world have had similar histories for the past decade. The British National Health Service (NHS) changed its orientation from right to left, from 'managed competition' to 'collaboration/ the day after the transition from Tory to Labour. The American system rushed into vertically integrated health maintenance organizations with mixed results. New Zealand assumed the role of guinea pig for extreme experiments in entrepreneurial health care with little apparent success (Government of New Zealand 2002). In this paper, we argue that most of these approaches to change are

22 Sholom Glouberman and Brenda Zimmerman Table 1.1 Simple, complicated, and complex problems Following a recipe

Sending a rocket to the moon

Raising a child

The recipe is essential

Formulae are critical and necessary

Formulae have a limited application

Recipes are tested to to assure easy replication

Sending one rocket increases assurance that the next will be OK

Raising one child provides experience but no assurance of success with the next

No particular expertise is required but cooking expertise increases success rate

High levels of expertise in a variety of fields are necessary for success

Expertise can contribute but is neither necessary nor sufficient to assure success

Recipes produce standardized products

Rockets are similar in critical ways

Every child is unique and must be understood as an individual

There is a high degree of The best recipes give good results every time certainty of outcome Optimistic approach to problem possible

Optimistic approach to problem possible

Uncertainty of outcome remains Optimistic approach to problem possible

based on a rational-planning approach with expert-driven design conceptions of strategy (Mintzberg et al. 1998,5). We also argue that health care and the systems within which it is delivered are best understood as complex adaptive systems (Begun 1994; Priesmeyer and Sharp 1995; and McDaniel 1997). The assumptions underlying rational planning are inconsistent with complex adaptive systems. Hence policies and strategies based on it can have significant unintended consequences when applied to complex adaptive systems (Zimmerman 1999). Although most of the experts and advisers have recognized that the health field and its problems are not simple, they do not exhibit an adequate understanding of the theoretical frames of complex systems and how to intervene in such systems. We begin by introducing a preliminary distinction between simple problems, complicated problems, and complex ones. Table 1.1 illustrates the distinction and identifies some of the characteristics of each type of problem. Simple problems like following a recipe may encompass some basic issues of technique and terminology, but once these are mastered, following the recipe carries with it a very high assurance of success. Com-

Complicated and Complex Systems 23

plicated problems contain subsets of simple problems but are not merely reducible to them. Their complicated nature is often related not only to the scale of a problem like sending a rocket to the moon, but also to issues of coordination or specialized expertise. Complicated problems, though generalizable, are not simply an assembly of simple components. Complex problems can encompass both complicated and simple subsidiary problems, but are not reducible to either (Goodwin 1994) since they, too, have special requirements, including an understanding of unique local conditions (Stacey 1992), interdependency (Holland 1995) with the added attribute of non-linearity (Lorenz 1993), and a capacity to adapt as conditions change (Kauffman 1995; Kelly 1994). Unavoidably, complex systems carry with them large elements of ambiguity and uncertainty (Wheatley 1992) that are in many ways similar to the problems associated with raising a child. Despite the uncertainty associated with complexity, all three kinds of problems can be approached with some degree of optimism: we do look forward to raising a child despite the complexity. Our contention is that many health care experts implicitly describe complex problems as complicated ones and hence employ solutions that are wedded to rational-planning approaches. These often lead to inappropriate solutions because they neglect many aspects of complexity. We are reminded of the old joke about the drunk who is stumbling around near a lamp-post. He is asked what he is doing and says that he is looking for his car keys. 'Oh, where do you think you lost them?' 'Down the block near my car/ he says. 'So why are you looking for them here?' 'Because the light is better/

The sophistication of our models, theories, and language for complicated problems can be as seductive as the lamplight. They provide better Tight' and clarity and yet can lead to investigations that are illequipped to address complex adaptive systems. Last year Canadians were shocked to learn that our health care system was ranked thirtieth in the world by the World Health Organization (WHO). Most of us can remember when medicare was a cornerstone of our social policy and a key aspect of our identity: 'Canadians are Americans with no guns and free health care/ How have things come to change so much in Canada? We will explore this ques-

24 Sholom Glouberman and Brenda Zimmerman

tion and a few others that come to mind. 'What can we learn from other ways of thinking about health care systems?' 'How can we once more be proud of a viable well-functioning health care system?' 'What would that system look like?' We will try to struggle with these questions closer to where the problems have arisen even if there is less light there. We will use ideas from 'Complex Adaptive Systems' theory to provide fresh accounts of how our system deteriorated and also to describe what a repaired health care system might look like. We hope that we can also gain some insights into how we might get there. We will use two case studies as we proceed: the rise of France to the top of the WHO ranking, and the unexpected Brazilian response to the HIV/ AIDS epidemic. How the Canadian Health Care System Came to Its Present State In spite of many experts' stated recognition of the complexity of health care, we contend that many of the problems associated with medicare result from the lack of a theoretical frame for understanding the complex nature and role of health care systems in our society. This has led Canada and other English-speaking countries to intervene in their health care systems in complicated ways with largely negative results. We can begin by considering changes in the Canadian system between 1990 and the year 2000 using a more complex filter. We all know that the Canadian medicare system has been an important part of Canadian federalism for more than thirty years. It is deeply embedded in the values and culture of Canadians. Its special role in Canada is well captured in the claim that medicare is part of the infrastructure of the country. The metaphor of infrastructure carries with it a complex array of values and expectations. Medicare is a symbolic replacement for railroads as that which holds the country together; it is an important privilege of citizenship and an indication of Canadian social generosity. To get some sense of this, it is worth contrasting such ideas with the metaphorical place of health care in the United States or the United Kingdom. In the United States, health care is increasingly seen as a commodity that can be bought and sold. Among the complex consequences of this metaphor is a much diminished desire to pay for someone else's health care, except in cases of dire need when it becomes a charitable act to provide it for the elderly and the poor. The British metaphor for health care is as a government service. Because

Complicated and Complex Systems 25

government services often have long waiting lists and a variety of nonmonetary costs, there is not too much resentment of those who are prepared to pay to get better service by 'going private/ These social and cultural differences mean that interventions in each of these systems will have different impacts on how the results are interpreted. As recently as fifteen years ago, Canadians were among the most satisfied people in the world with how they received health care, but by the late 1980s this had begun to change. In Canada, the United States, and Great Britain, there were growing fears surrounding the inflation of health care costs and the sustainability of the then existing systems. The result was an epidemic of retrenchment and massive restructuring that has lasted for more than a decade. All three countries have faired relatively badly in the WHO rankings and all continue to have serious problems with their health care systems into this new century. Public dissatisfaction with the Canadian system has been growing steadily and progressively from the time of the first interventions. By 1994, it was becoming clear that things would not get better quickly. Michael Decter summed up the Canadian experience by declaring that we were no longer smug about medicare (Decter 1994). Things have not improved. We remain worried about the current state and future prospects of medicare. There are a number of widely acknowledged problems beyond current questions about the economic sustainability of the system. Public confidence has worn down to the extent that many Canadians are not sure that the system will be there should they need it. Overcrowded emergency rooms, intolerable waiting lists, and crises in cancer care and even in the water supply fuel this anxiety. Canadians' generosity in accepting the financial and non-monetary costs associated with universal health care has been eroding in the face of these anxieties (Berger 2001, 2000, 1999). Health care professionals, too, are dissatisfied with the current state of the system (Barret 2000). They feel overworked, underpaid, and seriously undervalued for the services they provide. For example, 'Canada is experiencing a crisis in nursing' (Advisory Committee on Health Human Resources 2000, 2). Increases in funding to the system momentarily reduced public anxiety in the late 1990s but it was widely recognized that this was only a short respite. A renewed sense that the health care system is out of control, buttressed by terrifying economic projections in a period of recession, has once again led to questions of economic sustainability. Renewed attempts to bring the system under control have now led governments to swing from extravagant and rather impotent generos-

26 Sholom Glouberman and Brenda Zimmerman

ity to tight-fisted frugality. These swings appear to be getting shorter and shorter. We are reminded of the parent who, not knowing what to do with a difficult child, alternates between withholding and increasing the child's allowance but never getting to the underlying issues. Most significantly, there are apparent threats to the guiding principles and values that lie behind the system. All five principles of the Canada Health Act are threatened today, in one way or another. • As pressures reduce the scope of what is considered medically necessary, what is medically possible has expanded. This disparity creates tensions around the public's understanding and expectation of comprehensive care. • Some of the economic burden of illness has shifted from hospitals and the public system to individual citizens. Because the greatest out-of-pocket costs fall on those workers who do not have privateinsurance coverage, they experience exclusion from a system which for them has become less than universal, and find the system less accessible because of these indirect economic barriers. • Each province has dealt with the pressures on the boundaries of coverage in its own fashion, resulting in threats to portability. Former health minister Alan Rock said, 'We sometimes see different criteria being applied to decide what services a client needs, different health providers for different services and various approaches in determining how much the client should pay' (Rock 1998). • As private insurers and furnishers of health services, from physiotherapy to Positron Emission Tomography (PET scans), begin to administer a greater proportion of health, public administration comes under pressure. Public policy in the area of health care has been remarkably impoverished over the last number of years. The great successes of the past like the introduction of medicare, the publication of the Lalonde Report (Lalonde 1974), and the passage of the Canada Health Act have not been replicated. In fact, if health policy were to be measured by levels of public confidence, there has been a steady failure over the last decade (Berger 2001, 2000,1999). The outcomes of health policy interventions are not the only reason to think that health policy has been based on a weak understanding of the nature of health systems and organizations. A much better indication of this failure is the glaring fact that policy makers in different

Complicated and Complex Systems 27

countries have taken diametrically opposite approaches to solve similar problems. If repairing health care systems posed a problem like sending a rocket to the moon, then one should be able to ignore local history and politics and find the correct technical structures that would make any system work well. But, while most Canadian provinces were regionalizing their health care systems to eliminate independent institutions with the hope of making the system more efficient, the United Kingdom was busy dismantling a regional system to create independent institutions to foster competition with the expectation that this, too, would result in increased efficiency. Similarly, changes to funding streams indicated the same lack of understanding of the role of finance in health care systems. While the United Kingdom was separating the provider function from purchasing of health care because this would theoretically increase accountability, the United States was creating vertically integrated health systems where the insurance funders also began to own and control provider organizations. The failure of the large element of private payment for health care in the United States to contain costs hardly deterred Canadian policy makers from embracing this avenue in the current Senate studies. Many of these responses came from ideological commitments either to market forces or to strong regulatory mechanisms. With the perceived scale of the problems often determining the level of response, there was widespread and massive restructuring of health care systems in the English-speaking world. Yet this restructuring seems to have resulted in destabilization of health care systems. The press to increased efficiency of the system produced only economies that were passed from one sector to another - savings in hospitals shifted the burden of cost or care to other providers and consumers of health care. Intractable Choices These opposing views appear in many health care debates. It is characteristic of complicated problems that choices must be made between different branches of a decision tree. Policy makers, experts, health care professionals, and politicians are divided about which choice to make, their divisions rooted in, among other things, left-right ideologies, differing views about economic forecasts and organizational structures, and even differences about what constitutes evidence. Contrasting perspectives on the economic sustainability of medicare pro-

28 Sholom Glouberman and Brenda Zimmerman

vide an excellent example. One group of experts (Robson 2001) argues that, if we project recent spending over the next forty years, medicare will bankrupt us. The opposing group, using the same data but over a longer period of time, concludes that average expenditure has not risen significantly and projects a healthy future (Evans 2002). As a result of these and other interwoven disagreements, many of the problems associated with medicare have appeared to become either non-existent or intractable. We can identify a series of powerful tensions that have arisen between such issues as public and private financing, between hospital and community care, between primary and secondary care, and so on. There are strong views on both sides of these tensions and a multitude of opposition parties on one side or the other. Current debates seem to surround these tensions, with different groups of professionals, government officials, and other experts lining up on opposing sides of most of them to do battle. They are often presented as 'either-or' issues. Four sets of tensions are presented in Table 1.2. If a resolution of real and perceived problems in health care system is to emerge, we will need to clarify how we understand these tensions in the health field, find ways of dealing with them, and reconsider how to intervene. The good news is that new ideas about the nature of organizations and systems as well as planning and policy development have been emerging (Coveney and Highfield 1995; De Greene 1993; Prigogine 1984), though these ideas are only now beginning to be understood. Mintzberg (1994) and others have suggested that rationalstrategic planning models need to be rethought in the light of our better understanding of rapidly changing and significantly less predictable environments. We have spent the last five years immersed in this area as investigators and consultants, and have applied some of these ways of thinking to the health field. Our recommendations have been adopted in a number of organizations and published in several places (Glouberman 2001; Zimmerman et al. 1998), with particular emphasis on new ways of approaching the health field. In this study, we suggest some approaches that might resolve the current failures in health care policy. Complex problems are sometimes called 'wicked' problems because many of their characteristics are not reducible to their constitutive parts. When solved, the solutions do not function as recipes, which can be applied to other, similar problems. There are many good examples of such problems. Often they are problems of prediction. We have learned that there are definite limits to our capacity to predict the

Complicated and Complex Systems 29 Table 1.2 Four sets of tensions Knowledge tensions Specialized vs. general knowledge Raising vs. lowering professional boundaries Professional vs. lay knowledge Evidence vs. experience-based knowledge Instrumental vs. hands-on knowledge Data-based vs. narrative-based knowledge Allopathic vs. homeopathic medicine Economic tensions Sustainable vs. unsustainable medicare Public vs. private funding Funding leading-edge individual treatment vs. funding population health Value quality vs. efficiency Smooth running vs. heroic battles against disease Governance tensions Centralization vs. decentralization Competition vs. collaboration Rational planning vs. self organization Structural change vs. relationship development Strict accountability vs. self-accountability (clinical governance) Hierarchical vs. flat organizations Federal vs. provincial regulation Institutional tensions Institutional focus vs. patient focus Institutions vs. community Individual vs. collective Standardization vs. customization Primary vs. acute care Risk avoidance vs. risk management

weather, the stock market, or, indeed, the next drip of the faucet (Kauffman 1995). But they also include problems of how to intervene in complex situations (Arthur 1996). We know with some precision how to diagnose and treat certain acute diseases, but people who suffer from complex chronic conditions require much more individualized care. In fact, treating such a patient is far more uncertain. It may result in unexpected instability, failure of standardized approaches, or surprising successes. Halstead Holman, a Stanford physician, has made a distinction between even relatively complicated illnesses and complex chronic diseases. Table 3 illustrates some of these differences. We have argued

30 Sholom Glouberman and Brenda Zimmerman Table 1.3 Complicated acute diseases and complex chronic diseases (adapted from H. Holman) Complicated acute diseases

Complex chronic diseases

Abrupt onset

Gradual onset over time

Often all causes can be identified and measured

Multivariate cause, changing over time

Diagnosis and prognosis are often accurate

Diagnosis is uncertain and prognosis obscure

Specific therapy or treatment is often available

Indecisive technologies and therapies with adversities

Technological intervention is usually effective: cure is likely, with return to normal health

No cure, pervasive uncertainty: management, coaching, and self-care over time is needed to improve health

Profession is knowledgeable while laity is inexperienced

Profession and laity must be reciprocally knowledgeable to improve health

that problems relating to health organizations and systems, health policy, and health itself are complex rather than complicated problems that occur in the context of complex adaptive systems. Characteristics of Complicated and Complex Systems Tables 1.4, 1.5, 1.6, and 1.7 list some of the characteristics of the two kinds of systems. We have taken them from existing literature on complex adaptive systems (Axelrod and Cohen 1999; Cohen and Stewart 1994; Coveney and Highfield 1995; De Greene 1993; Kellert 1993; Prigogine 1984; and Waldrop 1992) and clustered them for heuristic purposes. There is not enough space in the present study to explicate all of them fully. Instead, we provide examples of each cluster as illustrations. The establishment of medicare is often seen as the progressive linear development of evidence-based policy, but Taylor's classic text recognizes that real history is much more complex and non-linear: 'It would be comforting to believe that the governmental process follows ... [a]... neat and logical course, but, obviously, it does not. Unforeseen obstacles appear, unpredictable events (such as an election defeat) occur, and feedback from the environment warns of the need for changes in policy or strategy ... The more one examines the roles of interest groups and national and provincial political parties in the formulation

Complicated and Complex Systems 31 Table 1.4 Theory cluster Complicated systems

Complex systems

Linear

Non-linear (inputs and outputs not directly correlated)

Noise, tension, and fluctuations suppressed

Opportunity seen in tension, noise, and fluctuations

Solution as external to system

Solution as part of system

Adaptation is to a static environment

Interaction with the rest of a dynamic environment

Table 1.5 Causality cluster Complicated systems

Complex systems

Simple causality

Mutual causality

Designed and intended outcomes

Adaptive and emergent outcomes

Deterministic

Probabilistic

Certainty

Uncertainty

Assumed predictability

Recognized elements of non-predictability

Focus on boxes

Focus on arrows

Structures determine relationships

Structures and relationships are interactive

of policies, ... the less the two-tier federal system resembles the traditional 'layer-cake' concept and the more it exhibits the idiosyncratic confusion of a marble cake' (Taylor 1978, xvi). An excellent example of this series of differences is in the boxes and arrows often used to describe the factors that contribute to health or illness. They give the impression that the statistical correlations that support them will provide a causal deterministic account of health with a high degree of predictability. This is not really the case. There is often little proof of the direction of the causality, uncertainties are ignored, and the resulting pictures tend to distort or ignore the picture of health as complex and probabilistic, with many factors interacting not only with the individual but also with each other (Glouberman 2000a).

32 Sholom Glouberman and Brenda Zimmerman Table 1.6 Evidence Cluster Complicated systems

Complex systems

Reductionism/analysis

Holism/synthesis

Averages dominate: outliers irrelevant

Outliers seen as possible key determinants

Classical economics ignores historical evidence because systems always tend towards equilibrium

History contains meaning of change and systems evolve in part based on where they have been

Measures of efficiency, fit, and best practice

Functioning of actual relationships and feedback loops

Table 1.7 Planning cluster Complicated systems

Complex systems

Convergent thinking

Divergent thinking

Reductive characteristics

Emergent characteristics

Decision procedure as an event

Decision as emergent

Environmental scan

Developing insights into own practice

Big issue needs big change

Butterfly effect - size of problem does not determine size of change

The problem of overuse of emergency rooms is most often treated as a complicated problem, and the variables considered are narrowly focused. Economic measures, however, do not consider the broader history of the problem and every iteration seems to worsen the situation. The increased pressure on emergency rooms usually results in increased resources that come to them and their hospitals. But the money comes by reducing less urgent services. If one considers this issue in the context of complex systems, one can quickly recognize the cycle that results (see Figure 1.1) (Glouberman 2000b). The headings for the methodology of the Health Services Restructuring Commission in Ontario are listed below (HSRC 1997). They offer an excellent example of how planning is viewed as a complicated problem. When one looks at them one can recognize some of the reasons for the recurrent difficulties with the predictions of cost savings.

Complicated and Complex Systems

33

Figure 1.1 The vicious cycle in ERs

There are clear though complicated decision procedures that take into account a limited set of replicable factors. The methodology, though complicated, must ignore significant features of health care systems such as the cultural role of hospitals. One can also see a clear linear flow from one step to the next. Each step is an event, either of data gathering or decision making. Convergence is reached and then the next step is begun. The implicit assumption is that feedback loops will be at most a minor irritation - an outlier perhaps - but if one sticks to the plan, the process will unfold as envisioned. One can quickly recognize the potential for a multiplication of error as one goes through the following steps: Step 1: Determine Net Expenses Step 2: Calculate Program and Related Transfers Step 3: Calculate Clinical Efficiency Savings Step 4: Determine Support-Service Efficiencies Step 5: Reallocate Other Expenses Step 6: Calculate-Site Closure Savings Step 7: Determine Administrative Efficiencies Step 8: Add Back Selected Expenses Step 9: Establish the Cost of the Reconfigured System

Interventions in complex adaptive systems require careful consideration and planning, but of a different kind than in mechanistic systems. It is more important to understand local conditions and to be aware of

34 Sholom Glouberman and Brenda Zimmerman Table 1.8 WHO 2000 rankings of Canada and France Rank item

France

System performance Health performance Health level (DALE) Health distribution Responsiveness level Distribution Fairness of financial contribution Overall goal attainment

4 3 12 7 16-17 3-388 26-29 6

1

Canada 30 35 12 18 7-8 8 3-38 17-199 7

Source: WHO 2000.

the uncertainty and feedback that accompanies any intervention. There are some good examples of successful interventions to resolve major health problems. Case Study 1: France to the Top of the WHO Ranking When the WHO did its ranking of world health care systems in the year 2000, France emerged at the head of the pack. There remains some degree of contentiousness about these results, but it is worth examining this case for points of comparison and difference between Canada and France. We compare the various rankings in Table 1.8, present a cross-section of similarities and differences in Table 1.9, and provide a glimpse of some changes between 1990 and 1998 in the two systems in Table 1.10. The comparisons can be interpreted in different ways and their significance can be questioned, but we do not wish to enter into this kind of discussion. Suffice it to say that the data are varied enough to provide material for many complicated debates. Some pieces of comparative data stand out and make us wonder about arguments to increase private funding, the threats posed by an aging population, and the percentage of expenditure on drugs or the number of doctors, nurses, and hospital beds. France has an even higher ratio of public to private expenditure than Canada, as well as a significantly older population; it also spends more on drugs and has a higher ratio of doctors and hospital beds per 1,000 population than Canada. Its outcomes do not seem to be worse despite a longer average length of stay in hospital. Its

Complicated and Complex Systems 35 Table 1.9 Some comparisons between Canada and France

Population Per cent 60+ Per cent GDP on health Per-capita international dollars Per cent public Healthy life expectancy Number of doctors Number of nurses Infant mortality

France

Canada

59,000,000 20.50 per cent 9.40 per cent $2,074 76.10 per cent 70.7 175,431 347,918 4.4/1,000

30,000,000 16.70 per cent 9.40 per cent $2,363 70.10 per cent 70.0 50,000 227,000 5.5/1,000

Source: WHO 2002.

infant mortality rate, for example, has become lower than the Canadian rate. Some of the French data reinforce Canadian worries. For example, out-of-pocket expenditures in health have increased significantly in Canada while they have remained flat in France. Some of the similarities are also of some interest. For instance, the percentage of GDP spent on health is the same in the two countries and healthy years of life are similar. What is critical here is that we are not arguing that these attributes of the French system have caused its success. Rather, we are intrigued by the fact that the French have, at least up to the date of this survey, not followed global trends and conventional wisdom about how to have an efficient, effective, and economically viable system. Our argument is that the attributes, statistics, and structures of the system are less intrinsically related to success than are the more complex understanding and assumptions about how systems work. Given some of the more recent changes France is making in health care, we would argue that its resistance to change was a result less of explicit understanding than of a series of implicit assumptions, and, as a consequence it has now become vulnerable to the global trends in health care reform. Our interest in looking at France is to see what insights we can gain about how France perhaps inadvertently created a more successful system than Canada by 2000. None of these tables adequately explains how the Canadian system has deteriorated over the last decade. In fact, the data suggest that the Canadian system has become far more efficient, with shorter lengths of

36 Sholom Glouberman and Brenda Zimmerman Table 1.10

Canada and France, 1990 and 1998 OECD data

1990 Canada Life expectancy at birth, females 80.4 Life expectancy at birth, males 73.8 Infant mortality, deaths per 1,000 live births 6.8 Total in-patient care beds, per 1,000 population 6.3 Practising physicians, density per 1,000 population 2.1 Average length of stay in in-patient care, days 13 Total expenditure on health, million National Currency Unit (NCU) 60,422 Total expenditure on health, per cent GDP 9 Public expenditure on health, million National Currency Unit (NCU) 45,071 Public expenditure, per cent GDP 6.7 Per cent total expenditure on in-patient care 49 Per cent total expenditure on Pharmaceuticals 11.4 Out-of pocket payments, per capita US$ 270 Private insurance, per cent total expenditure on health 8.9

1998 France

Canada France

80.9 72.7 7.3

81.5 76.1 5.3

82.2 74.6 4.6

9.7

4.1

8.5

2.6

2.1

3

13.3

8.2

10.7

569,782 82,821 8.6 9.3

796,986 9.3

436,594 58,082 6.6 6.5 46 43.1 20 15 207 392

606,348 7.1 44.8 22 209

10.9

11.2

12.6

Source: WHO 2002.

stay and a reduced complement of beds and staff. Moreover, public expenditure has gone down from 6.7 to 6.5 per cent of GDP. Yet, despite this apparent increase in efficiency, the sense is that the system remains in deep trouble - in terms of financial sustainability as well as of quality, and of public confidence. Andrew Pettigrew, a British expert on strategic planning, has pointed out that NHS managers have learned to manage (complicated) structural change but lost the capacity to manage (complex) health care (Pettigrew et al. 1988). It appears that, if we could only have the best structural features of another system, then we could resolve our problems. But it is difficult to see what we should emulate in France. Should we adopt the French ratio of bed complements and professional staff? Should we reduce out-of-pocket costs? Should we spend even more on drugs? It quickly becomes clear that we cannot have the same system as

Complicated and Complex Systems 37

the French and that the situation is far more complex. The more one looks at other systems up close the more it becomes apparent that health care institutions and systems are complex because they are deeply embedded in local politics, culture, and history, among other factors. There is much to learn from other complex health care systems, but it is not the case that we can simply adopt their structures or processes to resolve our problems. A good example of this in France is the array of programs and subsidies for mothers and children that provide support from pregnancy to nursery school. These programs have evolved over the last number of years. They include such things as training in parenting, nutritional support for pregnant women, and developmental day care for infants and toddlers. There are also financial subsidies to encourage good personal care and proper diet. These policies have come to be known by ordinary people as la maternelle, which translates literally as 'nursery school/ Their place in French social policy is similar in many ways to the role medicare once had in Canada - it marks a piece of the French soul and is celebrated by most French people, from ministers to cab drivers, who declare with a swagger the extent of their social generosity. We might easily transform the joke about Canadians to say, "The French are British who drink wine and love mothers and children/ Many of the programs are consistent with a French version of Fraser Mustard's vision for Canadian social policy. Although it is difficult to assign causality, the rate of infant mortality in France has gone from being higher than the Canadian rate to being lower over the last decade (OECD 2002). Perhaps we do not yet love mothers and children as the French do. And, although Quebec has adopted aspects of these programs, it has not captured the hearts of Quebecers as la maternelle has the French. It is a critical part of the solution of our present difficulties to find and regain our own maternelle - one that interacts with our culture and values in relatively complex ways. Certainly, we can learn more about the role of health and social policy in shaping and responding to the values and cultures of a country. We conclude that a major difference between the Canadian system and the French one is that there was less Draconian restructuring imposed on the French system. Important changes to the system, like la maternelle, did occur, but they were shaped in a context of relative structural stability and were largely programmatic. (Currently France is undergoing what is considered by many to be a major upheaval as the work week is shortened to thirty-five hours.) These apparently

38 Sholom Glouberman and Brenda Zimmerman

minor changes have had significant impacts on the effectiveness and efficiency of the French system, especially in comparison with other WHO-surveyed countries. Case Study 2: HIV/AIDS in the Developing World The Brazil Story In this section, we explore the story of a complex health issue: HIV/ AIDS in the developing world. We argue that, when HIV/AIDS is considered as a complex rather than a complicated problem, radically different options emerge for intervention and policy. We show how a complicated approach to the problem determines the types of questions posed and logically leads to a conclusion that the situation is intractable. We follow this by looking at Brazil's success in beating the odds and argue that its approach embraced the underlying principles of complex adaptive systems. This dramatically changed the nature of questions asked, the solutions found, and the interventions taken (Begun et al. 2002). Despite the fact that Brazil's annual per-capita income is less than $5,000 (Downie 2001), it has managed to disprove the World Bank prediction that it did not have the resources to resist HIV infection arid would have 1.2 million cases of HIV/AIDS by 2000 (World Bank 1997). Instead, it had 0.5 million (UNAIDS/WHO 2000). Brazil has been far more successful than South Africa in combating AIDS. In the 1980s, Brazil had one of the worst infection rates in the world (Darlington 2000), far surpassing South Africa. Today, South Africa's HIV infection rate is 25 per cent of the population whereas Brazil's is 0.6 per cent (UNAIDS/WHO 2000). Brazil challenged all of the prevailing complicated assumptions about how to deal with the HIV/AIDS epidemic. The Underlying Assumptions of a Complicated View In 1997 the World Bank reported that an estimated thirty million people had contracted the human immunodeficiency virus (HIV) and 90 per cent of those were in developing countries (World Bank 1997). HIV/ AIDS in developing countries is often assumed to be an intractable problem. Why? Here are the assumptions underlying the diagnosis. Cost of Drugs

Antiretroviral drug cocktails have transformed AIDS into a controlla-

Complicated and Complex Systems 39 ble, chronic disease for those who can afford the drugs in the developed world. However, their cost is out of reach for poor countries. 'Antiretroviral therapy, which has achieved dramatic improvements in the health of some individuals in high-income countries, is currently unaffordable and too demanding of clinical services to offer realistic hope in the near term for the millions of poor people infected in developing countries' (World Bank 1997). Prevention versus Treatment In developing countries with limited resources and high rates of HIV infection, choices need to be made as to how to spend limited health care dollars to fight the disease. Countries frequently decide that prevention of the spread of the disease is the best they can do; treatment is a luxury they cannot afford. So they opt to focus almost exclusively on prevention. Uneducated, Illiterate Patients Cannot Manage Their Own Therapies The drug treatment regime to control AIDS is a complicated routine. Different drugs need to be taken at specific times of the day. How can illiterate, uneducated people be expected to comply with such a sophisticated routine? In addition, sometimes a drug needs to be taken with food, which adds another challenge in poor countries where food is a scarce and unpredictable resource for the poor. Prevention Needs to Focus on the Fear and Severity of the Problem One of the tactics taken in prevention is to inform the public of the dangers involved in unsafe sex. The fear of disease and death will change people's behaviour to limit the spread of AIDS. Making a Dent in the HIV/AIDS Problem in Developing Countries Will Take a Generation or Two An almost exclusive focus on prevention, even if it is successful, will result in devastating losses in the current adult generation and perhaps the next if they were born with HIV/AIDS. The impact of today's prevention tactics will really be seen by the second or third generation. In the meantime, many millions will die of the disease. Integration Requires a Sophisticated Health Care System across a Country In the developing world, health care systems and public health infrastructures are frequently not consistent across a country. A national

40 Sholom Glouberman and Brenda Zimmerman Table 1.11 Brazil questions Questions that assume a complicated problem

Questions that assume a complex problem

What will drug costs be for our infected population? Or whom can we afford to treat?

How can we reduce costs so that we can provide treatment for all who need it?

Since illiterate poor people cannot be expected to comply with a complicated regime of therapy, what resources are needed to assure compliance for those treated (i.e., assumption of need for professionals to manage patients' regime)?

What methods of communication will work to convey drug-therapy routine to a patient - even a homeless, illiterate patient? If food is an issue, how can we use existing charities to provide food so that patients can get food at the right time for their drug regime?

With our limited resources, should we focus more on prevention or treatment? Or what are the resources for an effective prevention treatment?

How can we achieve our prevention goals while treating all of those currently infected?

What infrastructure do we need to implement our plans? What will this cost? What are the trade-offs? From what program/ service will we take the money to afford this infrastructure?

Where are the informal and formal networks/relationships that exist and that are consistent with our overall approach/ values? How can we help to strengthen these connections? What skills or resources exist already in our country and how can we help make them more visible and hence useful?

system of prevention and treatment requires a well-developed integrated health care system and infrastructure. Brazil's Approach to HIV/AIDS As a Complex Problem Brazil did not accept the analysis of the HIV/AIDS problem presented by the World Bank. It implicitly recognized that it was a complex problem that lent itself to different questions and hence led to less intractable conclusions. Tables 1.11 and 1.12 contrast the questions and answers in the two perspectives. A more detailed account of how Brazil dealt with the problem is presented below. Cost of Drugs

The government gives the drugs away for free to HIV/AIDS patients. Since 1994, Brazil has been manufacturing generic versions of the drugs in the antiretroviral drug cocktail. In doing so, it uses the contro-

Complicated and Complex Systems 41 Table 1.12 Brazil answers Conclusions that assume a complicated problem

Conclusions that assume a complex problem

Meaningful solutions require sophisticated, integrated, national health care systems.

We will find ways to use the resources we have to respond to the problem.

We cannot provide treatment to all when the costs are so high. Choices must be made. We cannot afford resources to manage treatment compliance. With our limited resources, we should focus more on prevention than treatment.

We will find a way to provide treatment to all who need it by dramatically reducing costs. We will use our informal system to train people to care for themselves. Prevention will be part of treatment and treatment will allow us 'access' to population for prevention strategies.

versial provision of the WTO that allows countries to violate patent laws in cases of national emergency (American Medical Association 2001). Brazil argues that the HIV/AIDS epidemic is and could become a national emergency. Although both the WTO and the United States challenged Brazil on its use of this approach, Brazil has continued, and in 2001 the United States dropped its lawsuit against Brazil. Each year, more of the drugs have been produced in Brazil in a generic form; by 2000, eight of the twelve necessary drugs were produced in generic form, at enormous cost reductions (Darlington 2000). While estimates of the cost reduction vary, the costs are being further reduced as more and more of the drugs are produced in generic form. At a minimum, the costs of the drug therapy per patient per year are 65 per cent lower than the $12,000 cost in the United States. Some estimate that it could be further lowered to be 90 per cent less than the U.S. cost. Prevention versus Treatment Brazil chose to use treatment as part of the prevention strategy. When people know they can get treatment, they are more willing to come to hospitals, clinics, or certain non-governmental organizations (NGOs) for tests (Rosenberg 2001). The situation is not deemed to be hopeless. While they are there for treatments or tests, they also get information and spread the ideas of prevention. Today, the bulk of the spending is on treatment, yet the prevention goals are being met.

42 Sholom Glouberman and Brenda Zimmerman

Uneducated, Illiterate Patients Nurses and other health care workers teach patients how to take the drugs. They use whatever methods they can to communicate the drug routine to the patients. They will draw pictures of the sun or the moon to denote different times of day. They will draw pictures of food on the labels of the pill bottles for those that need to be consumed with food (Rosenberg 2001, 30). In addition, they will help the poorest patients link up with NGOs, churches, and other organizations which offer free food. In spite of the high illiteracy rate in Sao Paolo, the adherence rate for the drug regime is at the same level as in San Diego. In both cities, 70 per cent achieve an 80 per cent adherence rate (Rosenberg 2001, 30). Prevention Needs to Focus on Fear and the Severity of the Problem Brazilian organizations have used a variety of playful approaches to sell the concept of condom use. The approaches include celebrating life and humour. Bright-coloured costumes on musicians playing music are used as an opportunity to share information and make the use of condoms acceptable among the high-risk groups. Pre-school children are taught about HIV/AIDS prevention in some parts of Brazil (Lehman 1999). Ads depict condoms as props. One ad shows three women sitting around a table, which has a condom for a tabletop. All the women are saying, 'Yes.' The caption is 'With a condom, they'll say "yes, yes, yes"' (Walbran 1998). Making a Dent in the HIV/AIDS Problem in Developing Countries Will Take a Generation or Two Brazil's efforts really began in earnest in the early 1990s. By 1994, Brazil was producing their first generic antiretroviral drugs. Within five years, it had made a major impact on reducing the spread of the HIV virus. In the 1980s, it was held out as an example of one of the worsthit countries for HIV/AIDS; today, it is touted as a model for developing countries fighting HIV/AIDS. Integration Requires a Sophisticated Health Care System across a Country At the time it began its campaign against HIV/AIDS, Brazil was not among the poorest countries in the world; it had an established infrastructure of hospitals, clinics and public health services. Yet its health system was hardly of the calibre of First World systems (Rosenberg 2001). There were huge differences in the services available across the country and to different segments of the population. Nevertheless, Bra-

Complicated and Complex Systems 43

zil's HIV/AIDS efforts have, to quite an extent, strengthened the health infrastructure, or web of connections, necessary to do the treatment and prevention work with HIV/AIDS. Brazil has used over six hundred existing NGOs and community-level care organizations to reach the country's poor (Center for Disease Control 2000). The country now has 133 testing and counselling centers. Health care clinicians work alongside NGOs and other organizations to provide the full range of services needed. 'It is a well-organized, well-formulated program that works because the government has managed to integrate the whole society especially NGOs' (Rosenberg 2001). The questions posed were of the kind of 'who has the current resources, skills needed to provide the treatment services or complementary needs for HIV/AIDS patients?' In other words, Brazil looked for the hidden resources and existing informal relationships in the country which would not show up on a Ministry of Health organization chart. It looked at what existed in reality rather than in theory. The country did not assume away the messy and complex nature of the 'system' but rather accepted it and embraced an emergent system and structure, holding on to a few key principles. Changing the Nature of the Questions in Brazil There are some key differences between a hypothetical set of questions that assume the Brazil case to pose a complicated problem and ones that assume its complexity. Both kinds of questions are based on a coherent set of assumptions about reality and change. Complicated problems are machine-like and complex problems more life-like. Each set of assumptions both illuminates and distorts certain aspects of reality. The machine metaphor of the complicated approach suggests that external managers - or mechanics - are needed to fix the system or its parts. The life metaphor suggests that solutions (and new problems) have the potential to emerge from within. The external role becomes more facilitative than mechanical. Our contention is that in Canada health care policy makers, administrators, and citizens have too often tried to reduce the complex problems of medicare to complicated ones. They have ignored many of its complex characteristics. The most prominent questions posed in the Canadian context tend to assume that the problems associated with medicare are complicated. The challenge is twofold: to articulate an approach to the complex issues and to resist reducing the analysis to one that assumes a complicated problem..

44 Sholom Glouberman and Brenda Zimmerman

How Ideas about Complexity Can Be Applied to Canada's Health Care Reform In this final section, we will build on the lessons drawn from the French and Brazilian cases and other examples when examined with a complexity perspective rather than with a complicated view of health care. Building on the Brazilian analysis, we demonstrate how changing the questions posed for Canada's health care system from a complicated to a complex viewpoint changes the solutions seen as viable. We show how changing the questions asked can shift the framework for intervention and lead to some clear directions, even some specific action plans, for policy makers. We conclude with some optimism that Canadians have the potential to make positive changes to the health care system. Examples of Questions for Canadians Reflecting on Health Care Reform Many of the questions currently raised about Canada's health care system assume a complicated system and hence constrain the possible answers. Some of the specific questions to which this study is a response are good examples of ones that assume that current problems must be viewed as complicated. Questions like 'Is it true that the success of reform depends on the means put in place to see it through?' carry with them the assumption that there is a process of 'seeing it through' to a definitive conclusion. It is such assumptions, which often imply that someone else is in control of the system and can take the blame for its troubles, that we are challenging. The provincial premiers and ministers of health often say that Ottawa is to blame because they have reduced resources. Hospital executives blame the provincial governments for lack of direction. Researchers blame politicians for meddling with their solutions. The other side of blame is control. If someone or something can be blamed for a failure, then it is logical to assume they can also be responsible for success. This fits clearly within the complicated assumptions noted above. An external party is responsible for the results and outcomes. Clear causal chains of inference can be made. Someone is ultimately responsible for the success or failure of the system. 'The buck stops here.' When leaders do not seem to have the solutions, other sources for the problems are exposed - again often from a complicated perspec-

Complicated and Complex Systems 45

tive. Money is a zero-sum game. Spending more on health care means less will be available for other services, as demonstrated in our example of emergency-room closures. The vicious cycle is repeated as 'parts' of a system are seen as both the problem and the solution. In Table 1.13, we have list some questions that assume that Canadian problems about health care are complicated. Learning from Brazil's lead, we can then reframe them if we assume that the issues are complex. The first set of questions is laden with constraints of limited resources, lack of leadership, trade-offs, external control of the future, and individual rights. They paint a picture of an intractable problem where compromises are necessary but not likely to satisfy many. In contrast, the complex questions have a more optimistic tone as they look for existing resources. We now present some possible answers to the four complex questions raised above. How Do We Build on Current Structures and Relationships to Stabilize and Enhance Medicare? At first glance, it appears that France and Brazil offer radically different approaches. France used incremental changes in la maternelle while Brazil made dramatic changes. Traditionally, it was thought that incrementalism is an evolutionary rather than revolutionary approach (Lindblom 1968). However, using insights from complex adaptive systems, which are non-linear in their reactions, new conceptions of change have emerged. Small, or incremental, inputs to a system can have dramatic or revolutionary effects (Zohar and Morgan 1995; Morgan 1998). Revolutionary change can happen within an evolutionary container. The cases of both Brazil and France suggest that it is important not to disrupt existing resources. A stable context can apply existing resources to useful change. Because many parts of the health care system interact, large- scale complicated interventions can destabilize the entire system, with chaotic results. In Canada, the health care system has undergone constant economic and structural changes for the last ten years. At the end of this period, everyone is unhappy and costs are less well contained. Complicated repairs have not worked and we would argue that they have destabilized the system. They have weakened relationships and Canadians - both providers and recipients of health care - are losing faith that the espoused principles or values of the Canada Health Act will be adhered to.

46 Sholom Glouberman and Brenda Zimmerman Table 1.13 Canadian questions Questions that assume a complicated problem

Questions that recognize the problem as complex

What are the structures we need to make the health care system sustainable?

How do we build on current structures and relationships to stabilize and enhance Medicare?

Can we afford increasing care and treatment for an aging population? What do we have to give up to support the most effective and advanced technology (or drugs)? How much should Canadians pay for their health care?

How can we provide care and treatment that makes everyone feel that the system will be there should their family need it? How can we help health care institutions and professionals enhance the quality of services and innovation in technology arid drugs? How can medicare contribute even more to the Canadian identity?

Therefore, we conclude that the first thing to do is to introduce relatively small interventions to stabilize the system and regain some measure of confidence in it. In the current situation, it is wise to introduce only those structural changes to the system that may stabilize and improve relationships. The system contains a large number of health care relationships that by and large work and are effective. These can be strengthened and reinforced. Other relationships have been disrupted by major changes and must be given time and care to adjust if we are to create and stabilize new ways of working. We will concentrate on a few examples of the kind of interventions that might be considered, and of others that should not be considered. All contain elements of cost containment and of system stabilization. Specifically, out program contains the following elements: • find the local informal relationships that already work in the system; • identify local providers and users of the system who can identify what makes it work; • legitimize what works by providing recognition and support for those who make the system work; • do not change structures to disrupt those relationships; • increase security of employment for nurses - they need to feel confi-

Complicated and Complex Systems 47

dent that they are not the first to be cut whenever cost containment is put on the table; • provide security of income for doctors - they need to feel confident that if they achieve savings, this will not threaten their livelihood either; • do not increase the burden of care for patients and families but work with existing care systems to enhance self-care rather than a sense of abandonment; and • do not decrease the support for health-related organizations and other professionals. How Do We Make Everyone More Confident That the System Will Be There Should They Need It? Many of the issues surrounding access have to do with several consequences of recent changes. A paradox of complicated solutions is that rationalized services reduce points of access to them. This often makes access worse by clogging other already overburdened parts of the system - an unintended and expensive outcome of a rational, complicated planning process which undervalued the cultural context of health as a local phenomenon. Increasing points of access is less expensive than concentrating them in emergency rooms; advice and social-support services provide more comprehensive and less expensive access than emergency rooms. This is a lesson that we are already beginning to learn. Telephone-based nursing services are helping to change the shape of the population that goes to the emergency room (Lattimer et al. 2000). Similar advisory and counselling services can help turn the vicious cycle described above into a more virtuous one (see Figure 1.2). Some examples of these kinds of relatively inexpensive interventions follow: • provide multiple access points - for example, tele-health and telecounselling - which paradoxically allow for centralized provision of care in a highly local context for citizens; • increase information points to the system (e.g., local, provincial and Canada-wide bed availability services); • make waiting times more transparent and safer (e.g., an emergencyroom clock for non-urgent cases, and a clearing house to post waiting-times for procedures and indicate safe margins for waiting);

48 Sholom Glouberman and Brenda Zimmerman Figure 1.2 The virtuous cycle in ERs

• multiply less urgent services inside emergency rooms (e.g., direct support for frail elderly patients who may need help while they wait); • increase support for care providers (e.g., more support for relatives of frail elderly); and • increase support for 'self-carers/ e.g., more tobacco-cessation programs to complement advertising). How Do We Recognize and Support Efforts to Improve Care? Organizations in the health field are almost always eager to learn and apply new knowledge and techniques. Most have a long history of assimilating innovation and improvements in treatment. Much learning is typically complex and non-linear. Efforts to build complicated solutions to learning have most often ignored the more complex aspects of learning and of health care and attempted to reduce complex health care education to complicated manageable components. Often a choice is made between introducing innovative improved care and containing costs. It is clear that solutions to Canadian problems must do both. We suggest that some interventions might begin to respond to the problem, and that continued efforts to solve the puzzle are worth pursuing. Clarifying boundaries can be as useful as lowering them in order to reduce the destructive rivalry that pits one profession against another. The effort of the Canadian Health Services Research Foundation (CHSRF) to increase research in nursing is a good example of this. All

Complicated and Complex Systems 49

the groups of professionals and paraprofessionals who provide health care should have more freedom (and responsibility) to do what they do best for patients: doctors to diagnose and treat illness, nurses to provide physical care and support for patients through illness, and so on. Many small points of innovations should be encouraged. The paradox in health care is that, although the acquisition of knowledge has benefited greatly from the division of labour, the integration of health care requires additional skills and attributes of connection (Glouberman and Mintzberg 2001). We must: • recognize and reinforce the altruistic motives of most providers; • respect the different complementary professional perspectives on care; • reward the differentiated knowledge that emerges from various professions; • avoid disrupting the division of labour in the acquisition of knowledge; • increase the cross-disciplinary recognition of the value of its results; and • dedicated funding for discipline-based research to stabilize costs. How Can We Restore Medicare to Reinforce Canadian Identity? We have been trapped into a narrow way of defining and responding to current issues as if they were merely complicated. Success in answering more complex questions will begin the process of revitalizing the Canadian health care system. It will help resolve many of the tensions which are now present in the system. The struggle to answer even the three questions we have posed can improve how health care is provided by stabilizing the system, recognizing the nature of health care knowledge, and providing opportunities to improve services in a cost-contained environment. These are clearly not all the questions, nor are we capable of providing all the answers. But this new perspective can help refine the questions to ask and provide indications of some of the kinds of answers we need. There is at least as much expertise and ingenuity in Canada to generate and answer complex questions as in Brazil and France. We are confident that Canadians can rise to the challenge to mobilize and use their existing resources to revive a truly Canadian universal health care system. In answering such complex questions, Canadians may

50 Sholom Glouberman and Brenda Zimmerman

become more confident that health care will be there when they need it and a fresh understanding of the principles of the Canada Health Aci may emerge. In this new understanding, the questions asked will closely link economic sustainability to values of social generosity, equity, and security that have been central to the Canadian identity. We need to be open to questioning the principles of the Canada Health Act, but with a deeper understanding of the complexity of hea Ith and health care. The fact that the principles have not been adhered to is not necessarily an indication that they are wrong. Rather, we need to look at the gems in Canada's health care system and determine what core values and principles they actually reflect. Our contention is that some of our best-functioning systems do indeed embody most of these. A healed system will contribute to health by reinforcing the universality of health care coverage, which is increasingly recognized as a critical determinant of health. It will also provide improved care, because the relationships among those providing care will be improved and the flow of work will be less fragmented. The struggle to contain costs will be a virtuous cycle spread throughout the system, rather than a vicious cycle of blame focusing first on one fragmented sector and then on another. REFERENCES

Advisory Committee on Health Human Resources. 2000. The nursing strategy for Canada. Conference of deputy ministers/ministers of Health. American Medical Association. 2001. Brazil may defy the United States and make more AIDS drugs. Journal of American Medical Association, HIV/ AIDS Resource Center, Reuters Health Information Available at: www.ama-assn. org / special.hiv / newsline / reuters / 02068951 .html Arthur, W. Brian. 1996. Increasing returns and the two worlds of business. Harvard Business Review 74 (4) July-August: 100-9. Axelrod, Robert, and Michael D. Cohen. 1999. Harnessing complexity: Organizational implications of a scientific frontier. New York: Free Press. Begun, J.W. 1994. Chaos and complexity: Frontiers of organizational science. Journal of Management Inquiry 3(4): 329-35. Begun, J.W., B.J. Zimmerman, and K.J. Dooley. 2000. The Berger population health monitor. Toronto. - 2001. Health care organizations as complex adaptive systems, S.M. and M. Wyttenbach, ed., In Advances in Health Care Organization Theory. San Francisco: Jossey-Bass.

Complicated and Complex Systems 51 Berger, Earl. 2000. The Berger population health report. Toronto. - 2001. The Berger population health monitor. Toronto. Center for Disease Control. 2000. Center for disease control daily news, 22, September 2000. Available at: www.ama-assn.org/special/hiv/newsline/cdd/ 091800g3.htm Cohen, Jack, and Ian Stewart. 1994. The collapse of chaos: Discovering simplicity in a complex world. New York: Viking. Coveney, Peter, and Roger Highfield. 1995. Frontiers of complexity: The search for order in a chaotic world. London: Faber and Faber. Darlington, Shasta. 2000. 'Brazil becomes model in AIDS fight.' 7 November. Available at: www.aegis.org/news/re/ 2000/re001107.html Decter, Michael. 1994. Smug no more. Paper presented at the King's Fund International Seminar, Banff, Alberta. De Greene, Kenyon B., ed. 1993. A Systems-based approach to policymaking. Boston, Dordrecht, London: Kluwer Academic Publishers. Downie, Andrew. 2001. Brazil: showing others the way. San Francisco Chronicle, 25 March 2001. Available at: www.aegis.org/news/sc/2001/ sc010310.html Evans, Robert. 2002. Keynote address. Paper presented at Diagnostics and solutions: 7th annual conference of the McGill Institute for the Study of Canada, Montreal. Glouberman, Sholom. 2000a. 'A Dynamic concept of health, In Sholom Glouberman, ed., Towards a new concept of health: Three discussion papers. Discussion paper no.HJB. Ottawa: Canadian Policy Research Networks. - 2000b. Why emergency rooms are overloaded: Health and everything. Available at: www.healthandeverything.org - 2001. Towards a new perspective on health policy. CPRN Study no.H_03. Ottawa: Canadian Policy Research Networks. Glouberman, Sholom, and Henry Mintzberg. 2001. Managing the care of health and the cure of disease - part I: Differentiation. Health Care Management Review 26 (1): 56-69. Goodwin, Brian. 1994. How the leopard changed its spots: The evolution of complexity. New York: Touchstone 2002. Government of New Zealand. 2002. Growing an innovative New Zealand forward. Auckland. Holland, John H. 1995. Hidden order: How adaptation builds complexity. Reading, Mass.: Helix Books. HSRC. 1997. GTA/905 health services restructuring report, appendix I. Toronto: Health Services Restructuring Commission. Kauffman, S. 1995. At home in the universe: The search for laws of self-organization and complexity. New York: Oxford University Press.

52 Sholom Glouberman and Brenda Zimmerman Kellert, Stephen H. 1993. In the wake of chaos. Chicago: University of Chicago Press. Kelly, Kevin. 1994. Out of control: The rise of neo-biological civilization. . Reading, Mass.: Addison-Wesley. Lalonde, Marc. 1974. A new perspective on the health of Canadians. Ottawa: Minister of Supply and Services. Lattimer, Val, et al. 2000. Cost analysis of nurse telephone consultation in out of hours primary care: Evidence from a randomised control trial. BM] 320 (7241): 1053-7. Lehman, Stan. 1999. Brazil AIDS official urges education. Associate Press. Available at: www.aegis.org/news/ap/1999/ap990609.html. Lindblom, Charles E. 1968. The policy-making process. Englewood Cliffs, N.J.: Prentice-Hall. Lorenz, Edward. 1993. The essence of chaos. Seattle: University of Washington Press. McDaniel, Jr R.R. 1997. Strategic leadership: A view from quantum and chaos theories. Health Care Management Review 22(1): 21-37. Mintzberg, Henry. 1994. 'The fall and rise of strategic planning. Harvard Business Review 72(1): 107-14. Mintzberg, Henry, Bruce Ahlstrand, and Joseph Lampel. 1998. Strategy safari: A guided tour through the wilds of strategic management. New York: Free Press. Morgan, Gareth. 1998. New directions in management: The art of using small changes for large effects. Journal of Innovative Management 4,1:34-42. OECD. 2002. OECD health statistics. Organisation for Economic Co-operation and Development, 1998,2002. Available at: http://www.oecd.org/oecd/ pages/home/displaygeneral/0,3380,EN-statistics-194-5-no-no-no194,FF.html. Pettigrew, Andrew, Lorna McKee, and Ewan Ferlie. 1988. Understanding change in the NHS. Public Administrationn 66: 297-317. Priesmeyer, H.R., and L.F. Sharp. 1995. 'Phase plane analysis: Applying chaos theory in health care. Quality Management in Health Care 4(1): 62-70. Prigogine, Ilya. 1984. Order out of chaos: Man's new dialogue with nature. Boulder, Colo. New Science Library. Robson, William. 2001. Will the boomers bust the health budget? A C.D. Howe Institute Report. Paper presented at the Canadian Institute of Actuaries Annual Meeting, Toronto, 18 June. Rock, Allan. 1998. Speaking notes for CMA 131st annual meeting. Paper presented at the CMA 131st annual meeting, Whitehorse, N.W.T., 7 September. Rosenberg, Tina. 2001. Look at Brazil. New York Times Magazine, 28 January,

Complicated and Complex Systems 53 26-54. Avaibable atwww.accessmed-msg.org/msf.accessmed/ accessmed2.nsf/iwpList4 Stacey, Ralph.1992. Managing the unknowable. San Francisco: Jossey-Bass. Taylor, M.G. 1978. Health insurance and Canadian public policy. Montreal: McGill-Queen's University Press. UNAIDS/WHO. 2000. UNAIDS/WHO epidemiological fact sheet on HIV/ AIDS and sexually transmitted infections. Update, Revised. Walbran, Shannon. 1998. The dance against AIDS in Brazil. Available at: www.changemakers.net/journal/98november/walbran.cfm: Changemakers.net Journal. Waldrop, M. Mitchell. 1997. Complexity: The emerging science at the edge of order and chaos. New York: Simon and Schuster. Wheatley, Margaret, J. 1992. Leadership and the new science: Learning about organization from an orderly universe. San Francisco: Berrett-Koehler. World Bank. 1997. 'Confronting AIDS: Public priorities in a global epidemic. Summary at www.worldbank.org/aids-econ/confront/confrontfull/ summary.html. Oxford: Oxford University Press. World Health Organization. 2000. the world health report 2000. Geneva: World Health Organization. - 2002. Archives, Available at: http://www. who.int/whr/2001/archives/. Zimmerman, B. 1999. Complexity science: A route through hard times and uncertainty. Health Forum Journal 42(2) March/April: 42-6. Zimmerman, B., C. Lindberg, and P. Plsek. 1998. Edgeware: Insights from complexity science or health care leaders. Irving, Tex.: VHA. Zohar, A. and G. Morgan. 1995. Achieving quantum change incrementally: Schulich School of Business working paper. Toronto: York University.

2 Financial Rules As a Catalyst for Change in the Canadian Health Care System IAN MCKILLOP

In 2001 it is estimated that almost $40 billion was distributed by provincial/territorial governments to health organizations with mandates that include the provision of institution-based health services (Canadian Institute for Health Information [CIHI] 2001a). Individually and collectively, these organizations have responsibility for the delivery of quality services designed to enhance and promote the health of Canadians. Designing organizational structures and policies that support the delivery of health services in a comprehensive and equitable fashion is a non-trivial endeavour that has long occupied the minds of researchers, funders, and practitioners alike (see, e.g., Palmer 2000; Coast et al. 2000; Anderson and Cosby 1998; Aas 1995). The fact remains, however, that there appears to be no single answer to how the delivery of health services should be funded or organized. Nonetheless, it is valuable to engage in a process of continuous evaluation, seeking to exploit features of existing approaches that work well while seizing opportunities to overcome challenges as they are identified. Many parties contribute to the fabric of the Canadian health care system. Many physicians, pharmacists, and others provide services directly to Canadians using a variety of for-profit business models. The services of these professionals are complemented by a large not-forprofit infrastructure characterized by organizations that often deliver health care in multiple sectors (acute care, mental health, community health, etc.) using common funding and governance structures. This paper focuses on the management control practices of the latter group. In the following pages, a selection of issues related to the rules and

Financial Rules As a Catalyst for Change 55

practices that currently govern funding and resource-allocation practices in Canada are explored. Specific attention is given to three areas of interest. These are: • the approaches used to allocate financial resources; • timing issues related to the budget-planning process used by health organizations; and • the capacity of health organizations to access alternative revenue sources. Each issue is explored from the perspective of the management control system to which it contributes, with a particular focus on how the current practices affect the prospects for implementing change. A number of issues worthy of further exploration are identified, and, where appropriate, recommendations are provided to stimulate discussion. The Concept of Management Control: Towards an Understanding of How Organizations Respond to Financial Rules Management control is the process of ensuring that the individuals within an organization conduct themselves in a manner that is consistent with what management desires to see done (Merchant 1997). In effective organizations, what management desires to see done is the successful execution of the strategy established by the governors of the organization. This execution of strategy is achieved through the development and implementation of a variety of management control processes, one subset of which are the 'financial rules' that form a focus for this paper. Collectively, these processes form a management control system. Seminal works such as that by Barnard in 1938 (Barnard 1970) and Simon's (1995) much-referenced work on levers of control have established the legitimacy of considering management control as a discipline. Otley (1994) recognizes the centrality of financial rules and policies but reminds us that these levers are but one element in a comprehensive management control system. Although non-profit organizations (such as the organizational structures responsible for managing the largest single portion of Canadian health expenditures) do not share the strategic goal of profit maximization with their market-driven colleagues, evidence suggests that the activities engaged in by management in both settings are remarkably

56 Ian McKillop

similar, and, as a consequence, management control concepts developed in the for-profit sector also have relevance in the non-profit sector (Anthony and Young 1995; Siciliano 1997). In the context of this discussion paper, there are two management control systems in play. The first is the management control system implemented by a funding agency (a ministry or department of health), where there is a desire to ensure that the system encourages (if not requires) the judicious and appropriate use of funds for the provision of health care by the organizations receiving funds. The second is the management control system employed by the funded agency, where mechanisms are implemented to ensure that the funds received are deployed in a manner consistent with that organization's strategic goals. While one might expect a natural alignment between the strategic goals (and therefore the management control systems) of these two actors, such is not always the case. For example, a funder may have a strategic goal of supporting a healthy population and, by extension, minimizing the need for hospitalization. The funder visualizes a system in which funds would be directed towards programs that promote healthy lifestyles, detect illness at an early stage, or support the delivery of services in the home. This strategy, however, may not be directly aligned with that of the agencies receiving funding. Consider the situation faced by a health organization with multi-sector responsibilities, such as an organization with responsibility for mental health, community care, and acute care services. While the organization appreciates that investing funds in programs that promote healthy lifestyles will reduce hospitalization rates in future years, the fact remains that implementing such a strategy in the short term will limit the organization's ability to achieve its strategic goals in the intervening years. (One such goal is probably to provide quality care to all those who seek care.) Faced with this prospect, the health organization feels that it is able to make only modest investments in preventative health strategies. Instead, in keeping with the organization's strategy, increased investment in cardiac care cind other high-cost services are made. While admittedly a contrived scenario, the example illustrates an important point that underpins the issues explored in this paper. Funders implement management controls (such as financial practices) in an effort to direct the behaviour of health organizations in a manner that is consistent with the funder's strategy. In the absence of these

Financial Rules As a Catalyst for Change 57

controls (or, at times, in spite of such controls), health organizations will take actions that are in their own best interests - in other words, actions that are consistent with the health organizations' strategy. Merchant (1997) provides a simple framework that can be used to reflect on the efficacy of elements of a management control system. Merchant argues that effective management control systems are designed in a manner that overcomes the three reasons that explain why actors (such as health organizations) exhibit behaviours that are not consistent with the stated goals of the broader system (in this case the funding agency). The reasons he identifies are applied here in the context of the discussion in this paper: • Lack of direction - Health organizations may not know what is required of them. It is important that the management control system ensures that health organizations understand the specific goals they are expected to achieve. These goals must include elements that can be objectively measured. While requiring a balanced operating position is an important first step, much more must be done in this area, particularly with respect to providing direction through the use of performance measures. • Motivational problems - Health organizations may know what is expected of them, but their own objectives are not aligned with those of the system. When funding agencies fail to address this problem in the design of their management control system, health organizations may take actions that benefit themselves at the potential expense of the system. 'Gaming the system' by exploiting a shortcoming in the design of a funding formula is an example. The financial rules contained in the management control system must be designed in a manner that aligns the motivational objectives of health organizations with the objectives of the system as a whole. • Personal limitations - Health organizations may know what is expected and be motivated to achieve what is expected but lack the expertise or knowledge to achieve the objectives. For example, in Canada's health system, financial-data quality issues, and the dearth of robust information on the cost of providing care, limit the ability of managers to make certain decisions in an informed manner (McKillop et al., 2000). In order to facilitate decision making with respect to change initiatives, it is important that personal limitations be addressed by the management control system (e.g., by facilitating access to reliable financial information).

58 Ian McKillop

Effective management control systems have three characteristics. The first is that they are future-oriented. The systems provide clear expectations and motivation to ensure a future state is achieved. Control is not about focusing on what has already occurred.1 Second, effective management control systems are objective-driven. The overriding purpose of the management control system is to ensure that goals are realized (Flamholtz, Das, and Tsui 1985). And lastly, effective management control systems balance the cost of implementing controls with the benefits expected. Good management control is not about imposing rules and constraints in the absence of an evaluation of the benefits to be gained by the rule or constraint. Accountability alone does not create efficiency in health systems, as evidenced by the experience in Norway (Pettersen 2000). A balance is required between these characteristics, as is found in the management control systems of all excellent organizations (Cawsey, Deszca, and Teall 1994). Guided by the principals of being future-oriented, objective-driven, and balanced, developers of effective management control systems rely on rules, policies, and practices designed to overcome problems related to a lack of management control. Following from Merchant's (1997) framework, these rules, policies, and practices are a combination of: • action controls - Control elements that cause health organizations to take actions that are consistent with the strategy of the health system as a whole (or prevent them from taking undesirable actions). Examples are requiring government approval before beginning a hospitalconstruction project or submitting an operating plan for approval. • results controls - Control elements that do not prescribe the means but, instead, focus on whether a desired end state is achieved. The decision by many jurisdictions to implement rules against operating deficits is one example, achieving a performance benchmark is another. • cultural or personnel controls - Control elements that build on individuals' desires to help themselves and others. Examples are a strong sense of shared professionalism reinforced by a code of conduct, engaging in continual professional training, and effective selection processes. Management control is about enabling, encouraging, and promoting good management practices. Since good management includes the abil-

Financial Rules As a Catalyst for Change 59

ity to embrace change, this paper posits that a well-designed management control system (and the financial practices it embodies) is a required precursor for change initiatives to be successful. Welldesigned management control systems are, in themselves, catalysts for, not barriers to change. If barriers to change are identified in a management control system, they point to a weakness in the design of the system itself. This must be addressed through changes in the design of the system if the emergence of dysfunctional behaviours is to be avoided. With this in mind, consideration is given to three elements of the management control system used in the Canadian health care sector. The discussion that follows focuses on health organizations whose mandate includes the provision of acute care-services. In most jurisdictions in Canada, these organizations also have responsibilities for a variety of care delivery missions in the broader health services sector. Funding Approaches as Catalysts for Change One of the most visible elements of the management control systems used by provinces and territories across Canada is the funding mechanism. Almost always, the funding mechanism contributes to the direction and motivation elements of the management control system. Each province/territory has developed a funding mechanism tailored to meet its individual needs. This has resulted in a wide variety of approaches being used. Nonetheless, all funding approaches share some common characteristics. Each • reflects the organizational structures found in the jurisdiction, • provides incentives to ensure the effective and efficient use of financial resources by fundees, and • respects fiscal realities in the jurisdiction. Although jurisdictions differ in the means by which they choose to implement their funding approach, similarities in the objectives and mechanisms of each permits approaches with similar characteristics to be grouped together for discussion purposes. Doing so supports a broader discussion of the impediments and challenges associated with specific approaches, without dwelling on the attributes of mechanisms used by individual jurisdictions. The taxonomy for classifying Canadian funding approaches released by CIHI in 2001 (McKillop, Pink, and Johnson 2001) is used for this purpose. For information purposes

60 lanMcKillop Table 2.1 Scope of the funding approach Scope

Description

Comprehensive

Comprehensive-funding approaches direct money to health service organizations with multi-sector responsibilities (some combination of long-term care, acute care, mental health, community care, etc.). These organizations typically have considerable freedom with respect to how they choose to distribute funds to each sector.

Institutional

Institutional-funding approaches direct money to specific acute care hospitals (or groups of acute care hospitals operating as a single corporate body) or other single-purpose health organizations (such as a long-term care facility.) Although these organizations may have some discretion over how money will be spent within their organization, they are usually not permitted to redirect money to other organizations.

Service specific

Service-specific funding approaches direct money to organizations to support the provision of a specific service or the care of a specific disease. The organization usually has a mandate to provide this care to residents drawn from a wide geographic area. Funds cannot be used for purposes other than the service or disease for which the funds were specifically granted.

Source: McKillop, Pink, and Johnson 2001,16.

only, a table summarizing the primary funding approaches used to allocate operating funds in the fiscal year 2000-1 is provided in the Appendix. Canadian jurisdictions employ funding approaches that are a combination of two kinds of decisions: a decision that is made with respect to the scope of the funding flow, and a decision that is made with respect to the method used to determine the relative share of the funding flow. Each of these characteristics has a profound influence on the receiving organization's prospects for implementing change. Challenges and Opportunities Related to Funding Scope The scope of the funding approach is determined by the mandate of the health organization being funded. The three funding scopes used in Canada are described in Table 2.1. Because scope is matched to organizational-delivery structures, and because almost all jurisdictions in Canada have moved to a system of supporting health organizations with multi-sector responsibilities, funding approaches with a compre-

Financial Rules As a Catalyst for Change 61 61

hensive scope are the most prevalent in Canada.2 Attention is therefore given to this approach. From a management control perspective, comprehensive-funding approaches offer a number of advantages.: • Comprehensive-funding approaches support a total system view, allowing strategies that reflect the interactions between different sectors of the health system to be embraced. • Comprehensive-funding approaches transfer responsibility for decisions related to the 'how and where' of care delivery from the province/territory to the health organization receiving the funding. • Comprehensive-funding approaches support a system-level perspective in decision making. Savings identified in one sector can be easily reinvested into service enhancements in another sector. • Comprehensive-funding approaches, by recognizing the shared strategic goals of the sectors managed by the health organization, should result in a shared sense of purpose, which is an important element of an effective management control system (Glouberman and Mintzberg 2001). • Comprehensive-funding approaches allow health organizations to adapt more easily to change than is the case with those funded using institutional or service-specific approaches. This is because the heavy reliance on results-based controls, and the wider service-mandate implicit in the use of a comprehensive approach, permits these organizations to reallocate resources between activities more easily. Looking Forward Management control problems related to lack of direction emerge when governments fail to ensure that health organizations have a clear understanding of the strategic goals and missions the government has in mind. This can manifest itself in health organizations pursuing initiatives that are not in the best interests of the jurisdiction as a whole. An example might be a smaller health organization deciding to offer a specialized surgical service on a province-wide basis. From a motivational perspective, this action is perfectly rational. The financial rules allow the health organization to be reimbursed when providing services to residents domiciled outside their region. The health organization expects this transfer to be in excess of the direct cost of providing the service and will use the excess to cross-subsidize other activities of the health organization. Equally attractive is the fact that, when offer-

62 Ian McKillop

ing the service to residents domiciled inside its region, the organization contributes to its own mandate by providing comprehensive care for its citizens. From a provincial/territorial perspective, however, this is not an ideal use of resources. Economies of efficiency and clinical effectiveness will be diminished as more and more health organizations choose to offer services that would otherwise be concentrated in a few centres of excellence. One solution is to implement action controls that require approval by the health ministry/department before such a service can be implemented. Another is to be more explicit in the direction given with respect to the mandate of the health organization. Best of all, however, is to consider carefully the unwanted behaviours that will be encouraged when developing a funding method. This is a step to which too little attention is paid. In the hypothetical example given above, a decision to include a component that ensures equity when one health organization provides services to residents belonging to another health organization is being exploited. A small change in the design of the financial rule would overcome this problem.3 A challenge that resonates through funding approaches of all scopes is that few health organizations have accounting systems sufficiently developed to permit a detailed understanding to be gained of how resources are consumed and costs are incurred.4 This limits the stewardship and monitoring activities of the funding agency and the decision-making capabilities of the health organization. If we are to foster change and make informed financial decisions about the impact of change, additional investment in the functionality and capacity of the financial-information systems used in our health system is imperative. This is particularly true in times of perceived or real crisis (Ezzamel and Bourne 1990). Efforts have been made in the past few years to broaden our financial systems' capacity beyond acute care. These efforts are welcomed and must be reinforced by health ministries/departments since they stand to realize benefits equal to those of the direct users of the systems. This is because better information leads to better decisions. Challenges and Opportunities Related to Funding Methods Features of the funding scope have an influence over prospects for implementing change within health organizations. The funding method in use, however, often has a more pervasive influence.

Financial Rules As a Catalyst for Change 63

Funding methods describe the process used to determine the actual dollars to be directed to a health service organization. Seven types of funding methods are used in Canada, with jurisdictions often using at least two different methods - one for apportioning operating funds and one for apportioning capital funds. The eight methods are presented in Table 2.2. For each funding method, an assessment of the relative ability of a health organization receiving funds to innovate or adapt to change is provided. An assessment of the relative stability of the funding method on a year-overyear basis is also given. The freedom of health organizations to implement change is inversely proportional to the tightness of the action, result, and cultural controls implicit in the funding process. Funding methods that tie apportionment merely to characteristics of the population being served (be these demographic or case-mix characteristics) provide the greatest opportunity to pursue change initiatives. The lack of tightness in the funding process allows the funded organization to explore the impact of reallocating funds between programs (such as by transferring savings realized in an orthopaedic program to renal dialysis) or between expense objects (such as by introducing rethermalization equipment and reducing in-house kitchen staff). This freedom is considerably constrained when methods such as line-by-line funding are used. This is because funding decisions in these cases are largely a function of what has been spent before, and line-by-line review implies a strong action-control orientation. Even when zero-based budget methods are introduced (requiring a critical examination of the rationale for each expense line item), the desire of actors to protect their relative share of the funding apportionment makes it difficult to identify savings in one line item that could be reinvested into another line item (Williams, Newton, and Morgan 1985). A number of funding methods (including population-based and case mix-based) have the potential to provide relatively stable funding allocations over time. When allocations do change, it is in direct response to changes in the underlying characteristics of the population served. As such, the response is a predictable and expected change in funding, and one that is consistent with the established delivery modalities of the organization receiving funds. Thus, although not explicitly designed as multi-year funding methods, these approaches have the potential to support longer-range planning horizons by health delivery organizations (Duckett 1995). This is in contrast to methods that involve multipliers (global and

Table 2.2 Types of funding methods

Method

Description

Population-based

Uses demographics or other characteristics of the population (such as age, gender, socioeconomic status, etc.) to determine the relative propensity of different population groups to seek health services. Uses characteristics of the organization providing care (such as size of organization, type of organization, geographic isolation, occupancy rate) to estimate the cost of operating a health service organization. Uses a profile of cases and/or service volumes previously provided (such as number of knee replacements, number of dialysis procedures) to estimate the cost to sustain a specified profile of cases and/or service volumes in the future. A factor is applied to a previous spending figure (or to a forecast spending figure) to derive a predicted spending level for an upcoming period.

Facility-based

Case mix-based

Global

Usually relies most upon

Relative ability of fundee to trespond to changes

Relative potential for year-to-year stability of funding

Results controls High

High

Results controls

High

High

Results controls

High

High

Results controls High

Medium

Table 2.2

(Concluded)

Method

Description

Line-by-line

Factors are applied on a line-by-line basis to previous cost experiences (or to forecasted costs) to derive a proposed funding level for each line item (such as housekeeping, inpatient nursing, etc.) for an upcoming period. Funding is directed to address specific policy initiatives of the Department or Ministry of Health. These policy initiatives affect the operation of multiple organizations within the jurisdiction, (e.g., funding designed to reduce waiting list time for a specific procedure). Funds are directed to a single health service organization in response to evaluating a proposal from that organization for one-time funding, often for a major expenditure. Ministerial-discretion methods are those where the decision on the specific dollar amount to flow to health service organizations is made by the minister of health.

Policy-based

Project-based

Ministerial discretion

Source: McKillop, Pink, and Johnson 2001, 2 (adapted).

Usually relies most upon

Relative ability of fundee to trespond to changes

Relative potential for year-to-year stability of funding

Results controls

Low

Medium

Results controls

Low

Low

Action controls

Low

Low

Action controls

Medium

Low

66 Ian McKillop

line-by-line) or the policy-based and ministerial-discretion methods. In these cases, year-over-year stability in the funding allocation is not predictable. This forces organizations to adopt planning horizons that are equal to or less than the annual budgeting process. Such methods are not conducive to promoting long-term change strategies. Instead, what emerges are a series of reactionary measures, implemented quickly, to respond to funding realities without the opportunity to explore longerterm implications. Looking Forward We must stress two important points that have a significant impact on the ability of almost all funding methods to direct funds at a rate that fully reflects the need identified. These points are particularly relevant for the funding methods that should offer the best advantages in terms of flexibility to respond to change and year-over-year stability in funding flow (i.e., population and case-based methods). The first is that, in all jurisdictions in Canada, funding methods are used to apportion a relative share of a funding envelope. The funding methods are not used to direct funds equal to the need the funding method actually identifies. For example, a case-based method may identify that a health organization with a current year budget of $150 million has a projected service profile that will have an expected cost of $200 million to service next year. The cases involved represent 1 per cent of the case mix in the jurisdiction. A health spending envelope of $1 billion is approved by the government. The health organization is allocated its proportionate share (1 per cent) of the envelope, or $100 million. This is $100 million less than the expected need, and $50 million less than the organization received the previous year.5 The health organization will find itself extremely constrained and will struggle in its ability to meet its service mandate. This is despite the fact that the funding method was designed to be objective, should have provided considerable freedom for the organization to respond to change, and should have furnished a relatively stable (or at least predictable) flow of funds over time. Thus, the financial rules used to operationalize otherwise quite rational and fair funding methods can themselves represent an impediment to change. Even if it is possible to develop management control systems that provide appropriate direction, motivate behaviours that are consistent with the strategy established, and that equip organiza-

Financial Rules As a Catalyst for Change 67

tions with the skills and knowledge needed to make effective decisions, the effort expended to develop the management control system is moot if the very same system apportions resources that are insufficient to realize the strategy. In the absence of a viable management control system, dysfunctional behaviours emerge as organizations attempt to cope as best they can with the resources they have available (Merchant 1990). The second important point is that we have limited experience in Canada with respect to the level and intensity of health interventions that are appropriate. Many of our funding methods are based on the premise that past levels and intensities of health interventions are indicative of actual need. To the extent that this assumption is flawed, many funding methods perpetuate the inequities of past practices in future funding decisions. A few ideas related to financial practices deserve further attention to help overcome these challenges. These ideas are: • If jurisdictions wish to use funding methods that rely on the propensity for Canadians to seek health care, then we must encourage research to determine the relationship between the propensity to seek care and objective clinical evidence of the population's need for care. This information is critical to the viability of the financial funding mechanism. • In jurisdictions where funding models are able to determine aggregate dollar requirements objectively, a threshold limit should be set for the gap that is allowed to exist between the identified need and the size of the funding envelope. Governments must recognize that, when this gap grows beyond the threshold, the management control systems used by health organizations will break down and dysfunctional behaviours will emerge. • Predictability and equity in funding should be favoured over yearto-year stability in funding. We have little history in Canada with using multi-year funding, other than for capital projects.6 Notwithstanding this point, there are credible reasons to avoid funding flows that vary wildly year-over-year, most notable of which is that it is almost impossible for health organizations to shed or acquire fixed costs over such a short time horizon (leading again to dysfunctional behaviours emerging and a collapse of the management control system). An idea worthy of exploration is to enhance funding mechanisms to distinguish between the funding of capacity and the funding of day-to-day activity. Capacity could be funded using a

68 lanMcKillop

longer term rolling horizon (such as three or five years), while activity would continue to be funded on a one-year horizon. Such an approach removes the dysfunctional motivational challenges triggered by constantly adapting to funding increases or decreases by making changes in operations at the margin. • The financial practices associated with existing funding mechanisms fail to recognize that investment is often required to realize savings. These investments may be in new technologies (such as digital imaging), new organizational structures (such as mergers), or new practices (such as care in the home). The existing practice is to require health organizations to identify surplus funds within existing operating funding to make these investments. Unable to do so, health organizations pressure governments into establishing special funds to which they can submit claims for adjudication. Payment is often at some reduced rate for each dollar spent. As management control theory would confirm, health organizations are motivated to maximize their share of the special fund by being generous in their interpretation of expenses that qualify. To better align the objectives of the funder and health organizations more effectively, it is suggested that in lieu of the current approach, governments establish innovation funds against which health organizations can borrow to invest in initiatives expected to generate savings in later years. As savings are realized, repayment to the innovation fund by one organization replenishes the balance for others. The Budgeting Process As a Catalyst for Change In most corporate entities, responsibility for management of the revenue stream rests with the same senior management team holding responsibility for management of the expense stream. This does not reflect how health delivery is organized in Canada. The revenue stream is determined by funding agencies that are separate from the health organizations tasked with delivering services. The separation of the management of the revenue and expense cycles causes jurisdictions to include a variety of other management control elements in their management control system. Key among these elements are prospective and retrospective monitoring practices. Prospective monitoring practices involve evaluating plans for expected future events. The preparation of business plans by health organizations, which are submitted to government for approval prior to funding decisions being made, are an example of a prospective mon-

Financial Rules As a Catalyst for Change 69

itoring practice and represent an action control. Budgets (a plan stated in monetary terms [Anthony and Young 1995, 442]) are a required element of operating plans in all jurisdictions in Canada. Retrospective monitoring practices involve evaluating past performance, either in absolute terms ('expenditures exceeded revenues by 5 per cent'), relative terms ('expenses by a health service organization were higher than those for another health service organization of similar size and scope'), or evaluative terms ('expenditures were 15% higher than the benchmark'). Jurisdictions in Canada focus much more effort on prospective monitoring activities than on retrospective monitoring functions. The latter is limited in many provinces/territories to the requirement that audited financial statements be submitted. There is, however, a growing interest in the role that performance monitoring activities could play in promoting and rewarding the effective and efficient delivery of health services, as evidenced by publicly available performance information emerging in a number of settings (see, e.g., CIHI 2001c; Helyar et al. 1998). As the ability to calculate these measures mature, they hold the promise of contributing in a valuable way to a results-based component of the management control system. Challenges and Opportunities Presented by the Timing of Budget Submissions The management control systems in all jurisdictions include components to support prospective monitoring. The most common component is the required preparation and submission of a budget. To promote fiscal accountability, almost all jurisdictions in Canada now also require health organizations to develop balanced budget estimates. This forces health organizations to explore carefully a wider variety of options as they seek to pursue their strategic goals. Table 2.3 outlines differences between selected jurisdictions with respect to the rules related to the timing of the budget submission. Two distinct strategies are observed with respect to the timing of the budget process. In some jurisdictions, budgets describing estimated expenditures for the upcoming planning period are submitted prior to the government making a decision about the size of the funding envelope. The other strategy is for a government to announce funding decisions and then request health organizations to prepare a budget outlining how funds will be spent. Each approach offers advantages as well as some distinctive shortcomings, as summarized in Table 2.4.

70 Ian McKillop Table 2.3 Timing of budget submission by province/territory (for the Fiscal Year 2000-1)

Province

Budget required? Must be submitted by

Budgeted deficit allowed? Yesa Nob N/AC Nod

Alberta British Columbia Manitoba New Brunswick Newfoundland and Labrador Nova Scotia

Yes Yes Yes Yes

Eight weeks after funding announcement 30 June following fiscal year end 1 June 2000, for the upcoming fiscal year Mid-January for the upcoming fiscal year

Yes Yes

Ontario

Yes

Prince Edward Island Quebec

Yes Yes

Saskatchewan Yukon

Yes Yes

Yese 1 6 October for the upcoming year Usually due in February for the upcoming Nof fiscal year and 1 0th of each month forupdates Date varies annually - usually in February for the upcoming fiscal year No9 January/February for the upcoming fiscal Noh year Three weeks after being advised of the No1 amount of the annual budget Yes' 1 5 May 2000, for fiscal year 2000-1 30 September for the upcoming fiscal year Nok

Source: McKillop, Pink, and Johnson 2001, 61. Specific conditions apply for deficit to be accepted. b The ministry works with health service organizations anticipating a shortfall. c In Manitoba, requests for additional funding are made with the budget submission. d The government of New Brunswick has issued a directive that regional hospital corporations will operate within approved budgets. Transitional funding to cover working capital shortfalls will be available to regional hospital corporations as required pending upcoming government decisions concerning system structure and governance. 6 In Newfoundland and Labrador, an action plan is being developed in an attempt to control deficits. f No planned deficit legislation exists in Nova Scotia. A planned deficit must be supported with a plan to eliminate the deficit within a specified period of time. 9 Ontario's policy is to not fund deficits, although some health service organizations have received special funding for this purpose because of financial pressures. Extended financial difficulty may trigger an operational review by the ministry. h In Prince Edward Island, regional health authorities are expected to submit a balanced budget. ' When a deficit is projected, the Quebec Ministry of Health and Social Services requires submission of a fiscal balance plan to correct the deficit. ' Saskatchewan Health is not committed to funding deficits. If a deficit budget is submitted, a plan for how this will be managed in the current year and eliminated in future years is required. k The Contribution Agreement in Yukon Territory requires the health service organization to operate within budget. Operating surpluses may be used to fund deficits from prior years or in future years. a

Financial Rules As a Catalyst for Change 71 Table 2.4 Advantages and challenges presented by financial rules related to the timing of budget submissions Budget submitted after funding is announced

Budget submitted before funding is announced

Advantages

Advantages

Knowing the actual dollars available facilitates the planning decisions that must be made by health organizations to adjust service and activity levels to match available funding.

Budget plans have a strong correlation with expected activity levels because they are the result of a detailed planning activity that is usually a required element of the management control system within each health organization.

The effort required to prepare comprehensive budgets is rewarded as these budgets can contribute in a valuable way to the organization's management control system. In many jurisdictions, announcing funding decisions before receiving budget plans causes the government to implement objective funding approaches using population-based or case-based methods. This is because the government is unable to use spending estimates to inform its decisions regarding apportionment. Challenges The government does not have the opportunity to inform itself of the magnitude of the gap between the amount of funding being made available and the perceived funding need of health organizations. Increased usage of service-specific funding approaches is required in situations where the government wishes to ensure that specific strategic initiatives are pursued by the funded organizations. (This is because the government will not be aware of strategic initiatives being pursued by health organizations until after the funding announcement has been made.)

The budgets provide the government with valuable insights into the strategic directions being pursued by the health organizations it supports. It is easy to identify the gap between the spending activity planned for in health organization budgets and the size of the funding envelope the government knows is available. Challenges In situations where the government is unable to provide funding at a level equal to the aggregate value of budgets submitted (which is common), the budgets prepared by health organizations become useless components of their management control systems. This is because the original budgets include revenues that will not be received, which will have implications for service volumes and potentially even the range of services the health organization is able to offer. The original budgeting process must be repeated using the announced funding. When this does not happen, the planning and control functions in the health organization are compromised.

72 Ian McKillop

From the perspective of encouraging effective management control, apportioning funds prior to asking health organizations to prepare a budget is the most attractive approach. This is because this sequence ensures that organizations evaluate their strategic plans in the context of the fiscal reality with which they are faced. In all health organizations, this requires difficult decisions to be made. But, once the strategic direction has been established, and decisions are made with respect to how this direction will be achieved, the organization is left with an operating plan and a budget that provides a realistic blueprint for achieving the strategy. Thus, the budgeting activity (which is almost always highly participatory in health settings) is legitimized and ownership for the decisions made is held throughout the organization. A budget with the characteristics of ownership, validity, and achievability is able to assume an important role in the design of an effective management control system. Looking Forward Sadly, the benefits presented by this approach are often overshadowed by two limitations in the way in which the budget process is implemented. These limitations are: • Funding announcements are often made either after the beginning of the fiscal year or too close to the beginning of the fiscal year for the health organization to adopt any meaningful strategy to react to the fiscal realities. In both cases, costs are being incurred before the health organization learns that program changes must be implemented. • Health organizations have a difficult time reconciling funding allocations with the service mandates they have set for themselves. This can manifest itself in a belief that it is impossible to reduce services to match fiscal realities of the funding allocation. At the other end of the scale, there is the possibility that the funding allocation may provide support at a level in excess of that which is actually required. Unless the management control system is designed in a manner that causes this information to become known to the funder, the information will remain private to the health organizations. These limitations represent an impediment to implementing change that would not exist were the planning process to be refined.

Financial Rules As a Catalyst for Change 73

For budgets to act as a catalyst for change, funding announcements must be made sufficiently ahead of the beginning of the fiscal period to allow health organizations to adapt to the financial reality they now know they face. One way to help lessen this impact is to move to the system advocated earlier, namely, of funding capacity on a multi-year rolling time horizon. This would align financial plans for capacity requirements with a planning horizon that presents realistic opportunities to adapt by shedding or acquiring these costs. An equally attractive feature of this approach, from a management control and political standpoint, is that governments would be able to make decisions regarding capacity using financial information that is not distorted by utilization rates.7 Alternative Revenue Sources As Catalysts for Change Given the fiscal reality that the funding provided by governments is rarely deemed adequate to meet all the strategic goals of health organizations, it is desirable for health organizations to consider how these funds can be supplemented with alternative revenue sources. Many health organizations have the freedom to seek alternate revenue sources for non-medically related services. For example, acute care hospitals often generate revenues through charges for preferred accommodations and the use of parking facilities, as well as revenues from public cafeteria operations. Long-term care facilities commonly assess residents a share of costs related to the provision of housing and food, and home care services may ask recipients to share in the cost of this service to the extent they are economically able to do so. In addition, most health care organizations also have formalized arrangements with a charitable foundation they have established under their corporate umbrella or work with in close partnership. The reality, however, is that with the exception of the significant contributions to fund investments in buildings and equipment made through foundation partnerships, these revenues represent a relatively modest proportion of overall operating revenues. For example, in 1998, community and small hospitals in Ontario generated only 12 per cent of their total revenue stream from sources other than the Ministry of Health and Long Term Care. Teaching hospitals were found to have a greater capacity to generate funds from alternate sources, at 18 per cent of total revenues, while non-acute hospitals demonstrated the greatest capacity, at 21.5 per cent (University of Toronto 1998).

74 Ian McKillop

While it is attractive for health organizations to pursue alternative revenue sources, organizations differ greatly in their ability to access these revenue streams. A number of reasons explain these differences: • There may be an inability or unwillingness on the part of potential customers to pay for services. For example, more and more companies are removing coverage for semi-private and private hospital accommodation from their benefit plans. Hospitals have noted a corresponding decrease in their ability to raise revenue through this source. But when hospitals are able to retain 100 per cent of the revenues they raise through this source, they will usually seek ways to ensure this revenue stream is optimized, often by dedicating a bedmanagement specialist tasked with ensuring that no request for preferred accommodation goes unmet because of inefficiencies in the bed-allocation process. • Market forces create differences in supply and demand for revenueproducing activities like parking lot operations. While a health organization in a major urban setting may command fees of $20 per day and attract 500 vehicles, health organizations in other settings may find that the supply of free neighbourhood parking negates their ability to raise any funds through parking tolls. • The health organization may not have excess management capacity to oversee revenue-generating activities. • The accounting information system used may be limited in its ability to provide accurate and timely information needed to manage revenue-generating activities. The way in which the management control system treats realized surpluses also has an influence on a health organization's incentive to find opportunities for savings. Some jurisdictions permit unrestricted access to surpluses, others limit this access to a certain degree, and yet others reclaim surpluses (see Table 2.5). These actions need to be considered in the context of the behaviours jurisdictions wish to promote. There is a certain logic behind a jurisdiction reclaiming surpluses. First, a surplus indicates a potential flaw in the planning process. Had the funding and planning process operated more effectively, funds in excess of those needed for operations would not have been advanced. As such, it is arguably fair that these, now visible, excess funds be returned to the provincial or territorial treasury so that they can be redistributed to other health organizations.

Financial Rules As a Catalyst for Change 75 Table 2.5 Treatment of realized surplus by province/territory

Province/territory

Can surplus be retained by hospital organization? Restrictions on use of surplus

Alberta

Yes

British Columbia

Yes

Manitoba

Yes

New Brunswick

Yes

Newfoundland and Labrador

NA

Nova Scotia

Yes

Ontario

Yes

Prince Edward Island Quebec

No Yes

Saskatchewan Yukon

Yes Yes

None - province recommends using for capital equipment None - province recommends using for capital equipment, except surpluses from specially funded services A maximum of 2 per cent of budget may be kept and used at their discretion Yes - a portion of surplus in transferable programs may be kept by Region retains surplus in non-transferable programs In recent years, surpluses have not existed Yes - all surpluses are reviewed by department and an action plan is taken accordingly None - province recommends using for capital equipment Approval of Health and Social Services Board is required before facility is permitted to use its surplus None None

Source: McKillop, Pink, and Johnson 2001, 64.

If, however, the consequence of demonstrating good stewardship as evidenced by a realized surplus is that the surplus will be reclaimed by the funding agency, it is not surprising to discover that surpluses rarely emerge. This is in keeping with management control theory, which suggests that actors have a natural desire to take actions that are in the best interests of their own organizations, even though these actions may not be aligned with the strategic direction of the system as a whole. It is expected that actors anticipating a surplus would engage in actions to eliminate the surplus by directing the 'surplus' funds to pro-

76 lanMcKillop

vide additional services within their organization or to offset costs of a future period. The net effect is to mask the true cost of providing services to the sponsoring jurisdiction. To the extent that the capacity to generate excess funds comes not from efficiency gains realized within the funding envelope but, instead, from ancillary operations or other revenue-generating activities of the health organization, a different problem emerges. It is one of perceived inequity. Organizations with greater capacity to generate funds (e.g., a health organization operating in a large metropolitan setting) benefit from having a larger proportion of total revenues from non-government sources. If all health organizations have received base funding using an approach that in itself is equitable, then the differences in the ability to raise additional revenues creates an advantage for health organizations so fortunate. In response to this point, health organizations with more limited revenue-generating capacity have argued that, in a system based on equity, it is only fair that health organizations able to generate revenues from non-government sources above a certain threshold should be required to share these excess revenues (possibly on some type of a sliding scale) with other health organizations. Resolution of this point is difficult in the context of the design of an effective management control system. While the arguments in favour of equity are persuasive, the literature suggests that attempts to 'tax' the revenue-generating capabilities of actors will reduce their incentive to find avenues to maximize these revenues. Jurisdictions that choose to reclaim all preferred revenue premiums see this lived out when they learn that their preferred accommodation occupancy rate is much less than in jurisdictions where this management control element is not in place. Conclusion This paper has addressed the question of how organizational financial rules affect change in the context of the organizations tasked with managing the largest component of financial resources dedicated to health care. It has been shown that financial rules are an integral component of comprehensive management control systems, and that when well designed, these systems enable, encourage, and promote good management practices. In doing so, these systems act as catalysts for change, providing motivation and incentives that cause actors within

Financial Rules As a Catalyst for Change 77

the health system to take actions which ensure that goals consistent with desired strategies are achieved. When barriers to change are encountered, these signal a weakness in the design of the management control system - a weakness that, by definition, implies the management control system will encourage behaviours that are inconsistent with the desired strategy. Specific attention was given to three areas of interest: • the approaches used to allocate financial resources; • timing issues related to the budget-planning process used by health organizations; and • the capacity of health organizations to access alternative revenue sources. It was shown that the timing of government funding announcements has a profound impact on the ability of organizations to develop budgets and plans able to contribute to effective management control systems. Insufficient lead time in funding announcements forces health organizations to make adjustments in service levels and activities without the benefit of careful thought and analysis. The inability of health organizations to shed or acquire capacity easily lies at the root of this problem. It is unknown whether organizations with multi-year budgets behave differently from those with single-period budgets since multi-year funding is not used by any jurisdiction in Canada for primary operating funds. Likewise, we currently have little understanding of the impact on health organizations when significant revenues are generated from non-government sources. Few organizations fit this profile in Canada. Nevertheless, there are issues related to financial practices to which attention can be turned with a view to ensuring that organizational financial rules do not impede change. These issues include the following. • If we are to foster change and make informed financial decisions about the impact of change, additional investment in the functionality and capacity of the financial-information systems used in our health system is imperative. Without good financial data we cannot make good financial decisions. • The use of evidence-based funding methods (such as those that are driven by population characteristics, case-mix data, etc.) must be

78 lanMcKillop

•

•

•

•

•

encouraged. These methods hold the greatest promise for promoting predictability and equity in funding, particularly when used by health organizations with multi-sector responsibilities. Funding methods that have the risk of perpetuating past or emerging inequities (such as line-by-line or multiplier methods) should be discouraged. It may be possible to promote stability in funding by encouraging the use of multi-year funding horizons for costs related to system capacity, while retaining single-year funding horizons for activity costs. Further research and an open discussion of options are required. A planned response must be developed by governments to address situations when there is a gap between the resource needs identified by evidence-based funding methods and the size of a jurisdiction's funding envelope. When the gap is large, simply sharing the resource shortfall among the players compromises the efficacy of everyone's management control system. While health organizations should be encouraged to demonstrate their capacity to generate funds from non-government sources, significant differences in their relative ability to tap these sources will create situations of real or perceived inequity. Before we proceed with determining expectations in this area, the impact of various options should be explored empirically. Limited working-capital resources constrain the ability of many health organizations to invest in projects that will yield efficiency gains. This limits a health organization's propensity to initiate change. To overcome this problem, an innovation fund should be established. Health organizations could borrow to implement efficiency initiatives and repay the fund using the savings generated. To help health organizations develop management control systems that ensure the goals and objectives of funders are realized, the continued development of performance measures is required.

And probably most important, when introducing new elements to a management control system (such as new financial rules or practices), it is imperative to give careful attention to the unintended behaviours these elements may promote. This is to ensure that the system encourages the judicious and appropriate use of funds for the provision of health services and the realization of the strategic goals that have been set.

Financial Rules As a Catalyst for Change 79

Appendix Primary operating funding approaches used, by jurisdiction (in effect for the fiscal year ended March 2001) Jurisdiction

Funding scope

Funding method

Alberta British Columbia Manitoba New Brunswick Newfoundland and Labrador Nova Scotia Ontario Prince Edward Island Quebec Saskatchewan Yukon

Comprehensive Comprehensive Comprehensive Comprehensive Comprehensive Comprehensive Institutional Comprehensive Comprehensive Comprehensive Institutional

Population-based Line-by-line multiplier Ministerial discretion Line-by-line and population-based Ministerial discretion Ministerial discretion Global Ministerial discretion Global Population-based Ministerial + global

Source: McKillop, Pink and Johnson 2001, 46. NOTES

1 In this respect, the position advocated in this paper differs from the view of control, advanced by Anthony and Young (1995), as being a component of feedback (or past performance). 2 Service-specific funding is used in many jurisdictions to support the activities of agencies such as cancer-care organizations that have mandates that exist in parallel with the multi-sector health organizations. 3 For example, establish the transfer price equal to the variable cost of providing the service. Fund the capacity costs of providing the province-wide service only to those health organizations the government wishes to see providing this service. (Alternatively, a service-specific funding approach could be used.) 4 Many jurisdictions are to be recognized for the significant efforts they have made to develop MIS-compliant reporting systems (after CIHI 2001b); however, to a large extent these systems remain unable to provide valuable information at the service-recipient level. In addition, only a few jurisdictions have explored the development of information systems able to gather data on episodes of care that extend to include interactions with physicians, pharmacy systems, and the broader health sector. 5 Numbers have been chosen to facilitate visualization of the concept, not to reflect typical magnitude.

80 lanMcKillop 6 The use of multi-year operating funding was not identified in any province or territory in the McKillop, Pink, and Johnson (2001) study. 7 Using existing financial approaches, if an MRI cost $lmillion/year to own, and was placed in a community where 500,000 images were made per year, this MRI would be deemed to cost twice as much to operate per image ($2.00) as the same machine in a community where 1,000,000 images were made ($1.00 per image). Because adjustments have not been made for capacity, it is not actually known whether the per-image cost differs between communities. And even assuming that it might only cost $1.00 per image in the first community, it may be politically and medically appropriate to 'subsidize' the presence of an MRI in the community in the amount of $500,000. (All numbers are simplified for purposes of illustration.)

REFERENCES

Aas, I. H. 1995. Incentives and financing methods. Health Policy, 34(3): 205-20. Anderson, M., and J. Cosby. 1998. Evaluating an alternative funding plan.' Healthcare Management Forum 11(2): 28-32. Anthony, R.N., and D. Young. 1995. Management control in non-profit Organizations. 6th ed. Boston: Irwin McGraw-Hill. Barnard, C.L. 1970. The functions of the executive, with an introduction by K. Andrews. Reprint, c.1938; 30th edition. Cambridge, Mass: Harvard University Press. Canadian Institute for Health Information (CIHI). 2001a. Provincial/territorial government health expenditures, by use of funds, Canada 1975 to 2001 (Table C.4.1).' National Health Expenditures Estimates. Ottawa: CIHI. - 2001b. Guidelines for management information systems in Canadian health service organizations. (CD) Ottawa: CIHI. - 2001c. Hospital report 2001: Acute care. Ottawa: CIHI. Cawsey, T., G. Deszca, and H.D. Teall. 1994. Management control systems in excellent Canadian companies. Management Accounting Issues Papers (no.5). Hamilton, Ont.: Society of Management Accountants of Canada. Coast, J., et al. 2000. Conceptual and practical difficulties with the economic evaluation of health service developments. Journal of Health Services Research Policy 5(1): 42-8. Duckett, S.J. 1995. Hospital payment arrangements to encourage efficiency: The case of Victoria, Australia. Health Policy 34:113-34. Ezzamel, M., and M. Bourne. 1990. The roles of accounting information systems in an organization experiencing financial crisis. Accounting, Organizations and Society 15(5): 399-424.

Financial Rules As a Catalyst for Change 81 Flamholtz, E.G., T. Das, and A.S. Tsui. 1985. toward an integrative framework of organizational control. Accounting, Organizations and Society 10(1): 35-51. Glouberman, S., and H. Mintzberg. 2001. Managing the care of health and the cure of disease - part II. Health Care Management Review 26(1): 70-84. Helyar, C, et al. 1998. Benchmarking comparisons of the efficiency and quality of care of Canadian teaching hospitals. Hospital Quarterly 1(3): 14-25. McKillop, I., G. H. Pink and L. Johnson. 2001. The Financial management of acute care in Canada: A review of funding, performance monitoring and reporting practices. Ottawa: Canadian Institute for Health Information (CIHI). McKillop, I., G.H. Pink, J. Porter and E. Schraa. 2000. An examination of how hospitals use the reporting framework prescribed in the Ontario hospital reporting system. Technical Report 00-03-TR. Toronto: Institute for Clinical Evaluative Science (ICES). Merchant, K.A. 1990. The effects of financial controls on data manipulation and management myopia. Accounting, Organizations and Society 15(4): 297313. - 1997. Modern management control systems. Toronto: Prentice-Hall. Otley, D. 1994. Management control in contemporary organizations: Towards a wider framework. Management Accounting Research 5(4): 289-99. Palmer, G.R. 2000. Evidence-based health policy-making, hospital funding and health insurance. Medical Journal of Australia 172(3): 130-3. Pettersen, I.J. 2000. Accountable management reforms: Why the Norwegian hospital reform experiment got lost in implementation. Financial Accountability and Management 15(3/4): 377-96. Siciliano, J.I. 1997. The relationship between formal planning and performance in nonprofit organizations. Nonprofit Management and Leadership 7(4): 387403. Simon, R. 1995. Levers of Control: How managers use innovative control systems to drive strategic renewal. Boston: Harvard Business School Press. University of Toronto, Hospital Research Report Collaborative. 1998. Hospital report '98: A system-wide review of Ontario's hospitals. Toronto: Department of Health Policy, Management and Evaluation, University of Toronto. Williams, J.J, J. Newton and E. Morgan. 1985. The integration of zero-based budgeting with management-by-objectives: An empirical inquiry. Accounting, Organizations and Society 10(4): 457-76.

3 Governance and Management of Change in Canada's Health System JEAN-LOUIS DENIS

Pressure for significant transformations in health care delivery is present in the health care systems of all OECD countries and is giving rise to numerous reform projects (Saltman, Figueras, and Sakellarides 1998; Contandriopoulos and Souteyrand 1996; Contandriopoulos 1994). Although there is considerable consensus on the inevitability of change in health care delivery, it is important to acknowledge our considerable unfamiliarity with the processes by which to achieve such transformations. In addition, we know little about the nature of the changes actually achieved. Furthermore, we note that a number of these changes, including changes to the payment system for physicians, greater coordination of health care and services, and inter-professional collaboration, have not been successfully implemented. At the same time, many ideas concerning the need for and the relevance of certain changes have been suggested and are becoming more widely accepted. Thus, our analysis seeks to gain a better understanding of the conditions that would make it possible to implement the desired changes and transformations. The commission on the Future of Health Care in Canada was concerned by the impact of governance structures and management approaches on the ability of the health system. This is an important question because it relates directly to our ability to shape the organization of the health care system. More specifically, it enables us to address the factors that are conducive to the production, distribution, and institutionalization of change in this system. This more general question is complemented by three specific questions concerning the appropriate-

Governance and Management of Change 83

ness of the procedures used to monitor health care and services; the system's ability to manage the integration of health care and services; and the system's ability to change on a broader sense. These specific questions will be addressed throughout this paper. We concentrate primarily on the organizational and inter-organizational dynamics that can foster change to varying degrees. To this end, we first identify the problem of organization in the health care system, and put forward a concept of governance aimed at a clearer understanding of the relationship between policies and organizations and the ability to change. Next, we discuss three cases we feel are illustrative of the relationship between governance and change. The first concerns significant structural change in a health care system linked to the implementation of regional governance. The second involves experimentation as a strategy for producing change, using the example of the Health Transition Fund (HTF). The third presents an entrepreneurial method of producing change, focusing on organizational leadership. These three change strategies open up a debate on the relevance of forms of governance, the revitalization of change, and the role of management in producing change. Change and Governance For the purposes of this study, change is defined as the modification of a system of values - of agents' perceptions of a situation, structures, or practices in a given organizational field. Governance relates to the systems and practices that allow agents to develop a plausible perception of their future, to design and implement effective change strategies, and to rely on values that create trust and solidarity (Hatchuel 2000; Aggeri and Hatchuel 1999; Gallon, Lascoumes, and Barthes 2001; Chevalier 1996; Lascoumes 1996). Saltman and Ferroussier-Davis (2000) suggest that the legitimacy of governance will be even greater if it relies on a strong appreciation of collective responsibility. The theme of organizational change is the subject of numerous works (Brock, Powell, and Hinings 1999; Powell, Brock and Hinings, 1999; Hinings and Greenwood 1988; Greenwood and Hinings 1996; Pettigrew, Ferlie, and McKee 1992; Kitchener 1999). Change goes hand in hand with the uncertainty in clinical and administrative practices and in agents' behaviour. The problem of change is a central dilemma for reform, in part because of the multiplicity and complexity of the relationships between principals and agents (Eisenhardt 1985) and the

84 Jean-Louis Denis

difficulty in establishing appropriate mandates for organizations and professionals (Denis and Valette 2000). Furthermore, it is important to note that the process of change is not separate from the social and organizational contexts in which it takes place. Evaluations of health policies and programs have provided many examples of the impact of the organizational context on the outcome and effectiveness of interventions (Scheirer 1981,1987; Giacomini et al. 1996). A number of sociology studies (Perrow and Guillen 1990; Meyer and Zucker 1989) rely on the concept of 'organizational failure' to account for the difficulty that many organizations have in adapting to their environment. Thus, the concept of change relates to our collective ability to implement the new policies and programs necessary for a sustainable development of the health care system. The concept of change is in itself quite imprecise. For example, basing themselves on the work of Miller and Friesen (1984) and Tushman and Romanelli (1985), Greenwood and Hinings (1996) propose a distinction between convergent and radical change, on the one hand, and evolutionary and progressive change, on the other. These distinctions are important for the purposes of our study. Radical change means changing the direction of organizations and of the health care system. It disturbs both the organizational structure and the agents' perceptions of their roles, practices, and fields of activity. Convergent change consists in making adjustments within existing structures and perceptions. It is more akin to a social- reproduction process (Bourdieu 1980; Bourdieu and Wacquant 1992; Giddens 1984). This distinction leads to a fundamental question for future health care: Should radical changes be implemented or should we rely on convergent change in order to ensure sustainable development? The distinction between progressive change and revolutionary change relates to the speed at which change is implemented. Do we need to act quickly or should change be introduced gradually in order to ensure sustainable development? Authoritarian or coercive (top-down) strategies or participatory strategies can be used in order to promote change. Analysis of the three cases reveals a change process that varies in terms of the nature of the change, the rate of change, and the degree to which it is hierarchical or authoritarian in nature (Minzberg, Ahlstrand, and Lampel 1999). We suggest that a realistic change for health care organizations and for the health care system in general would focus on radical changes proceeding in a progressive manner, by relying on strong, emergent dynamics. Quite simply, our initial hypothesis is that the considerable and recurring effort

Governance and Management of Change 85

aimed at reforming health care systems in various OECD countries (Saltman and Figueras 1997; Drache and Sullivan 1999) are indicative of dissatisfaction with the changes that have actually been achieved. Bureaucratic or political reform projects that seek to introduce rapid change to the entire health care system result in major upheavals, without necessarily effecting any in-depth change in practices and procedures (Ferlie and Fitzgerald 2002). Thus, it seems important to consider alternate ways of effecting change that focus more on incentives and learning. Several works in the field of political science seek to define the structural conditions that govern the evolution of national health care systems (Ruggie 1999). These give a good description of the evolutionary and convergent nature of change over long periods of time and they provide valuable information about the bases of current organizational methods. Our approach involves a shorter time-frame and seeks to identify resources and processes to effect change in a deliberate manner, in order to limit the problems associated with the operation and legitimacy of the health care system. Thus, our study concerns the efforts directed at transforming Canada's health care system. These have had to deal with a double imperative: to renew or adapt the provision of health care while at the same time controlling costs more effectively. There is a constant tension between these two objectives and there is no easy solution to the dilemma. In addition, we have decided not to analyse specifically changes to the health care system that could have resulted from challenging the very logic of this system. If we were to address health determinants or population health, this could significantly alter our perception of health care and, consequently, our perception of the role and operations of the health care system (Hayes and Dunn 1998; Evans, Barer, and Marmor 1994). To meet the need to regulate organizations and the health care system, governance relates to the organizational design of the health care system and the sharing of responsibilities and activities among its various entities, the production and information-distribution systems and mechanisms, and the modes of financing organizations and professionals (Contandriopoulos, Denis, Touati, et al. 2001). In order for governance to be effective, it must also address the consistency of decisions made and actions taken at the clinical, organizational, and general policy levels. The way in which health care systems will structure governance is a social and political issue in itself. For some, governance is, first and foremost, a matter of democratization of political

86 Jean-Louis Denis

and administrative structures (Bickerton 1999; Bohman 1996). It is an instrument for acting on the distribution of influence in a given system and for producing radical change. In addition, governance clearly poses the problem of managing uncertainty. Exercising governance means getting involved and forming new relationships between agents in order to determine appropriate action and ensure the viability and legitimacy of the system (Gallon, Lascoumes, and Barthes 2001). In this respect, democratic government includes a dimension of learning, since it makes it possible to generate new political options for a given social system for which one cannot assume that any agent possesses all the resources and expertise required to design and implement solutions to collective problems (Hatchuel 2000; Lascoumes 1996; Alter and Hage 1993; Bryson and Crosby 1993). These considerations cast the role of governance in terms of strategies that lead to change to the health care system on a functional level, and in terms of acting on the understanding that social agents have of the conditions for legitimacy of this system (Bickerton 1999; Hatchuel 2000; Gallon, Lascoumes, and Barthes 2001). To renew governance, that is, the ability of health care organizations and systems to assess, induce, and direct change, several strategies or levers can be used: incentives, influence, regulation, formal authority including structural change, and moral commitment (Valette, Contandriopoulos, and Denis 2000; Contandriopoulos, Fournier, Denis, et al. 1999; Hood 1983; Ferlie et al. 1996). Our analysis concerns three variable strategies to revitalize governance: regional governance, which consists in structural modifications to the health care system in order to create new structures capable of processing and regulating demands at mid-level between the front line of services and the stragetic summit; which relies primarily on incentives and commitment; and leadership, which uses influence, vision, and formal authority to produce change. Establishment of Regional Governance and the Ability to Change By creating new organizations, regionalization policies seek to establish enhanced strategic capacities at the intermediate level of the health care system (Denis and Valette 1997). Establishing regional governance involves both centralization and decentralization (Lemieux 1997). The central government delegates some of its decision-making and action powers to regions. Local communities or health care institutions also

Governance and Management of Change 87

delegate part of their decision-making and action powers to regional authorities. Some view regionalization as a means of performing various functions more effectively, including resource allocation services planning, and facilitating public participation in governance (Smith 1979; Rondinelli, 1981). In Canada, regionalization has been a major form of experiment in health policy. Nine of the ten provinces have adopted a regionalization policy, thus relying on the potential of this structural change to bring about reform (Bickerton 1999; Church and Barker 1998; Mahtre and Deber 1992). Various studies of this policy reach radically different conclusions. For some (Church and Barker 1998), while regionalization has not yet proven itself in terms of coordination and integration of health care and services, it is a rational instrument for achieving the objectives of efficiency and effectiveness in service delivery. For others (Bickerton 1999; Mahtre and Deber 1992), regionalization has not brought about the greater democratization of administrative structures and service systems expected. According to them, regionalization is an instrument for government renewal, which makes it possible to avoid exchanging a traditional bureaucratic government for a competitive one likely to relinquish its responsibilities for social policy (Bickerton 1999; Trottier et al. 1999). These macroscopic analyses offer an overall assessment of regionalization policies, showing that such policies have pursued objectives difficult to reconcile with each other and that the various Canadian provinces have adopted diverse organizational arrangements to establish regionalization (Denis et al. 1998). They also suggest that regionalization has had imperfect results in producing change, although it may have made it possible to initiate reform and to provide an avenue for action in tandem with central governments. Other works (Lomas, Veenstra, and Woods 1997a,b,c; Lomas, 1997; Lewis et al. 2001) address the attitudes, motivations, and perceptions of board members of regional authorities. These studies provide a portrait of regionalization in some Canadian provinces from the perspective of regional stakeholders. They demonstrate that board members take their mandate seriously and devote a considerable amount of their time to it, that they are middle-class, and that they have had previous experience as members of other types of boards. Board members recognize that they defend particular geographical or group interests. They feel less equipped technically in terms of needs analysis and planning; however, they are well informed about the processes and rules of delib-

88 Jean-Louis Denis

eration. They feel that they are better informed about costs and service utilization than about service benefits, citizens' preferences, and the opinions of key players in the community. These results suggest that regional boards are more likely to respond to the wishes of the central government than to local preferences. Although such boards acknowledge that they do not have all the information they deem useful for decision-making purposes, they believe that, on the whole, they are still able to make good decisions and support reform projects. In addition, they demonstrate a lack of consensus on the central government's role in terms of support for or obstacles to local reform. Most respondents believe that they are held accountable for elements that are not entirely within their control and that the central government's rules are too constraining. In general, these empirical works on the viewpoints of agents of regional governance suggest that regionalization policies have produced some changes. They have made it possible for regional agents to participate in boards, to become involved in reform projects, and to manage a health care policy rather than a health policy. Regionalization As a Lever to Produce Structural Change Drawing on empirical studies conducted in different regions of Quebec and France (Denis, Langley and Contandriopoulos, 1995; Denis et al., 1998; Denis, Langley, and Contandriopoulos 1998; Denis, Contandriopoulos, Langley, et al. 1999; Denis and Valette 1997, 2000; Denis et al. 2001), we analysed the role of regional authorities in restructuring the provision of health care. Our findings reveal certain dynamics that link regional governance and change, namely, the role of reformist ideologies and structural changes in reform. In Quebec, regionalization found its legitimacy in part by evoking a specific reformist ideology (Rocher 2001), that is, population health. This trend also had a strong impact on the reformist position of several Canadian provinces (Hayes and Dunn 1998; Mahtre and Deber 1992; Davidson 1999; Casebeer and Hannah 2000). Such a position, if translated in operational policies, may represent major changes for organizations and the health care system. It can lead to substantial change in the resource-allocation scheme within a region, hence the distinction usually made between the preventive and curative sectors of service delivery, or to a focus on cross-sectoral policies and programs in areas such as access to income, housing, or education. Within this perspec-

Governance and Management of Change 89

tive, the regional authority effects change by relying on a stronger health care policy that focuses on the efficiency and effectiveness of health care and ,by assuming a more assertive advocacy role towards other authorities for pro-health actions (Evans, Barer, Marmor 1994). At first glance, Quebec regional authorities took strong action to restructure health care provision and, through their public health management, played an advocacy role. In concrete terms, the strengthening of regional governance in Quebec in the 1990s led primarily to tight control over resources allocated to health care as well as a reallocation in the areas of social services, prevention, and out-of-hospital health care programs. In other words, regionalization coupled with a strong policy to control central-government expenditures made it possible to limit the development of in-hospital health care delivery. By way of illustration, between 1995 and 1998, the Montreal area experienced a 16 per cent decrease in hospital budgets and a 25 per cent decrease in the number of hospital beds (Observatoire sur la transformation des organisations de sante: www.santemontreal.qc.ca). In addition, the 1990s were marked by a number of institutional mergers or consolidations, which significantly changed the autonomy of these organizations and the structural landscape of the entire health care system. There is no need to outline all of these changes here. Among other objectives, structural changes sought to augment the use of outpatient practices, including day surgery, and to curb the imbalance between the provision of hospital care and community care. Some of these changes are in line with trends in the health care systems of other industrialized countries, while demonstrating the ability of regional authorities to induce structural change and, in so doing, to promote change in hospital practices (Denis, Langley, and Contandriopoulos 1995). Overall, in spite of a reformist ideology in Quebec focused on 'population health/ regional authorities have been a special instrument of structural change at a time of significant resource constraints (Lomas 2001). They likely helped to speed up certain trends in the evolution of hospital care provision, such as the development of outpatient health care services. In addition, outside large urban areas, they promoted a territorial health care framework on a smaller scale by merging health care institutions serving small areas. Such changes lead to two observations about the regional government's ability to produce change. First, regional health authorities took strong action in order to restructure health care delivery without necessarily being able to pro-

90 Jean-Louis Denis

mote or effect change in the practices associated with health care and services coordination between collaborating institutions (Leatt, Pink, and Guerriere, 2000; Denis, Lamothe, Langley, et al. 1999). In order for regional governance to be effective, it is important that the various levers, including the compensation of medical resources and the funding of medical practices, come under the regional authority (Contandriopoulos, Denis, Lamarche, et al. 2002; Contandriopoulos, Denis, Touati, et al. 2001). Some provinces have chosen to integrate health care institutions under the regional authority, and thus to abolish independent boards. To our knowledge, there are few, if any, studies that address systematically the potential of this integration measure while comparing its benefits and limitations with a regionalization model where institutions preserve their autonomy. Logically, it would seem that an integration such as this would promote the coordination of health care and services yet adversely affect the identity of organizations and the motivation of professional and managerial teams (Denis, Lamothe, Langley, et al. 1999). Second, the province of Ontario, without having established regional governance, was also able to intervene decisively in health care structures (see the HSRC report, March 2000). There are, therefore, several governance options to bring about structural change. However, it is important to remember that, in Quebec at least, such actions directed at health care institutions had previously been impossible. The limitations of restructuring for adapting the provision of health care suggest that there can be significant change without necessarily transforming the relationships between organizations and professional groups in order to coordinate health care and services more effectively. Regionalization As a Democratization Tool As we mentioned at the outset, for many, the issue of governance is essentially a problem of citizen participation in the conduct of public affairs. According to this view, the desired change will enhance citizen participation in governance. In a previous study, we addressed the question of the role of public consultation in the decision-making process for difficult decisions such as closing public hospitals (Denis, Langley, and Contandriopoulos 1995, 1998). This study of a decisionmaking process in a large metropolitan area revealed that the regional authority supported public consultation within a rigorous technocratic framework. The consultation process was well planned and used for

Governance and Management of Change

91

purposes of information (Forest et al. 2000) as well as persuasion, while leaving the regional authority open to demands from various segments of the community. Public decision making was a way of demonstrating the merits of the policy advanced by the regional authority and of limiting the pressure that different groups could exert directly on the decision-making process. We also observed at that time that a crucial component of a regional authority's ability to handle external pressure was the unwavering solidarity of its management team and board. Thus, in this case, social technology associated with democratic governance, that is, public consultation, coexisted with a competent and cohesive technocratic and administrative apparatus. That observation led to the suggestion that regional governance is definitely a hybrid mode of action seeking to combine the advantages of democratic participation, managerial control, and political negotiation with influential groups in order to have a legitimate impact on the provision of health care (Denis 1999; Denis, Contandriopoulos, Langley, et al. 1999; Denis and Valette 1997). When the regional authority has no power of taxation, the central government's directions and desires largely control its governance capacity. In addition, the regional authority sometimes requires the cooperation of institutions to bring about change, while minimizing the negative impact of certain constraints on the availability and accessibility of health care services. This is where political negotiations with institutions present on its territory come into play. Among other considerations, public participation could be used to anchor the regional authority in a local political culture in order to establish its legitimacy and possibly reduce its dependence on the central government. This proposal is consistent with the analyses of Bickerton (1999), who views regionalization as a means of renewing government and who refers to a realistic conception of public participation in health care systems (Contandriopoulos, Denis, and Langley 2001). Regional Governance and Change Does regional governance promote or permit significant changes in citizens' relationships with the government and its administrative apparatus and in the provision of health care? The potential of regional governance cannot be assessed without considering a central government's initiatives and capacity for action. In Quebec's case, decisions made by the central government, such as

92 Jean-Louis Denis

the early retirement of health care professionals, including nurses, adversely affected the freedom of regional authorities to bring about change. The tensions created by such decisions in organizations' operations and the potential patient dissatisfaction with service availability and accessibility have dealt a severe blow to the legitimacy of reform. In fact, one basic problem of governance and change is maintaining the legitimacy of regional-reorganization projects when decisions from higher administrative or political levels threaten coherence and the resources allocated to these projects. In addition, the evolution of regionalization in Quebec demonstrates quite clearly that regional action in the health sector has become more fragile and less legitimate in the eyes of the central government (Turgeon 2001). This trend is also evident in other provinces (Lomas 2001). The Quebec example shows that ensuring the political conditions that make it possible to renew lasting governance is a fundamental problem in the management of change. Once again, we do not think there are simple solutions to ensure sufficient political support for regionalization and decentralization policies. In general, the relationship between regional governance and change can be expressed as follows. Establishing strategic abilities at the regional level made it possible for the health care system to come through difficult times and bring about significant restructuring. However, health care organizations and professionals have been severely affected by these change processes, which have been rapid rather than progressive. The structural changes are probably convergent in the sense that they can strongly threaten the existing equilibrium without significantly reconstituting the provision of health care, including the coordination of health care and associated services. The approach used to bring about these changes was authoritarian or hierarchical in several cases, because regional authorities had quickly to impose strong constraints on health care organizations. Regional governance was built, according to the circumstances, with or without the presence of independent institutions and was exercised within the context of maintaining considerable independence for the medical profession. This analysis of the relationship between regional governance and change leads us to make recommendations that convey a particular vision of organizational change. Following a significant phase of structural change, regional governance must move towards a supportive and catalytic role in terms of the dynamics that bring about changes in practice, enhanced cooperation between agents and organizations,

Governance and Management of Change 93 and, consequently, more effective integration of health care and services. In order for such changes to take place, regional governance must have strong and lasting legitimacy in the health care system, and it must approach structural change as the next stage after the adoption, by agents, of an ideology that promotes the integration of health care and services. Proposal 1: To produce transformation, regional governance needs to target organizational and professional practices. Proposal 2: To produce transformation, regional governance, in addition to carrying out its responsibilities relating to control, needs to play a supportive and catalytic role with regard to the organizations within its territory. Proposal 3: To produce transformation, regional governance needs to rely on a combination of strategies in order to revitalize and channel change, including incentives, training, political negotiation, and structural modifications. Proposal 4: To produce transformation, regional-governance projects need to receive consistent and sufficient support from the central government. Proposal 5: To produce transformation, regional governance needs to gain control of various levers that enable it to promote the integration of health care and services in the community. These include the remuneration of medical personnel, closer cooperation with highly specialized health care facilities, and methods of funding that promote institutional cooperation. Proposal 6: To enhance its ability to effect change, regional governance needs to stimulate local, public participation so that reform projects can realize their full potential and be sustained as needed without exhausting human resources in the health care system. Experimentation As a Mode of Producing Change in the Health Care System This section deals with a particular mode of producing change, experimentation through funding pilot projects in the health care system. Experimentation has long been viewed as a special learning mode in complex social systems (Campbell 1969). In order to discuss the relationship between experimentation and change, we will mainly use the results of a cross-section analysis of Health Transition Fund projects carried out in Quebec (n = 40 projects, Desbiens, Dagenais, Joubert, et al. 2001; Denis 2001; Denis, Lamothe, and Langley 2000, 2001) and the

94 Jean-Louis Denis

results of studies of specific projects (Touati et al. 2001; Beland et al. 2001a,b; Lamarche et al. 2001a,b). This is not an in-depth look at the HTF experience in its entirety but we think it offers some valuable observations on the relationship between experimentation and the production of change in health care systems. Here, we are interested in a situation quite different from the one analysed previously; we are dealing not with projects that create new structures and forms of governance but with projects that seek to produce local change by maintaining a relative consistency in the current boundaries and rules of the health care system. Thus, the analysis of the HTF program makes it possible to assess the potential of convergent and progressive changes in transforming health care provision. These changes can, through a domino effect, produce more substantial change in the long term in the organization and operation of the health care system. However, as we shall see, local initiatives often require transformations on other levels in order to promote lasting, substantial change. Some of the interest in these projects lies in the frequent involvement of health care professionals or providers in dynamics of change that are more subtle than those of major reform projects undertaken by central governments and public bureaucracy (Ferlie and Fitzgerald 2002, Brunsson and Sahlin 2000). In this section, we begin by briefly presenting the context and diversity of HTF projects in Quebec. Then, we discuss the transformation potential of practices and organizations associated with these projects, ending with a discussion of the resources and factors that are conducive to producing change in the health care system and the role of experimentation in these processes. Presentation of the HTF Experience HTF projects took place within a specific economic and political context. They represent a special approach adopted by the federal government in Canada in recent years to finance targeted programs in a context where its financial involvement in the health care system has decreased. In addition, provincial health care systems and organizations that received funding during this experiment were at a relatively low ebb when the HTF program began in 1997. Thus, change represents an enormous challenge. Two important questions underlie the HTF program: How can we ensure that the resources allocated to innovative projects are not redirected into the standard elements of the system and, in particular, not used to fill the gaps caused by peri-

Governance and Management of Change 95

ods of fiscal austerity? How are we to ensure that human resources, under great pressure to deliver health care and services, become involved in innovative approaches? Finally, assessment processes associated with all HTF projects allow us to draw conclusions about the relationship between the production of knowledge and the production of change. Most of the projects (31 /40) financed by the HTF in Quebec involved the integration of health care in order to enhance the efficiency or quality of services provided. A number of others (17/40) involved the introduction of technological tools such as new information systems or clinical-practice guidelines. In addition, the projects also differed in their choice of intervention targets as priorities. These concerned such things as a specific clientele (e.g., the frail elderly, or monitoring persons with a specific disease), or the organization of health care and services for a community or territory with mixed clienteles and pathologies (Desbiens, Dagenais, Joubert et al. 2001). The diversity of project types highlights the considerable degree of interdependence at work in the process of effecting change in health care systems. In fact, it is difficult to limit change to one specific field - to clinical practices, for example without having to make changes at some time or another to governance, including resource allocation and economic incentive (Contandriopoulos, Denis, Touati, et al. 2001; Contandriopoulos, Denis, Lamarche, et al. 2002). This empirical observation is consistent with the idea that organizational change is an overall process requiring new alignment between agents' values and interests and organizational design in the broad sense of the word (Greenwood and Hinings 1996; Mintzberg, Ahlstrand, and Lampel 1998,1999). Different Experimentation Modes to Produce Change HTF projects in Quebec gave rise to three distinct experimental approaches: prototype grafting, the introduction of new interventions with no direct patient impact, and the renewal of current practices and procedures. Here, we discuss change processes associated with prototype grafting and with renewing current practices. The experimentation mode that involves introducing new technologies or interventions for professionals will be discussed in more detail below in the section analysing internal-change processes in health care organizations. Prototype experimentation consists in grafting a new service-delivery model onto the usual operation of the health care system (Beland et al. 2001a,b). The SIPA project, for example, was a pilot project aiming at

96 Jean-Louis Denis

developing integrated care for the frail elderly in some participating community health centers of the Montreal area. Prototype experimentation is appealing because it seems to offer a more controlled vision of change and encourages stakeholders to invest considerable energy into project implementation. With this mode, it may seem easier to intervene in favour of change during the experimentation. However, prototype experimentation raises certain challenges. The prototype itself represents an alternative to usual organizational modes and health care and service-delivery modes. Thus, revisiting the model at the end of the experimentation clearly raises the question of the capacity for replacing the usual organizational or intervention modes by the prototype replacing the usual home care model by a SIP A-type model. Implementing a prototype during experimentation may also be incomplete. Several HTF projects involved testing a capitation-financing formula which is difficult to achieve locally. It also raises the issue of harmonizing a health care and services model that is developed for a specific clientele with the health care system as a whole. Several pilot projects represent just one specific component of the more integrated health care models (Denis and Lamothe 2001). As an example of prototype experimentation, the SIPA project clearly illustrates the interdependence involved in effecting change. It demonstrates that a planned and conscious approach to change under the responsibility of committed professionals can produce practice-related changes during experimentation. Yet, in order for it to continue and remain consolidated, local change must be based on systemic changes. Thus, there are major uncertainties as to the capacity of organizations and local agents to institutionalize significant changes. To act towards change, it seems useful to consider the complementarity of the participative and emergent local modes of change using a more directive and centralized approach with the potential to modify some health care system rules and enhance the system's ability to adopt innovations. Unlike prototypes, other experimental projects push change towards resources and current health care system structures. These projects may be based on an injunction from a governing authority, for instance, to seek alternatives to institutional mergers in order to attain the objectives of resource downsizing and enhanced coordination of health care and services (Lamarche et al. 2001 a); on perceived threats in the community such as the possible closure of a small hospital (Touati et al. 2001); or on a strong conviction on the part of local stakeholders that it is possible to do more with the resources at hand. It involves a change

Governance and Management of Change 97

that is, by nature, locally entrenched and has strong emergent dimensions, in the sense that the model evolves through time according to local and relatively unplanned dynamics. It appears less planned than the prototype experimentation and seems more like a resourceful do-ityourself experimentation. Several lessons may be drawn from the dynamics associated with this type of experimentation (Lamarche et al. 2001a,b; Denis, Lamothe and Langley 2000; 2001a). Various initiatives may be launched to produce change; however, structural change seems to have limited potential. In fact, in order for structural changes to yield results in terms of health care and services coordination, they must be in synergy with the existing dynamics of cooperation among stakeholders. Approaches to structural change have a greater chance of being productive if they complement emergent changes already in place. Moreover, the changes that benefit patients are changes to professional practice - the way professionals provide health care and services. This is also one of the strong points in experimenting with new prototypes of health care and services delivery. In addition, in order for the dynamics between agents to converge, including cooperation among doctors, organizations, and other professionals, incentives must become the focus of change. In this respect, change cannot be strictly structural or voluntarist, it must rely on a synergy between these two factors and the incentives (Denis, Lamothe, Langley, et al. 1999). As well, involvement of local promoters in the various projects facilitated consolidation of one vision of changes to be made and lent legitimacy to developing and setting up implementation strategies. We will revisit this question below in the section dealing with organizational leadership. A plausible strategy for managing change is being developed in the health care system and organizations using these two modes of experimentation (prototype and do-it-yourself). HTF projects have led to changes that are closer to an emergent mode, that is, a mode that distances itself from the bureaucratic downsizing which gave precedence to the structural-change mode in the health care system (Denis, Lamothe, and Langley 2000). They clearly show the importance of not approaching change strictly as a top-down process, but instead relying on the emergent (bottom-up) dynamics created by professionals working closely with service and care delivery (for example, the Umbrella Project in Alberta). This more flexible approach to change, combined with the need to assess innovative projects, may provide considerable new knowledge and the emergence of new alliances between practising and

98 Jean-Louis Denis

academic circles (Denis, Lamothe, and Langley 2001 a). It is also consistent with an adaptive perspective of organizational design that is more in line with the imperatives of professional production and knowledge-based management of organizations (Alvesson 2001; Anderson and McDaniel 2000; Empson 2001 a,b; Lowendhal, Revang, and Fosstenlokken 2001; Lamothe 1996). However, the question of the dissemination and durability of local change remains unanswered. Capacity to Institutionalize Change Three dilemmas limit the potential for change associated with experimentation in the health care system (Denis, Lamothe, and Langley 200la). The first dilemma concerns the very nature of the organizational models or structures arising out of experimentation. Several projects aim to enhance coordination between organizations and must constantly adapt to an essentially vertical structure in the health care system. The second dilemma lies in the conditions for committing the stakeholders to the experiment. In several projects, uncertainty as to the continuity of an experiment, which is a fundamental characteristic of this mode of change, undermines the commitment and may adversely affect the conversion of experimentation resources into permanent transformations to organizational modes in health care and practices. A third dilemma deals with the issue of the fragmented health care system, raised by numerous health care commissions in various Canadian provinces (Fyke, Clair, etc.). The HTF program gave rise to multiple experiments that are difficult to understand and integrate into an overall policy for the health care system. Therefore, experimentation requires the structuring and controlled use of the emergent dynamics and the more centralized and generalized change phases in the health care system. This culture of experimentation must be both strengthened and maintained as a special objective of governance in the health care system. To be more precise, strategies must be developed to enable an ongoing, effective interaction between governance, which is responsible for ruling on the relevance, sustainability, and funding of experiments, and promising local initiatives, without unnecessarily restricting their possible development. Courses of action are being developed in order to renew the roles of regional governance, studied in the previous section on, among other things, the relationship between knowledge production and knowledge use, the training of agents for change, and the decision-making process in experimental programs.

Governance and Management of Change 99 Resources for Change

It is possible to define resources for change from an analysis of the HTF projects (Denis, Lamothe, and Langley 2001a; Denis 2001). The HTF experience shows the importance of bringing together operators, professionals, and other staff to develop change projects. Physician cooperation is essential yet difficult to achieve in many projects. For the most part, HTF projects involve a decentralized mode of change, which has certainly been instrumental in introducing diversity into a system that is rather focused on uniformity. HTF projects have often focussed on the emergence of flexible structures, such as coordinating or consultative committees that can promote change when they have access to the necessary resources. These flexible structures are obviously much less disruptive for the agents than more rigid structural changes. Moreover, they are more effective in an experimentation context, which makes it easier to monitor certain resources through such structures. The HTF experience likely teaches us little about the effectiveness of these flexible structures in a usual context in which such structures have to mobilize resources under institutions' control. In addition, the HTF projects are rather clear evidence that agents need resources and time in order to develop changes to professional practices. Change is also fostered through the presence of promoters or leaders who have a vision of the transformations to be effected. This vision is akin to a reformist ideology but one whose outlines are locally defined and thereby enable agents to identify with the change project. These resources can make it possible to get the most of the experiment. The Complexity of Projects and the Potential for Transformation Therefore, HTF projects have represented an extremely varied program in terms of change management. In the health field, and to consider the interdependence involved in the production of change, we defined three areas where change must come about: the clinical system, which includes the way of providing health care and the relations between professionals; the administrative system, which includes the terms and conditions for funding the management system and the information system; and the collective system of representations and values (Contandriopoulos, Denis, Touati, et al. 2001). An experimentation project will be all the more complex as it expressly seeks to make changes to the three areas listed. In light of the analysis of the HTF

100 Jean-Louis Denis

experience, it appears to be difficult to make any significant change in one area without disrupting the others. An analysis of the HTF experience also allows us to form a hypothesis about the relationship between the complexity of a project and the production of change. Projects dealing with the clinical and administrative aspects, with the values and relationships between the local and the overall operation of the health care system, are eminently more complex than those limited to a localized clinical dimension. A simple project, such as an education program for asthma patients, will produce predictable changes but will probably be a source of only rather moderate transformation of the health care system. Conversely, a complex project that seeks to make changes in the different areas that we have just described, such as the restructuring of health care and services of an entire rural territory, will have a high potential to transform the health care system but may run into a number of difficulties in fulfilling this potential. These statements clearly show the importance of incorporating change projects in the long term since that is when the most significant transformations may come about. Including change on a time horizon is appropriate in that arguments may be made about the ability to produce more radical changes in a progressive manner. To date, the health field has provided several examples of the difficulties involved in producing radical change at the administrative level and in terms of relations between professions. A governing authority's support of experimental projects could likely play an instrumental role in making radical and progressive change in the health care system a reality. Yet this begs the question of whether governing authorities have the capacity to participate in the long term in implementing such changes. From our perspective, when projects are as ambitious as the ones we have just described and when they are based on strong professional dynamics, there is a potential for major transformation. One important target of managing change in the health care system is an enhanced ability to monitor local initiatives and their relations with changes that must be produced at other levels of human- and financial-resources management. This analysis of experimentation as a mode of producing change leads us to make the following proposals: Proposal 1: To produce transformation, it is important to promote local and emergent change. Proposal 2: To produce transformation, it is important that strategies be

Governance and Management of Change 101 developed to ensure the involvement of professionals and others in local change projects. Incentives appear to be essential in ensuring the commitment of independent professionals. Proposal 3: To produce transformation, experimentation alone will not suffice; there must be reliance on the roles, resources and skills of central and regional governance authorities in order to promote, monitor, disseminate and institutionalize experimentation. Proposal 4: To produce transformation, it is important for all governing authorities to develop 'venture capital' type programs in partnership with health care organizations. Proposal 5: To produce transformation, experimentation must rely heavily on the development of agents' cognitive capacities. By cognitive capacities, we mean a capacity to visualize complex change processes. In this respect, the training of agents involved in change management and implementation appears to be an important aspect. Proposal 6: To produce transformation, experimentation must be part of a cyclical process alternating between the emergence and planning phases of change management. Proposal 7: To produce transformations, it is important to manage the coherence of the local project and its place within the systemic context. Leadership As a Strategy for Change Organizational leadership represents an entrepreneurial mode of producing change. In this section, we examine the factors that are conducive to the emergence of an organizational leadership that favours change. Strategic Change and Leadership in Health Care Organizations For over ten years now, we have followed change processes in health organizations, particularly in university hospitals (Denis, Langley, and Cazale 1996; Denis, Lamothe, and Langley 1999,2001b; Denis, Langley, and Pineault 2000). Recent works (Brock, Powell, and Minings 1999) suggest that professionals, including doctors, see the context of their practices undergoing transformation while managing to maintain strong autonomy. Currently, the recruitment of medical workers and the mobility of this professional category suggest a growing autonomy of doctors vis-a-vis organizations. This raises the question of what organizational conditions are required for producing and controlling change in such a context.

102 Jean-Louis Denis

Several studies shed light on the role of managers in steering organizations and implementing changes (Hambrick and Mason 1984; Finkelstein and Hambrick 1996). This work is often developed around an individualistic and grandiose vision of organizational leadership, a vision ill-suited to the workings of complex organizations marked by a fragmented authority structure. In contrast, we focus on the specific characteristics of leadership in these types of organizations. The concept of collective leadership highlights the importance of a multiplicity of agents in the development of change projects. It is based on the necessity of ensuring that the clinical and administrative areas are represented in leadership positions. Organizational leadership would be based on the promotion both of diversified skills and of different sources of legitimacy. To harmonize these different resources for the exercise of leadership, the functional properties of a collective leadership can be defined as follows (Denis, Lamothe, and Langley 2001b). To enhance the potential for change, collective leadership must be distinct, specialized, and complementary. The differentiation of leading roles is related to an effective division of roles among the leaders of the organization. This differentiation could prevent unproductive rivalries among those individuals. There may also be a greater capacity for leadership if leaders have specialized and complementary roles, so that they may deal with all of the issues posed by change. Issues in the Development of a Collective Organizational Leadership The collective-leadership approach allows for an understanding of the role of change agents. There needs to be strong leadership for change to occur. However, the existing leadership may be a source of inertia; in such situations, it must be reconstructed and considered as a prime target for change. Three factors threaten the stability and effectiveness of a collective leadership: rivalry between leaders to dominate organizational processes; separation of the leaders' strategic project from the organizational base; and weakness of the change project in relation to environmental requirements. These factors can prevent an organization from building enough leadership skills to manage significant changes. Specific organizational processes lie behind these three factors that threaten the practice of a collective leadership. Rivalry between leaders may be linked to structural factors such as the scarcity of opportunities or the identification of leaders with different segments of the organiza-

Governance and Management of Change 103

tion. The separation of the leaders' project from the organizational base may depend on the inability to mobilize professionals in collectiveleadership development. Lastly, the significance of power struggles and divergent opinions on the future of an organization may push leaders to make significant compromises with respect to the scope of a change project. These compromises may jeopardize the consistency of the organizational strategy with environmental trends. The importance of the vision in managing change processes (Kotter 1995) takes on its full meaning insofar as it can prevent change projects from drifting too far off track. Change Management and Organizational Leadership Acknowledging the importance of leadership in health care organizations has several consequences for change management. It shows how the reconstruction of management teams in organizations is closely linked to the ability to produce change. An empirical analysis of the processes of merging teaching hospitals also suggests that any major change is accompanied by destabilization of relations between the different components of the organization, in this case between administrative agents and professional or clinical agents. We need to know more about the mechanisms that are conducive to generating or maintaining trust between these key agents of the organization. The effective mobilization of leaders from various professional groups may help to secure trust among people involved in a given change process. Moreover, the social and organizational dynamics surrounding the deployment of a collective leadership cannot produce change by themselves. If change projects are significant enough, they will have varying impacts on agents and groups of agents. Ideally, these impacts will maximize benefits for the greatest number of agents in an organization and in their environment. These benefits, already mentioned in the HTF analysis, can be in the realm of learning and autonomy in the completion of new models of health care organizations and practices. However, it is also clear that the benefits arising from the introduction and implementation of major changes will not be the same for all groups and agents (Miller, Greenwood, and Hinings 1999). Collective leadership in favour of change - as consolidated as it may be - must be able to rely on incentives that are consistent with the desired changes. Therefore, it appears that change management is a complex process within organizations, and it becomes more complex when it involves

104 Jean-Louis Denis

several organizations that are distinct yet required to cooperate more closely. The dimensions of a functional collective leadership and the mechanisms likely to foster the coexistence of multiple identities in one single organizational project will be more difficult to produce in interorganizational spaces (Denis, Lamothe, and Langley 2001b). The mobilization of leaders is central but must be promoted through effective incentive systems. Management tools such as quality strategies and performance indicators may also help to channel the behaviour of autonomous agents towards change (Berry 1983; Oakes, Townley, and Cooper 1998). However, so that these approaches may bring about change, they must be supported and guided by a strong organizational leadership. The dynamics of leadership in favour of change lead us to make the following proposals: Proposal 1: To produce transformation, it is important to achieve solid cooperation between agents in the clinical field and in the administrative field in steering change. Proposal 2: To produce transformation, it is important to develop a collective vision of organizational leadership. Proposal 3: To produce transformation, agents in a leadership position must share their roles effectively and accept changes to those roles over time. Proposal 4: To produce transformation, leaders must pay special attention to communication and to the validation of change projects with the organizational base. This validation includes an inventory of costs of change for the various agents and groups of agents. Proposal 5: To produce transformation, leaders must be in a position to strike a balance between the agents' preferences within an organization and outside pressures for change.

Conclusion In this study, we have analysed different basic processes for building capacities for change in health care systems and organizations. We quickly realized the interdependence of governance and change capacities. Change of a local, voluntarist, and emergent nature is not sufficient in the health care system. If this were the case, there would be no operational problems in the current health care system, nor any shortcomings in the renewal of practices. Throughout this study, we have worked on the assumption that a radical change should come about

Governance and Management of Change 105

progressively through the standardization of policies and of initiatives at the macroscopic level of government, as well as at the organizational and clinical levels in the health care system. Convergent change comes about through fine-tuning and does not significantly modify agents' perceptions of their areas of activity, current modes of operation and practice, and the degree of cooperation and coordination between the different components of the health care system. Radical change affects all of these elements so as to alter significantly practices, organizational forms, and the interpretation by agents of critical issues for the sustainable development of the system. The recent evolution of the health care system and organizations has converged towards a need for radical change. The problem associated with the integration of health care and services is a good indicator of the importance of implementing such changes. Moreover, recent experience with health care reforms demonstrates the need for wisdom in the speed of implementation of change. The health care system in Quebec shows that it is possible, in a short time, to make major structural changes and, in so doing, to destabilize the health care system and organizations. Such upheaval did not facilitate the implementation of radical changes to the fundamental operating logic of the system. It is through these observations that the idea of a progressive radical change appears to produce results. By favouring such an approach to change, the role of governance gains importance. To bring about transformation, governance must take its supporting, expert, and incentive role seriously in emerging change projects in the health care system. An analysis of regional governance shows potential for significant intervention, but one that is geared towards promoting experimentation. An analysis of the HTF experience shows its limitation to strictly local-change production. The incentive of professionals and organizations and the development of human resources through change training should have preferential status in governance practices. A leadership analysis shows the need to develop an organizational elite able to assume the control of change and adapt their roles to the progressive character of that change. In all of the processes we have analysed, the costs and benefits of change for human resources must be considered. In a professional context, these benefits can involve a key learning component. However, learning cannot take place if changes lead to excessive and seemingly disruptive upheavals in clinical work. This leads us to make the following recommendations. These rec-

106 Jean-Louis Denis ommendations translate the results of our analysis into courses of action for the federal government. They are based on the assumption that the federal government is maintaining, and will seek to maintain in the future, a role in the evolution of the health care system in Canada. They are also influenced by the notion that there must be diversity in a system so that it can renew itself. A support role for the federal government in the renewal of the health care system thus seems reasonable. The experience with the HTF program suggests that it is possible to promote experimentation; special attention should be paid, however, to the sustainability of initiatives that appear to be promising. This likely compels greater synergy between the various governance levels in the health care system. Recommendation 1: To help produce the transformation required in the health care system and organizations, the federal government must pay special attention to the need for consistency between the initiatives taken at the different levels of policy and administrative decision making. This recommendation is based on the importance of the interlocking logic of change production in complex systems. Recommendation 2: To help produce the transformation required in the health care system and organizations, the federal government needs to rely on human-capital development and invest resources in the training of agents for change. This recommendation is based on the critical role human capital plays in professional organizations, on the necessary autonomy of agents, and on the importance for agents to develop favourable perceptions of the implementation of change. Recommendation 3: To help produce the transformation required in the health care system and organizations, the federal government must invest in a decentralized risk-capital program to promote experimentation with new organizational models and practices. This recommendation is based on the importance of making resources available so that agents can commit to change and of the role of local-governance structures in revitalizing change. This risk-capital program must be integrated with other governance structures instead of directly addressing the local level. Recommendation 4: To help produce the transformation required in the health care system and organizations, the federal government needs to foster the emergence of inter-regional and inter-provincial spaces within which a collective leadership may be exercised in favour of change. This recommendation is based on the acknowledged limitations of local change and on the need to create new political spaces to resolve crucial

Governance and Management of Change 107 issues related to the funding of organizations and professional practices and to the availability of human resources. Recommendation 5: To help produce the transformation required in the health care system and organizations, the federal government needs to invest in a monitoring system to document, follow up, and share information on innovative experiments. This recommendation is based on the importance of information and knowledge in influencing agents in favour of change.

NOTE

I would like to express my sincere appreciation to the following people for their valuable comments on this report: Francois Champagne, Department of Health Administration and Interdisciplinary Health Research Group, Universite de Montreal; Andre-Pierre Contandriopoulos, Department of Health Administration and Interdisciplinary Health Research Group, Universite de Montreal; Paul Lamarche, Department of Health Administration and Interdisciplinary Health Research Group Universite de Montreal; Lise Lamothe, Department of Health Administration and Interdisciplinary Health Research Group, Universite de Montreal; Ann Langley, Ecole des Hautes Etudes Commerciales; Pascale Lehoux, Department of Health Administration and Interdisciplinary Health Research Group, Universite de Montreal; Raynald Pineault, Department of Social and Preventive Medicine and Interdisciplinary Health Research Group, Universite de Montreal; Daniele Roberge, Charles LeMoyne Hospital Research Centre, and Interdisciplinary Health Research Group, Universite de Montreal.

REFERENCES

Aggeri, F. and A. Hatchuel. 1999. Innovation et production de la norme: les nouvelles formes de gouvernementalite dans le domaine de I'environnement, in Les figures de I'institution et de la norme: entre etats et sodetes civiles. Proceedings of the ADSE conference. Paris: 1'Harmattan. Alter, Catherine, and Jerald Hage. 1993. Organizations working together. Newbury Park, Calif.: Sage Publications. Alvesson, M. 2001. Knowledge work: Ambiguity, image and identity. Human Relations 54(7): 863-886. Anderson, R. A. and R.R. McDaniel. 2000. Managing health care organizations:

108 Jean-Louis Denis Where professionalism meets complexity science. Health Care Management Review 25(1): 83-92. Beland, Francois, et al. 2001a. Evaluation d'un systeme de services integres pour personnes agees en perte d'autonomie (SIPA), in J.-C. Henrard et al., ed., Personnes agees dependantes en France et au Quebec. Qualite de vie, pratiques et politiques. Paris: INSERM, 237-58. Beland, Francois, et al. 2001b. Un systeme de services integres pour personnes agees en perte d'autonomie - Evaluation de la phase 1, juin 1999 a mai 2000. Final report, Health Transition Fund. Ottawa. Berry, Michel. 1983. line technologie invisible? L'impact des instruments de gestion sur revolution des systemes humains. Paris: Management Research Centre report, Ecole polytechnique. Bickerton, James. 1999. Reforming health care governance: The case of Nova Scotia. Journal of Canadian Studies/Revue d'etudes canadiennes 34(2): 159-91. Bohman, J. (1996). Public deliberation: pluralism, complexity, and democracy. Cambridge, Mass.: MIT Press. Bourdieu, Pierre. 1980. Le sens pratique. Paris: Les editions de Minuit. Bourdieu, Pierre, and L.J. Wacquant. 1992. Reponses. Pour une anthropologie reflexive. Paris: Editions du Seuil. Brock, David, Michael Powell, and C.R. (Bob) Hinings. 1999. The restructured professional organisation: corporates, cobwebs and cowboys, in David Brook, Michael Powell, and C.R. (Bob) Hinings, ed., Restructuring the professional organization: Accounting, health care and law. London: Routledge, 215-29. Brunsson, N. and Andersson K. Sahlin. 2000. Constructing organizations: The example of public sector reform. Organization Studies 21(4): 721-^6. Bryson, John M., and Barbara C. Crosby. 1992. Leadership for the common good: Tackling public problems in a shared-power world. San Francisco: Jossey-Bass Publishers. Gallon, Michel, Pierre Lascoumes, and Yannick Barthe. 2001. Agir dans un monde incertain. Essai sur la democratic technique. Paris: Editions Seuil. Campbell, D.T. 1969. Reforms as experiments. American Psychologist 24:409-29. Casebeer, Ann, and Kathryn Hannah. 2000. Transforming a health care system: Managing change for community gain. Canadian Journal of Public Health 91(2): 89-93. Chevalier, J. 1996. La gouvernabilite. Paris: P.U.F. Church, John, and Paul Barker. 1998. Regionalization of health services in Canada: A critical perspective. International Journal of Health Services 28(3): 467-86. Commission d'etude sur les services de sante et les services sociaux. 2000. Emerging solutions. Quebec.

Governance and Management of Change 109 Contandriopoulos, Andre-Pierre. 1994. Reformer le systeme de sante: une utopie pour sortir d'un statu quo impossible. Ruptures 1(1): 8-26. Contandriopoulos, Andre-Pierre, Jean-Louis Denis, Nassera Touati, et al. 2001. Integration des soins: dimensions et mise en oeuvre. Ruptures 8(2): 38-52. Contandriopoulos, Andre-Pierre, Jean-Louis Denis, Paul Lamarche, et al. 2002. Les soins et services de premiere ligne dans un systeme de sante public. Final report prepared and submitted to the Confederation des syndicats nationaux (CSN). Contandriopoulos, Andre-Pierre, Marc-Andre Fournier, Jean-Louis Denis, et al. 1999. Les mesures incitatives et le paiement des ressources, in Clermont Begin et al., ed., Le systeme de sante quebecois. Un modele en transformation. Montreal: Presses de 1'Universite de Montreal, 123^47. Contandriopoulos, Andre-Pierre, and Yves Souteyrand, dir. 1996. L'hopital stratege: dynamiques locales et offre de soins. Paris: John Libbey Eurotexte. Contandriopoulos, Damien, Jean-Louis Denis, and Ann Langley. 2001. The definition of the public in health care reform. Workshop in organizational discourse: Theory and analysis. Academy of Management Annual Meeting, Washington, D.C. August. Davidson, Alan R. 1999. British Columbia's health reform: 'New Directions' and accountability. Canadian Journal of Public Health 90 (Suppl. 1): S35-S38. Denis, Jean-Louis. 2001. Comment integrer le changement dans le systeme de sante: les legons tirees des 40 projets quebecois finances par le Fonds d'adaptation des services de sante (PASS). Guest speaker at the XXIIth Jean-YvesRivard conference: L'innovation et le changement. Les innovations produisent-elles toujours du changement? Montreal, 8-9 November. Denis, Jean-Louis. 2001. Entre le central et le local - les dynamiques de regionalisation en sante. Un nouveau paradigme pour la sante publique. 1st international conference on local and regional health programs. Amiens, France, 23-26 October. Denis, Jean-Louis. (1999). Trois modeles et trois terrains pour penser la decentralisation, in J.P. Claveranne et al., ed., La sante demain: vers un systeme de soins sans murs. Paris: Economica 49-66. Denis, Jean-Louis, et al. 2001. Histoire et bilan d'une innovation institutionnelle: les Agences Regionales de I'Hospitalisation. INSERM/MIRE/CNRS convention, July. Denis, Jean-Louis, et al. 1998. Les modeles theoriques et empiriques de regionalisation du systeme sociosanitaire, rapport de recherche CQRS. Denis, Jean-Louis et al. 2000. The regionalization of the Quebec health care system - an analysis. ASA-IFSAM symposium 2000, Montreal, 8-11 July.

110 Jean-Louis Denis Denis, Jean-Louis, and Annick Valette. 2000. Changement de structure de regulation et performance du mandat. Finance, controle et strategic 3(2): 57-79. Denis, Jean-Louis, and Annick Valette. 1997. La regulation regionale au concret: 1'experience des DRASS. Politiques et Management Public 15(4). Denis, Jean-Louis, and Ann Langley. 2001. Power, legitimacy and shifting archetypes in health care reform. ASAC 2001: A management odyssey. 29th Annual Conference, 26-29 May, London, Ont. Denis, Jean-Louis, Ann Langley, and Andre-Pierre Contandriopoulos. 1998. From institutional discourse to organizational action: The dynamics of legitimation. Academy of Management Meeting, San Diego, August 1998. Denis, Jean-Louis, Ann Langley, and Andre-Pierre Contandriopoulos. 1995. Le role des instances regionales dans la transformation du systeme de sante au Quebec - le cas de la Regie regionale de Montreal-Centre. Revue Frangaise d'Administration Publique 76:599-608. Denis, Jean-Louis, Ann Langley, and Linda Cazale. 1996. Leadership and strategic change under ambiguity. Organization Studies 17(4): 673-99. Denis, Jean-Louis, Ann Langley, and Marc Pineault. 2000. Becoming a leader in a complex organization. Journal of Management Studies 37( 8): 1063-99. Denis, Jean-Louis, Damien Contandriopoulos, Ann Langley, et al. 1999. Theorie et pratique de la regulation des regies regionales de la sante au Quebec. Report funded by HEALNet. Denis, Jean-Louis, and Lise Lamothe. 2001. Le rapport de la Commission Clair: renovation ou transformation du systeme de sante quebecois? in Quebec 2002: annuaire politique, social, economique et culturel, under the direction of R. Cote et al. Quebec: Editions Fides. Denis, Jean-Louis, Lise Lamothe, and Ann Langley. 2000. The emerging shape of professional organization in health care. Symposium on professionals as knowledge assets in organizations - reality, time and power. Academy of Management Annual Meeting, Toronto, August. Denis, Jean-Louis, Lise Lamothe, and Ann Langley. 2001. The dynamic of collective leadership and strategic change in pluralistic organizations. Academy of Management Journal 44(4): 809-37. Denis, Jean-Louis, Lise Lamothe, and Ann Langley. 2001a. Governmentfunded experiments as resources for renewal in health care? All Academy Symposium: How governments matter: the unanticipated consequences of health reform. Academy of Management Annual Meeting, Washington, D.C. August. Denis, Jean-Louis, Lise Lamothe, and Ann Langley. 1999. The struggle to implement teaching hospital mergers. Canadian Public Administration 42(3): 285-311.

Governance and Management of Change

111

Denis, Jean-Louis, Lise Lamothe, Ann Langley, et al. 1999. The struggle to redefine boundaries in health care systems, in David Brock, Michael Powell, and C.R. (Bob) Minings, ed., Restructuring the Professional Organizations. London: Routledge, 105-30. Desbiens, Francine, Christian Dagenais, Pierre Joubert. et al. 2001. De I'innovation au changement. Final report submitted to the Health Transition Fund (HTF), under the direction of Pierre Joubert jointly with members of the steering committee on evaluation. Quebec City: Ministere de la sante et des Services sociaux. Drache, D., and T. Sullivan T. 1999. Market limits in health reform: Public success, private failure. New York: Routledge. Eisenhardt, K.M. 1985. Control: Organizational and economics approaches. Management Science 31(2): 134-49. Empson, L. 2001a. Introduction: Knowledge management in professional service firms. Human Relations 54(7): 811-17. - 2001b. Fear of exploitation and fear of contamination: Impediments to knowledge transfer in mergers between professional services firms. Human Relations 54(7): 839-62. Evans, Robert, G., Morris L. Barer, and Theodore R. Marmor, ed. 1994. Why are some people healthy and others not? The determinants of health of populations. New York: Aldine de Gruyter. Ferlie, E.L., et al. 1996. The new public management in action. Oxford: Oxford University Press. Ferlie, Evan, and Louise Fitzgerald. 2002. Le changement dirige par le haut au Royaume-Oui: un poiny de vue polique. Gestion 27(3): 125-33. Finkelstein, S., and D. Hambrick. 1996. Strategic leadership. St-Paul, Ninn.: Southwestern College Publishers. Forest, Pierre-Gerlier, et al. 2000. Participation de la population et decision dans le systeme de sante et de services sociaux du Quebec. Department of Political Sciences and the Research Group on Public Policy (GRIG). Report submitted to the Conseil de la Sante et du Bien-etre du Quebec. Giacomini, M., J. Hurley, V. Bathia et al. 1996. The many meanings of money: A health policy analysis framework of understanding financial incentives. Centre for Health Economics and Policy Analysis, Hamilton, Ont, McMaster University. Giddens, A. 1984. The constitution of society. Berkeley: University of California Press. Greenwood, Royston, and C.R. (Bob) Hinings. 1996. Understanding radical organizational change: Bringing together the old and the new institutionalism. Academy of Management Review 21(4): 1022-1054.

112 Jean-Louis Denis Hambrick, D., and P.A. Mason. 1984. Upper echelons: The organization as a reflection of its top managers. Academy of Management Review 9:193-206. Hatchuel, A. 2000. Perspective et gouvernance: quelle theorie de 1'action collective? in E. Heurgon and J. Landrieu, edv Perspective pour une gouvernance democratique. Paris: Editions Aube. Hayes, M.V., and J.R. Dunn. 1998. Population health in Canada: A systematic review. CPRN Study no.HOl. Ottawa Canadian Policy Research Networks. Health Services Restructuring Commission. 2000. A legacy. Toronto: Ministry of Health. Hinings, C.R. (Bob), and Royston Greenwood. 1988. The dynamics of strategic changes. New York: Basil Blackwell. Hood, Christopher C. 1983. The tools of government. London: Macmillan. Hughes Tuohy, Caroline. 1999. Dynamics of a changing health sphere: The United States, Britain and Canada. Health Affairs 18(3): 114-34. Kitchener, Martin. 1999. 'All fur coat and no knickers:' Contemporary organizational change in United Kingdom hospitals, in David Brook, Michael Powell, and C.R. (Bob) Hinings, ed., Restructuring the professional organization: Accounting, health care and law. London: Routledge, 183-99. Kotter, J.P. 1995. Leading change: Why transformation efforts fail. Harvard Business Review 73(2): 59-67. Lamarche, Paul, et al. 2001a. L'integration des services: enjeux structurels et organisationnels ou humains et cliniques. Ruptures 8(2): 71-92. - 2001b. Projet d'action concertee sur les territoires (PACTE): une gestion locale des services du reseau de la sante et des services sociaux en reponse aux besoins des personnes agees. Research report QC-423, Health Transition Fund (HTF), Health Canada. Lamothe, Lise. 1996. La structure professionnelle clinique de facto d'un hopital de soins ultraspecialises. PhD thesis, McGill University, Montreal. Lascoumes, Pierre. 1996. Rendre gouvernable: de la 'traduction' au 'transcodage.' L'analyse des processus de changement dans les reseaux d'action publique, in La gouvernabilite. Paris: P.U.F., 325-38. Leatt, Peggy, George Pink, and Michael Guerriere. 2000. Towards a Canadian model of integrated health care. HealthCare Papers 1(2): 13-35. Lemieux, V. 1997. La decentralisation. Quebec: Les presses de 1'Universite Laval. Lewis, Steven J., et al. 2001. Devolution to democratic health authorities in Saskatchewan: An interim report. CMAJ, 164(3): 343-47. Lomas, Jonathan. 1997. Devolving authority for health care in Canada's provinces: 4. Emerging issues and prospects. CMAJ 156(6): 817-23. - 2001. Past concerns and future roles for regional. CMAJ, 164(3). 356. Lomas, Jonathan, John Woods, and Gerry Veenstra. 1997a. Devolving author-

Governance and Management of Change 113 ity for health care in Canada's provinces: 1. An introduction to the issues. CMAJ 156(3): 371-7. Lomas, Jonathan, Gerry Veenstra, and John Woods. 1997b. Devolving authority for health care in Canada's provinces: 2. Backgrounds, resources and activities of board members. CMAJ 156(3): 513-20. Lomas, Jonathan, Gerry Veenstra, and John Woods. 1997c. Devolving authority for health care in Canada's provinces: 3. Motivation, attitudes and approaches of board members. CMAJ 156(5): 669-76. L0wendahl, B.R., O. Revang, and S.M. Fosstenlokken. 2001. Knowledge and value creation in professional service firms: A framework for analysis. Human Relations 54(7): 911-32. Mhatre, S., and R. Deber. (1992). From equal health care access to equitable access to health: A review of Canadian provincial health commissions and reports. International Journal of Health Services (22): 645-68. Meyer, Marshall W., and Lynne G. Zucker. 1989. Permanently failing organizations. Newbury Park, Calif.: Sage Publications. Miller, D., and P.H. Friesen. 1984. Organizations: A quantum view. Englewood Cliffs, N.J.: Prentice Hall. Miller, Danny, Royston Greenwood, and C.R. (Bob) Hinings. 1999. Creative chaos versus munificient momentum: The schism between normative and academic views of organizational change. Gestion 24(3): 158-64. Mintzberg, Henry, Bruce Ahlstrand, and Joseph Lampel. 1999. Transforming the firm. Gestion 24(3): 122-30. Mintzberg, Henry, Bruce Ahlstrand, and Joseph Lampel. 1998. Strategy safari: A guided tour through the wilds of strategic management. New York: Free Press. Oakes, Leslie S., Barbara Townley, and David J. Cooper. 1998. Business planning as pedagogy: Language and control in a changing institutional field. Administrative Science Quarterly 43(2): 257-92. Perrow, Charles, and Mauro F. Guillen. 1990. The AIDS disaster: The failure of organizations in New York and the nation. New Haven: Yale University Press. Pettigrew, Andrew, E. Ferlie and L. McKee. 1992. Shaping strategic change: Making change in large organizations: The case of the National Health Services. London: Sage Publications. Powell, Michael J., David M. Brock, and C.R. (Bob) Hinings. (1999). The changing professional organization, in David Brook, Michael Powell, and C.R. (Bob) Hinings, ed., Restructuring the professional organization: Accounting, health care and law. London: Routledge, 1-19. Rocher, Guy. 2001. Le 'laboratoire' des reformes dans la Revolution tranquille. Desjardins conference, given as part of McGill University's Quebec Studies Program, 6 November. Montreal: McGill University.

114 Jean-Louis Denis Rondinelli, D.A. 1981. Government decentralization in comparative perspective. International Review of Administrative Sciences 47(2): 135. Ruggie, Mary. 1999. The US, UK and Canada. Convergence or divergent reform practices? in Daniel Drache and Terry Sullivan, ed., Market limits in health reform: Public success, private failure. London: Routledge, 127-44. Saltman, Richard, and J. Figueras, directors and authors. 1997. European health care reform: Analysis of current strategies. Copenhagen: WHO Regional Publications, European Series, no.72. Saltman, Richard, J. Figueras, and C. Sakellarides, dir. 1998. Critical challenges for health care reform in Europe. Buckingham: Open University Press. Saltman, Richard B., and O. Ferroussier-Davis. 2000. The concept of stewardship in health policy. Bulletin of the World Health Organizations 87(6): 732-9. Scheirer, M.A. 1981. Program implementation: The organizational context. Beverly Hills, Calif.: Sage Publications. Scheirer, M.A. 1987. Program theory and implementation theory: Implications for evaluators, in L. Bickman, ed., Using program theory in evaluation. San Francisco: Jossey-Bass, 59-76. Smith, B.C. 1979. The measurement of decentralization. International Review of Administrative Sciences 45(3): 215. Touati, Nassera, et al. 2001. Une experience d'integration des soins dans une zone rurale: les enjeux de la mise en ceuvre. Ruptures 8(2): 93-107. Trottier, Louise-Helene, et al. 1999. Contrasting vision of decentralization, in Daniel Drache and Terry Sullivan, ed., Market limits in health reform: Public success, private failure. New York: Routledge, 147-65. Turgeon, Jean. 2001. Panorama quebecois: passe et prospectives. Guest speaker at the XXIIth Jean-Yves-Rivard Conference: L'innovation et le changement. Les innovations produisent-elles toujours du changement? Montreal, 8-9 November. Tushman, M.L., and E. Romanelli. 1985. Organizational evolution: A metamorphosis model of convergence and reorientation, in L.L. Cummings and B.M. Staw, ed., Research in organizational behavior. Greenwich, Conn.: JAI Press, vol. 7,171-22. Valette, Annick, Andre-Pierre Contandriopoulos, and Jean-Louis Denis. 2000. Les processus de restructuration hospitaliere et leur gestion. Report. France: MIRE.

PART TWO CHANGE AND PROVIDERS

This page intentionally left blank

4 Planning for Care: Approaches to Health Human Resource Policy and Planning in Health Care PAT ARMSTRONG AND HUGH ARMSTRONG

Shortly after the Commission on the Future of Health Care in Canada asked us to address the question 'How do health human resource practices and policies promote or inhibit change?' the Canadian Institute for Health Information (CIHI) released its report Canada's Health Care Providers (CIHI 2002). The Canadian Policy Research Networks have also just completed a substantial report summarizing the literature and issues related to human resource planning for physicians and registered nurses (RNs) (Koehoorn et al. 2002). Instead of repeating what is in these reports, we take as given the wealth of information they provide and set out to complement them. Our emphasis is on identifying principles and approaches to planning for change, rather than on providing details or recommending particular solutions. Many of the latter are available to the commission and to the general public. This paper suggests ways for assessing them. Addressing the commission's question means, in the first instance, setting out what kinds of change are at issue. This, in turn, means identifying both the kinds of change now under way and the options available for future change. The choices made about change necessarily imply choices about the nature of care. Only with a model for care in view can one assess whether human resource planning initiatives in health care are or will be successful. Hence, we devote our attention in the next section to a discussion of two distinct models for care. The extent to which one or the other model is adopted will shape the criteria by which the success of planning initiatives may be measured, and more specific recommendations assessed.

118 Pat Armstrong and Hugh Armstrong

This is not to suggest that the criteria for successful planning can be simple or straightforward. Even with agreement on a preferred model for care, and thus on the desirable directions for change, human resource planning in health care is inevitably somewhat messy and incomplete. The contexts and dynamics of change, the tensions among criteria that are all compatible with the preferred model, and the very complexity of the vast field of activity known as health care all serve to complicate matters. It takes a wide diversity of providers to deliver health care. The rise or fall of any one group of providers, as well as changes in the relationships among these groups, will have profound effects on the kinds of care available. So, too, will the education of providers, both because it influences the skills and perspectives they bring to their work and because it is a vital factor in determining how many providers are available. Another important ingredient in human resource planning is the nature of decision making, regulation, and control in health care. This element has a significant impact on the quality and kinds of care and on the ways in which providers respond to change. Subsequent sections of the paper consider the current human resources for care and the relationships among groups of providers, education for care, and accountability and governance for care. The range of these influences on human resource planning in health care, as well as the tensions within and among them, all contribute to the inevitable messiness and incompleteness of this planning. Before ending this section, however, we wish to comment briefly on the kinds of change now under way. For at least a century, the history of health care in Canada has been one of constant evolution but, during the last decade, changes in health care have been more dramatic. These changes take four obvious forms. The first is the restructuring of hospital care to focus on acute, short-term, technology-dependent interventions. This development is in turn linked to a second, the significantly altered care provided in long-term facilities. People in long-term care now have complicated medical needs that are often combined with mental-health problems. With hospital and long-term facilities now focused on sophisticated medical care, more and more complicated care is provided at home. This third aspect of change has, like the others, been made possible by developments in technologies and drugs, as well as by new ideas about management, responsibilities, costs, evidence, and the nature of care. A fourth kind of change is now receiving increasing attention - primary care and telehealth. All these changes are taking place within significantly changed international and national contexts,

Planning for Care 119

about which more will be said later. These contexts shape many of the options available and the consequences of selecting among them. In our view, many proposals for human resource planning in health care fail to take these contexts sufficiently into account. This is not to suggest that current reforms and contexts tightly determine the options for future change. There are choices to be made in how care is best understood and pursued. We now turn to a discussion of contrasting approaches to the nature of care and to some of their implications for human resource planning. The Nature of Care One approach to human resource planning in this field begins by recognizing the specificity of health care. Within this perspective, health care is not a business like the rest, and people working in health care differ in some significant ways from those employed in other sectors. Many of the differences may seem obvious - merely common sense but they are important to identify because they provide the backdrop for any consideration of change and resource planning. Perhaps most obviously, health care is about human life. This means that the risks, and consequences, cannot be assessed exclusively or even primarily in economic terms. Health care is also about individuals, each with his/her own specific complex of health issues and each shaped by particular cultural, social, and economic contexts. Although it is clearly important to search for patterns in population health and for evidence about the efficacy of particular interventions, it is equally important to understand that such patterns do not necessarily play out in the case of a particular individual in a particular social context. Indeed, evidence in health care is primarily about what works relatively well, most of the time, for a significant proportion of people. Decisions about actual care, however, are necessarily made about individuals, who may be among those who do not fit the pattern. Health care needs and the methods of addressing them can never be precisely, scientifically established. As a result, the notion of being able to determine accurately what is the right thing to be done by the right person to the right person at the right time in the right place is simply inappropriate in health care. It requires a skilled practitioner to connect evidence to the specific case and some trust that this will be the case. 'Whatever the technologies, medicine depends on the quality and credibility of interpersonal relationships between clinicians and patients and the organizational forms that sup-

120 Pat Armstrong and Hugh Armstrong

port them' (Mechanic 2002,466). The same applies to nurses and others who provide most of the daily care. Moreover, care is an interactive process, with both the care provider and care recipient engaged in ways that necessarily have an impact on what can and will be done, as well as on the effect of what is done. The best science cannot predict exactly how individuals will react to treatment, because care is about a relationship among persons, each of whom brings his/her own attitudes, experiences, and situation into the exchange. Nor is it a simple matter to determine what an individual provider will do in a particular case. The health of both patients and providers is determined by an array of factors, including their psychological, social, physiological, economic, and physical situations, and so are their responses. And adverse events, including the process of dying, create particular needs for support and care in both patients and providers. These specific characteristics of care have consequences for care work. They mean that providers must have knowledge of whole people and their particular contexts, as well as knowledge of the complex and different ways individuals respond to care. This requires a range of complementary skills that are either integrated within individual providers or possessed by different individuals working together to provide care. These specific characteristics mean as well that many of those providing care need have not only a range of skills but also both the autonomy required to make judgments about the particular needs of the person requiring care and the possibility for collaboration or consultation on care. Evidence provides a guide rather than the rules for care. It is combined with experience and skill to create the expert (Benner 1984). Autonomy, in turn, requires that providers take responsibility and be trusted to take responsible action. In addition, the nature of care means that care delivery involves a wide range of skills and locations. Providers typically view themselves as working in health care whether they perform the surgery or clean up after the surgery, whether they administer regional health authorities or book patient appointments in clinics, whether they are paid well to provide care in a hospital or paid nothing at all to provide care at home. And the research findings on the determinants of health support their view. It would be wrong, however, to suggest that this understanding of health care is universally shared. Indeed, many reforms over the last decade have been based explicitly or implicitly on the assumption that health care is a business like any other and that employees within the

Planning for Care 121

system can be managed like any others. Such reforms have been adopted without evidence that they are appropriate for care providers and sometimes even without evidence that they have worked well in other sectors. The amalgamation of hospitals mimics similar amalgamations in the corporate sector. Across Canada, more than thirty major teaching hospitals have recently been merged into giant organizations (Levine 2000, iii) and many small community hospitals have been closed or transformed. Markham and Lomas (1995) argue that there is no empirical evidence to demonstrate economic efficiency, quality, or human-resource gains with multi-hospitals, and some evidence suggests that costs may increase, flexibility and responsiveness to individual patients' needs decline, and relationships with employees deteriorate. Similarly, downsizing in health care services imitates a strategy used in other sectors, even though research on downsizing corporations indicates that a majority of those initiatives did not increase productivity, a significant number raised expenditures, and both morale and trust usually declined (Appelbaum, Everard, and Hung 1999). Moreover, 'efficiency' for one organization often turns out to be simply the off-loading of costs onto another organization, or to individuals and households where women typically have to pick up the burden of extra unpaid work. In the words of Janet Gross Stein (2001, 71), 'efficiency is an intensely political concept.' 'Between 1994 and 1996, 85 per cent of Canadian hospitals reduced their workforce by more than 10 per cent' (Wagar and Rondeau 2000, iv). The largest decrease was in management and executive positions, but 14 per cent of maintenance staff and 12 per cent of nursing staff also lost their jobs. Some of this reduction reflected new technologies that made both day surgery and complex home care possible, as well as a philosophical shift that defined care at home not only as better care but also a matter of public preference. Not surprisingly, with job reductions have come lower employee satisfaction and more conflict for those who remain (Wagar and Rondeau 2000, iv). New jobs in the community have not made up for the job losses in the institutional sector (Kazanjian 2000, 6), and little has been done in long-term care facilities to accommodate the transformed care needs. Meanwhile, the shift to home and long-term facility care has often meant a literal shift to care as a business, given that a growing number of these services are provided on a for-profit basis, as part of large corporations. Human resource strategies have also been imported from outside

122 Pat Armstrong and Hugh Armstrong

the health care sector, and these, too, have contributed to dissatisfaction and conflict. 'Total quality' management practices have been among the most popular, in part because their emphases on participation and teamwork, on improving processes and customer satisfaction, and on 'multi-skilling' and innovation all seem to fit well with health care (Hassen 1993). However, in a context of cost-cutting, downsizing, and treating health care as a business like any other, these promises have not been fulfilled (Armstrong et al. 1997; Lam and Reshef 1999). Instead of teamwork, there is often a fragmentation of care work into a series of tasks performed by the lowest-skilled and lowest-cost care provider, or by quickly trained unpaid providers, most of whom are women. More managerial control, exercised especially through measurement techniques and care pathways,1 and the emphasis on the elimination of waste defined as extra beds or time not used in the performance of medical tasks have been central features of the reforms. This assumption of control often conflicts with the traditional exercise of autonomous clinical judgment by health professionals. At the same time, flattened hierarchies tend to concentrate control at the top while delegating responsibility, with fewer resources, to those delivering care. One place where the conflict becomes obvious is in the use of information technologies. Computer-based patient-records technology, for example, which was 'initially sold as facilitating nursing work, improving coordination between nursing and medical activities, improving quality of care, and lowering costs, resulted in information overload and standardization, clerical tasks load increases, work organization rigidity, and expert autonomy negation' (Sicotte, Denis, and Lehoux 1998, 440-1). The main problem was identified as the failure to base technologies on the way nursing is practised - in other words, the failure to recognize the specific nature of care work. Task-measurement strategies borrowed from the private sector face similar problems, because of the assumption that a specific task will take the same time without regard to the particular patient or provider. The consequences of these strategies are obvious in the high rates of illness and injury among nurses and assisting occupations. Health sector workers are over 50 per cent more likely than other workers to miss work because of illness or injury. If the absenteeism rate of RNs were reduced to that of all other workers, the equivalent of almost 5,500 more nurses would be at work full-time each year (CIHI 2002, 87) almost exactly the number of RN diploma and baccalaureate graduates

Planning for Care 123

in 1999 (CIHI 1999, 31). There are high burnout rates, feelings of job insecurity especially among less experienced nurses, and work-family conflicts' (CIHI 2002, 87). Part-time and casual employment is much more common than in other sectors, and so is multiple job holding. These patterns, too, can be linked to managerial strategies that fail to take the specificity of health care into account. Noting that current management practices have produced growing job dissatisfaction among nurses, the Final Report of the Canadian Nursing Advisory Committee (CNAC 2002) is blunt on the urgent need to address issues such as workload, overtime, absenteeism, illness and injury, turnover, paperwork burden, front-line management by nurses, and the elimination of chief nurse positions. It concludes that the need is 'not to repair nursing, but rather to renew and repair the work environments in which nurses practice' (CNAC 2002, 25). In other words, it is about business practices that fail to recognize the specific characteristics of care work. The same holds for the working environments of other health care providers. Of course, Canada is not alone in these developments. Management strategies and the accompanying technologies are shared globally, and so increasingly is management personnel. More for-profit involvement in Canada has also meant more foreign corporations involved in care. Just as there is evidence of an international move towards defining care as a business like the rest, so, too, is there evidence of widespread problems in workplace design (Baumann and O'Brien-Pallas 2001). A study of nurses in five countries concluded that, in emulating industrial models, hospital management has created problems in work design that are contributing to an uneven quality of care, medical errors, and adverse patient outcomes, as well as increased nurse dissatisfaction and high levels of stated intentions to leave the profession (Aiken et al. 2001; see also Burke and Greenglass 2000). It is not clear how far international trade agreements will force Canada in this direction. What is clear, however, is that they set limits on how much we can experiment, and that failed experiments would be costly in terms of both actual expenditures and loss of control over care policy (Gold 2002; Sanger 2001; Pollock and Price 1999). It is also clear that, for those negotiating and enforcing international trade agreements, health care is a business and one that is growing rapidly. This international context must be taken into account when developing strategies for health care reform, and thus for human resources in health care. The more or less enthusiastic embrace by governments, here and abroad, of the care-as-another-business model does promote the in-

124 Pat Armstrong and Hugh Armstrong

creased participation by for-profit firms in health care funding and delivery.2 However, a distinction must be made between comparing the models and describing the public and private sectors. In this section, we compare models rather than describe sectors. Although our preference for the care-as-different model will be obvious, the main point of the section is to argue that the choice between models has implications for the kind of workforce that is required. The care-as-different model stresses skill acquisition, continuous learning through practice, clinical autonomy, accountability through judgments based on evidence as well as peer or (less often) citizen review, and collaboration through teams with complementary skills. The care-as-another-business model emphasizes a division of labour based on quickly learned tasks, accountability through evidence-determined practices and managerial control, and substitution of lowerskilled for higher-skilled providers, as well as flexibility in assigning providers to tasks. The two overlap, of course. For example, both encourage more care by nurse practitioners and midwives, although their reasons differ. In the former model, these providers are supported because they focus on the whole person and spend more time on health promotion, while in the latter model they are supported because they are thought to be less expensive substitutes for physicians who would otherwise provide the care. The former entails an expansion of skills; the latter, a denigration of skills. Both also encourage evidence-based decision making, although the former sees evidence as a means of giving providers more control and enhancing their care, while the latter sees evidence as a means of controlling providers and making sure the least expensive care is provided. Each broad model also encompasses competing perspectives about who should do what. Many physicians, for example, understand primary care in terms of a medical model, with physicians assisted by others, while RNs tend to favour a more collaborative approach with nurses playing expanded roles. These necessarily brief and simplified descriptions are intended to demonstrate the importance of models and contexts in structuring care. Before planning for change begins, the models need clarification, and decisions must be made about which model prevails and to what extent it prevails. Such decisions are in turn linked to those about the nature of the care to be provided, where, and within what kind of structure. The introduction of a public home care or palliative care program across the country, for example, would mean that more resources are required, and the model guiding the program would determine how much of

Planning for Care 125

this care is provided by RNs, licensed practical nurses (LPNs), or other paid and unpaid providers, educated for the care or not. The Human Resources for Care The labour force in the health sector is highly organized, with 62 per cent belonging to a union and an additional number represented by professional organizations (Akyeampong 2001, 52). This workforce is highly structured along complex, hierarchical lines, although the structures have been flattened to eliminate many middle-management positions in recent years and some teamwork has always been part of care delivery. Doctors still define what is medically necessary but their authority is increasingly challenged by managers, patients, and other providers. More than thirty occupations and professions are regulated under various pieces of provincial or federal legislation (Koehoorn et al. 2002,11), but they account for a minority of the workforce employed in the sector and there are no regulations covering unpaid care. Moreover, there is considerable diversity among jurisdictions in terms of rights for individual providers, unions, and employers. As well, there is considerable diversity in rules and practices linked to different work locations. Equally important, there are often significant differences between what people actually do and what the regulations or policy documents say they do. Home care providers and hospital employees, for example, operate under different regimes, as do doctors and nurses working in the same hospital. Nurses in the north of the country have a wider range of duties than those working in the south, and care that in a facility can be provided only by a regulated provider is often provided at home by someone with little or no formal training. These structures mean that the organizations representing providers must be involved in change, and that there is a wide range of organizations and regulations to take into account. While these organizations press for their members' interests, it does not necessarily mean that they resist change or fail to take patient needs into account. Research on a major hospital strike, for example, demonstrates that the predominantly female employees were motivated primarily by concerns for their patients (White 1990), and research on physicians indicates that they were much more open to change before recent reforms than after (Woodward et al., 2000, 14). It is, then, the kind of change that is at issue, along with the speed of change and the involvement of providers in decision making about change.

126 Pat Armstrong and Hugh Armstrong

The health sector encompasses a wide range of jobs. Twenty per cent of paid employees in this sector are nurses. Along with doctors, who remain the most powerful group in care work, nurses have received the most policy and research attention. Some of the others are dentists, social workers, therapists, and technologists, but a significant proportion is accounted for by housekeepers, laundry and clerical workers, trades people, and dietary aides. The Canada Health Act recognizes that everyone employed in a hospital is necessary to care. More recently, however, support services in hospitals have been equated with hotel services and contracted out, explicitly rejecting notions of their specificity. Yet research shows that 'the technical sophistication of hospitals and the responsibilities of support workers require a different set of skills and training than would be required in a hotel/ especially as care becomes more complex (Cohen 2001, i). Moreover, the work is integral to that done by recognized professionals and makes it possible for them to do their jobs. The farther care moves from the hospital, the less it is provided by regulated professionals and the more likely it is part of a business. However, research in British Columbia indicates that support work is also critical to the health of those cared for at home (Hollander and Tessaro 2001). Human resource planning in health care, then, must take support workers into account not only in terms of numbers but also in terms of training and regulation. This is particularly the case as care moves away from hospitals and away from regulated professions. The labour force in the health sector is also highly gendered. More than 1.5 million people work in the paid health and social services labour force, and more than 80 per cent of them are women (CIHI 2002, 40). Women account for a similar proportion of unpaid primary care providers, although their actual numbers are more difficult to determine. We do know that those providing care as family, friends, and volunteers significantly outnumber paid providers, and that paid services complement unpaid ones, not the other way around (Armstrong and Kits 2001). Although women have moved in increasing numbers not only into the labour force but also into traditionally male-dominated health care jobs, and some men have moved into female-dominated work, nursing and support occupations remain women's work and so does care at home. Indeed, the National Forum on Health (1997, 19) used the term 'conscripted' to label the requirement that women provide unpaid home care to household members. This means that many women providing paid care are increasingly expected to provide unpaid care at home as well.

Planning for Care 127

Even without a disabled child, frail parent, or ill spouse at home, women are more likely than men to experience conflicts between family and paid work. Given that most men still earn more than most women, women are also more likely to be tied to their spouse's place of work. However, with their partners employed and with changing values about parenting, men, too, are less willing to work hours that make their home life difficult. Moreover, women are socialized to feel responsible for care and struggle hard to make up for the growing gaps in the system in ways that both make the consequences of cutbacks less visible and undermine their health. Equally important, they are expected to provide care. Delivery systems are based on this expectation. As providers of most of the hands-on care, women are held responsible for care by their employers, patients, and professional organizations. Similarly, those providing care at home, without pay or training, are also held responsible for their family members' care. The female domination of this work has contributed to the invisibility of many skills involved in care, as well as to the invisibility of the paid and unpaid care work now being done by women to make up for the gaps left by health care reforms (Morris 2001). These patterns have significant implications for pay, workforce distribution, and care location, as well as for openness to change. Policies to promote change need to take both paid and unpaid care, as well as the gendered nature of the workforce, into account (Armstrong, et al. 2001). The labour force in the health sector is also aging. As a result of women staying in the labour force and of reductions in both hiring and education, the average age of those working in health care is well over forty for the first time. This means that many are finding it increasingly difficult to work under current conditions. And many are near retirement or contemplating early retirement. Equally important, intensified working conditions limit their possibilities for sharing what they know with new recruits on a daily basis. This means not only that planning must contemplate replacements for the large numbers soon to retire, but also that strategies are required now to ensure that those currently employed stay until retirement age and share what they know during that time (Schetagne 2001, 20; Kazanjian 2000, 6; CNAC 2002, 37). It means as well that there are important differences within each occupational group that need to be considered in planning. Moreover, cutbacks throughout the last decade mean that there is an age gap in the labour force, with few people between the oldest and youngest groups. In addition, resources include those who are currently unemployed or underemployed in care. There are people who have already left

128 Pat Armstrong and Hugh Armstrong

because they were laid off or found working conditions too difficult, and others who have seen their hours reduced to part-time or have been shifted to casual work. There are also those who are employed but denied the opportunity to use the knowledge and skills they possess. They constitute a reserve that could be drawn back into care if conditions are changed to offer secure, full-time employment as well as enough support and time to provide appropriate care. Foreign-trained providers represent another reserve, one that we are increasingly tapping. However, this means that other countries are paying for the educated labour force we recruit and may be left without the resources they require. Furthermore, there are additional questions about the culture of care learned in other countries and the regulations regarding professional practice in Canada. Finally, the nature of the labour force is linked to the nature of the required care. Care needs are changing. The most talked-about change is the aging of the population. There are intensive debates about its impact. Some warn of catastrophic consequences, while others maintain that seniors today are healthier than their ancestors and that the costs of care are in many ways created more by the care we provide than by the inevitable demands of aging (Canadian Health Services Research Foundation 2001). From this latter perspective, changes are required primarily in the kind of care provided rather than in the amount of resources dedicated to care. One strategy, which may reduce costs while increasing comfort and dignity, would be to reallocate some resources to palliative or hospice care. In any case, more geriatric services will be required. More people with severe disabilities are living longer, as are others with chronic diseases. Some of these persons are demanding more control over their providers, and all need some care. More of the population has migrated from Asia, Eastern Europe, Latin America, and the Caribbean, and more of the population has Aboriginal origins. At the same time, there are fewer social supports provided by government, fewer children around to support family members in their old age, more women in the labour force, and fewer spouses as a result of higher divorce rates. Demand is also created independently of these demographic changes. The growth in advertising for drugs and treatments on traditional mass media and the Internet helps shape the demand for care, sometimes in inappropriate ways. 'The result of this deluge of information on the supposedly pervasive risks to personal health is that people feel much more vulnerable' (Petrie and Wessely 2002, 690). All

Planning for Care 129

these developments have implications for the structure of care delivery and the kind of care required, and thus for the kinds of change providers will have to address. To conclude, a successful human resource strategy in the health sector must include cooperation among employers, unions, and professional associations; recognition of the diversity of care settings, populations needing care, and occupations in health care; acknowledgment that people in unregulated occupations or providing unpaid care for family, friends, and neighbours are vital to the system; gender-sensitive approaches that take explicit account of the competing demands placed on women in particular; initiatives to retain older health care workers and help them share their experience with entrants to the sector; efforts to make fuller use of those who are underemployed in health care, who have left what they deem an unattractive field, or who have foreign credentials we could better recognize; and ongoing attention to shifts in the population needing care. Educating for Care The history of education for care is in part a history of increasing specialization, years of formal education and continuing studies (CIHI 2002, 18-29). At the same time, however, a growing number of care providers are quickly trained or have little or no training, especially on-the-job training. While most formal training happens in public institutions, there is also some formal workplace training. Approaches to education vary not only over time and among jurisdictions but also with approaches to care. Planning for care involves the consideration of all these. A central issue in planning for the future is who should receive formal health care training. Although there is a consensus that health is shaped by a host of factors, including the physical and social environment, health care is often treated as a determinant separate from the rest. Thus, those defined as health care professionals receive specialist training while those involved in ensuring the environment for care is safe have seldom been required to take courses on health. Defining support functions as hotel services reflects this approach, but there is at the same time a growing recognition that these providers, too, require particular skills. For example, research from Montreal demonstrates the importance of hospital cleaners in disease prevention (Messing 1998), and hospitals in British Columbia prefer to hire cleaners who

130 Pat Armstrong and Hugh Armstrong

have completed a three-month community college program (Cohen 2001, 6). Similarly, research on clerical staffing concludes that inconsistencies and the lack of training standards undermine both efficiency and quality of care, leading to recommendations for training programs that 'close the skills gap' (Hospital Employees Union et al. 1999, 13). Even though a great deal of care is now provided at home by family, friends, and volunteers, there is little formal training and even less consistency in addressing this skills gap. The lack of training, like that in some paid support services, reflects both the invisibility of the work and the assumption that women who do the bulk of the work are natural caregivers. The health consequences for providers and care recipients are often negative. At the same time, the notion that health care is a business like any other means that managers in health care are not required to have health care education. Indeed, some come directly from other sectors. For Henry Mintzberg, arguably Canada's foremost management expert, the cult of measurable efficiency has had immoral consequences. He illustrates his position with reference to health care, where 'we're starting to find out what we lost, but it took years to find out. They [i.e., managers who use cost-benefit analysis but lack substantive knowledge of health care] knew what they were saving instantly' (quoted in Swift 1999,19). The need to consider education for the entire range of people working in health care is linked in turn to education for teamwork. Teamwork has a long tradition in this area, especially among nurses. Today, nursing shortages, efforts to cut costs, and new management philosophies are encouraging strategies that integrate more LPNs, care aides, and others into care teams (Canadian Practical Nurses Association 2000). Similarly, there is renewed interest in promoting teams for primary care, involving nurse practitioners, midwives, therapists, social workers, pharmacists, and others. This could mean strategies to train existing staff by upgrading skills and preparing for teamwork. The Hospital Employees Union, for example, maintains that 'support for team nursing and increased practical nurse education will go far in alleviating those pressures and enhancing the quality of bedside care' (Hospital Employees Union et al. 1999, 3). Bridging programs would not only allow providers to acquire appropriate skills, but also permit providers to move from one category to another without undermining skill requirements. A focus on teamwork could also include more interdisciplinary and shared courses in initial education programs, as well as more training on working in collaborative teams.

Planning for Care 131

There is concern, however, that these strategies could lead to substitution rather than complementarity, and serve to reduce education and training periods instead of recognizing that all providers need a solid background in health. There is also concern that such courses would take time away from critical-skills training. As well, there is concern that one occupation and care model would dominate, with physicians and their bio-medical model in particular viewed with suspicion by some. In Ontario, mid wives opted for a separate university program in part because they feared physician dominance. Finally, the costs of ongoing training and education are also an issue, particularly for women, few of whom have the time or the money to take courses outside their paid work hours. Shared courses, interdisciplinary courses, and training in collaborative teamwork are not the only suggestions in course content for providers that would help them adapt to change. Flexner's famous report on health care, released early in the twentieth century, recommended that humanities courses would promote communication with patients and other providers. Only small progress has been made in this direction, in spite of many more subsequent inquiries supporting this development. Communication skills have received more attention and become increasingly important with a multicultural and highly educated population. The Internet, too, has contributed to the need for these skills, as has the focus on informed choice (Mechanic 2002). Courses on the structure of the health care system have also been suggested, especially in these times of rapid change. With inequality, culture, sex, and gender now recognized as determinants of health, these subjects, too, warrant attention. And the aging of the population, combined with a shift to community and palliative care, means that more emphasis needs to be placed on care for the elderly and care at home. A shift to health promotion strategies, along with more evidence-based practices, is also on the agenda. The problem here, of course, is that there is limited time in any program, although these issues could be addressed by integrating them into all course materials. Such material may also be uncomfortably placed within a curriculum that favours certain kinds of evidence (Muzzin 2001). Another partial solution would be to make greater use of procedures for prior learning assessment and recognition. Additional skills and content could be introduced in workplace education programs, with a focus on the particular needs of the location and population served. Workplace education can help providers learn

132 Pat Armstrong and Hugh Armstrong

about new technologies and new patient needs. Requiring that a certain proportion of the payroll be dedicated to education, as now happens in Quebec, could encourage this (Schetagne 2001, 20). Research on physician education suggests that the location of educational institutions also matters. People trained in rural areas are more likely to work there. The same may be true for those who receive much of their training in home care or long-term care facilities. Teamwork is not exclusively or even primarily learned in a classroom. Indeed, much of what providers need to know is learned tacitly, as they work with others and outside formal learning structures. In addition, the rapid changes in health care organizations and the continual introduction of new technologies mean that learning must be a regular aspect of all providers' work. Some professions require their members to take continuing education courses. Some are considering ways of assessing skills on an ongoing basis. Some employer-union contracts build formal learning programs into the workplace. But there is little consistency in continuing education policies across health care occupations and little research on their effectiveness. Little attention has been paid to educating those in long-term or home care for their new demands. Moreover, the move to care in isolated households means that there are few opportunities to learn from and share with others unless there are organized attempts to create such opportunities. Along with the emphasis on continuing education has come a renewed interest in tacit learning and mentoring. At the same time, however, it has become increasingly difficult for experienced providers to share what they know or learn new ways of practising. This is mainly because a business model focused on raising short-term productivity means that workloads have increased in ways that leave little time or energy to share or to learn. Research on nursing leads to a recommendation for greater flexibility in allowing older staff to work shorter hours while retaining their permanent, regular positions. "This flexibility for older staff would ensure that they continue to work and would allow proper mentoring of the new recruits by senior RNs, improve working conditions, and enhance quality of care. Such a policy requires sustained collaboration between employers and unions' (Kazanjian 2000, 6). Similarly, a report on skill shortages recommends that employers negotiate clauses in collective agreements that allow a gradual move to retirement while providing a 'period during which the transfer of knowledge to younger workers could be accomplished' (Schetagne 2001, 20). Of course, while this approach may conflict with

Planning for Care 133

the 'business-like' emphasis on measurable tasks and cost-cutting in care, it would mean more time for younger workers to learn and would also allow older workers to learn from younger ones. Although the focus in tacit learning is usually on individuals, groups also learn and do so within specific organizational contexts that influence how they learn. Examining the introduction of new cardiac surgery technology, a U.S. researcher has concluded that 'it may even be counter productive to encourage organizations to quickly adopt new technologies that require tacit knowledge for improvement and present a need for independence among users. This is because it is hard to explain how to improve teamwork and coordination' (Anonymous 2002, 26). Instead, alternative means such as participation in other, more experienced teams must be developed in order to transfer knowledge. New areas such as telehealth also mean new requirements for learning. Currently, phone-service lines rely on experienced nurses trained to provide other kinds of care. Assessments have focused more on whether these services reduce the use of emergency rooms and doctors' offices than on the kinds of skills required (Lattimer, et al. 2000). Other forms of telehealth such as monitors for home care use to replace provider visits and video-phone consultations among physicians while procedures are under way, have received even less assessment focusing on the skills required or the full range of patient needs. For example, there is little research on what happens to social supports when patients receive care through telecommunications systems. This still leaves the question of how learning is assessed and how graduates are recognized. There are wide variations across the country in how people move from education to practice and few ways of tracking movement after they do graduate. There have been calls for a single standard in education and in admission to practice in some fields, combined with a single registration number that would make it possible to track practitioners. This would require much more coordination among jurisdictions and with organizations representing providers. It would also involve the assessment of costs and benefits. Standardization is not supported by all. It could undermine the diversity that has developed in response to regional differences in needs and it could promote competition among jurisdictions for providers. Less attention has been paid to prior learning assessment and recognition. This procedure enables us to put to use knowledge and skills acquired through experience or in other countries, while benefiting from the diverse sources of such knowledge and skills. Bridging pro-

134 Pat Armstrong and Hugh Armstrong

grams, which facilitate the movement of practitioners from one health occupation to another, could also help address skill shortages. If the future holds more long-term and home care, more primary care teams and telehealth, and more technologically sophisticated interventions as well as more health promotion, then education programs within and outside the workplace will need to provide an environment that enhances the kinds of skills and relationships required for these quite different forms of care. Given that the future necessarily involves serving a more culturally and economically diverse population, and a gendered one, education programs must prepare providers to be sensitive to this diversity. Accountability and Governance in Care Debates about accountability in health care are as old as the Hippocratic oath. For much of the twentieth century, the emphasis was on ensuring that those defined as professionals had the necessary skills and adhered to ways of practising determined by their peers. Policies on scope of practice linked to licensing, certification, and registration3 reflected what professions defined as required skills. Indeed, the very definition of a profession was, to a large extent, based on the notion of self-regulation and an acquired body of theoretically based knowledge. The history of the twentieth century was also a history of processes whereby more occupations successfully sought professional status, gaining them both more recognition and more control over their work (Johnson 1972). From the 1960s onward, claims that professions were more about self-interest than protection of patients, combined with an increasingly educated public, led to demands for more involvement by patients in care decisions and more citizen participation in the regulation of professions (Blishen 1991). In response, lay members were added to professional regulatory bodies, their processes were made more transparent, and some patient rights were recognized (Flood and Epps 2001, 4). More recently, task-based regulatory models have been introduced to limit licensing to tasks considered to involve potential for serious harm. There is an increasing interest in formal skill maintenance and testing, as well as in surveys of patient satisfaction. Today, the emphasis is on evidence-based practice, on cost control, and on greater flexibility for managers in assigning tasks. Comparable 'calculative technologies make it possible to render visible [activities] of individuals, to calculate the extent to which they depart from norms of

Planning for Care 135

performance and to accumulate such calculations in computers and files and to compare them' (Peter Miller, quoted in Exworthy and Halford 1999,5). Accountability often means counting, and strategies such as those that base physicians' income on a fixed rate for each patient signed up for the service (capitation and rostering) seek to ensure that doctors are financially accountable for care. Much less attention has been paid to the unregulated occupations, although their responsibilities and their compensation are usually delineated in contracts. Unpaid providers seldom face any kind of formal accountability mechanisms. Accountability, then, is a complex term with quite different meanings. There are debates about whom providers are accountable to, for what actions, and through what processes. Are they accountable to their peers, to their patients, to their employers, to citizens who pay taxes, to governments that decide policies, or to all of these? Is accountability about money, practice, opinions, responsibilities, and relationships with colleagues? Is it about reporting and regular assessment, self-regulation and complaints, regular upgrading and education? There are also debates concerning how rigidly lines between providers' scopes of practice should be drawn. There are no simple answers to these questions, but rather tensions that must be balanced in determining strategies that will make it possible for positive changes to happen. First, there are tensions between managers/employers/governments and providers over control and the nature of their practice. Managers and governments are promoting a flexible labour force in ways that imitate similar practices in business. Multi-skilling would allow easy deployment and would mean new approaches to scope of practice, allowing more providers to do currently restricted tasks. And task division would allow both the substitution of some workers for others, freeing the most skilled for specialized tasks, and faster training for tasks defined as least skilled. The farther care moves away from hospitals and from regulated professions, the more flexibility is promoted. Against this position are those who support collaborative, integrated services based on a recognition of specialized, complementary skills and the need for comprehensive patient care, as well as on the desirability of self-regulation. For example, nursing organizations argue that care in long-term facilities requires regulated providers and skillupgrading courses to accommodate new care needs and changing technologies. Among both governments and provider groups, there are also disagreements about the degree of specialization required. RNs,

136 Pat Armstrong and Hugh Armstrong

for example, have long been divided over whether or not a BScN degree should be the minimum educational requirement. This continuing debate reflects, in part, different views of care. There is considerable evidence supporting the need for regulation and for highly skilled workers, especially in terms of care by RNs (CNAC 2002, 24). Licensing rules linked to scope of practice do help protect patients and providers. Indeed, there is a case to be made for extending regulations to emerging categories. Pharmacy technicians provide just one example (Peartree Solutions 2001). Yet variations across the country in terms of how scope of practice is defined and who is regulated, as well as variations within provinces in terms of who does what, suggest that the lines cannot be rigidly drawn. So does the overlap in areas of services covered. There is evidence indicating that nurse practitioners and midwives can safely do some of what doctors do. There is much less research on the impact of multi-skilling in health care or on the use of other providers. Again, pharmacy technicians provide an example. 'There is little disagreement that the organized use of properly trained unlicensed personnel can be beneficial to organizations, patients, and health care practitioners' (Backman 2000,33). What is required is a balance between skill protection and flexibility. Tensions between managers and providers arise not only over skills but also over control. Managers seek both to have greater control over what providers do as a means of planning and cost cutting and to have providers in general, and physicians in particular, take more responsibility for cost control. Bed reductions and care pathways are examples of the former, capitation and rostering of the latter. Salaries for doctors represent a middle ground between capitation and fee-for-service, with cost control delinked from procedures or number of patients, and with more potential for choice based on need. At the same time, providers maintain that peers are the only ones in a position to assess their care and that autonomy is required to ensure that they can use their judgment to apply evidence in an individual case. Some see autonomy as integrally linked to payment methods. Support workers, too, argue that they need flexibility in their schedules to allow them to decide on the quality required in particular cases. Moreover, trust is a critical and necessary part of the system, because providers cannot be under constant surveillance and because rules cannot be developed to cover every case. Here, as well, tensions need to be balanced rather than addressed by choosing one approach over the other. What is clear is that, when and to the extent that providers and their organizations

Planning for Care 137

participate genuinely and fully in designing and implementing change, they are less likely to resist it. More important, they contribute to better decision making about whether change should occur, how quickly, and in what directions (see, e.g., Carrier and Tolbert 1998, especially 20-3, 28). Second, there are tensions between patients and providers. Many, but not all, patients want to influence decision making and gain some say over their own treatment, while many providers seek to maintain the autonomy necessary to direct patient care based on their knowledge. Policy makers are encouraging patient and citizen participation in order to promote the appropriate use of cost-effective services, reduce dependence, improve responsiveness, and increase understanding or acceptance of policy decisions (Coulter 2001). Participation can be one means of addressing the growing demand for services. Patient surveys, decision aids, citizens' juries, and patient bills of rights have all been offered as ways to improve care and keep providers accountable. Critics suggest that patients are often not in a position to judge, because they lack the knowledge, time, or inclination and because their participation may reinforce prejudices against specific groups such as homosexuals or the mentally ill (Rosen 2001). Most patient assessment is summative rather than formative; that is, it grades the care rather than offering means to improve it. Administrative data on performance also have severe limitations, creating the impression of quality assessment while offering data that are often of little use in distinguishing among services (Naylor 2002). A U.S. study of hospital report cards found 'ratings poorly discriminated between any two individual hospitals process of care or mortality rates' (Krumholz, et al. 2002,1277). The Maclean's ranking of hospitals is similarly 'limited to simple parameters that cannot reflect the multidimensionality of treatment-effectiveness indices or health-care environments as these exist in the real world' (Page and Cramer 2001, 297). Moreover, patients may have few choices among services, especially in rural areas, so data may simply make them anxious about care while blaming providers for problems that are the result of the system rather than of a particular provider's practice. Indeed, various forms of participation may be primarily used to legitimate rationing decisions and shift blame. Patient participation makes no sense if there are no means for the system to respond, and is too often based on a 'consumer' or 'market' model of care that defines quality and access exclusively in terms of individual choice and purchasing care. Equally important, a

138 Pat Armstrong and Hugh Armstrong

strategy focused on patient demands ignores the ways the private, forprofit sector seeks to increase demand as a means of expanding sales. Direct-to-consumer advertising for prescription drugs provides just one example (Mintzes, Barer, Kazanjian, et al. 2002). There is a distinction to be made between individual patient participation in their own education, diagnosis, and treatment, on the one hand, and more collective, citizen participation in decision making regarding health care, on the other. There are also distinctions to be made between consultation and decision making, and between regular participation and occasional input. The case has successfully been made for the ongoing involvement of citizens on regulatory and governing bodies, where they gain experience and knowledge while bringing their own expertise to bear on the process. How they are selected (by appointment or election) and whom they represent (themselves or particular groups) are more contentious issues, as is the decision-making process itself. So is citizen consultation on what services should be provided for whom. While citizens serving on regional boards decide broad policy on services, providers still usually retain the right to define what is needed in the particular case, including what is defined as medically necessary. Similarly, there are issues related to the accountability of individual practitioners and those related to their participation as a group in decision making. In addition to promoting job satisfaction, their participation in management and governance allows the system to benefit from their expertise and encourages providers to be more open to change (Backman 2000, 28). Critics warn, however, that such groups mainly defend their own interests, and some fear that physicians in particular will dominate. What needs to be balanced here is not only providers' judgment and patients' or citizens' rights but also the responsibility of policy makers to decide, as well as take responsibility for their actions, and the responsibility and rights of patients. Third, there are tensions among providers. Some doctors have fought hard against the licensing of midwives, RNs against the use of LPNs, and so on. In doing so, they are protecting not only their selfinterest but also (their perceptions of) the quality of care provided under current conditions. For example, the literature suggests that care quality in hospitals and long-term facilities is directly related to the proportion of RNs: the higher their share of the work, the higher the quality of care (Shamian and Thomson 1999). Furthermore, 'the ability of nurses to practice their profession according to the professional standards and values is a key determinant to their satisfaction and commitment' (Backman 2000, 28). Breaking down care into tasks reduces their

Planning for Care 139

capacity to provide comprehensive care across the full spectrum of caregiving. Employment insecurity, combined with the effort to make care a series of easily learned tasks and declining autonomy, are thus important factors in the resistance to redefining scope of practice. Yet a move towards greater flexibility and teamwork need not require abandoning scope of practice regulations or undermining provider quality. It is possible to organize comprehensive care by expanding the group involved in care and by basing care on their combined skills and collaboration. It does require appreciation of the skill sets of other providers and a sharing of power. Such a strategy can simultaneously ensure quality while allowing a wider range of providers to participate and to share both knowledge and responsibility. Working in teams also makes daily work visible, thus encouraging accountability of providers to those who know about quality care. However, this kind of strategy requires employment and income security, as well as recognition of the need for training and the importance of skills. And teamwork cannot be the only solution, because teams are difficult to organize or inappropriate in some areas. Building accountability into change requires recognizing that accountability is about more than finances, and that finances involve values. It is also about the quality of care in both the short and the long term, and about trust in both the providers and the system. It is difficult to base accountability for care on a market model of competition and choice. In a market model, too much is short term and confidential, and too little is secure and continuous. If care is to be integrated, there will be too few organizations available to make choices among competitors meaningful. And the stakes are much too high and the means of assessment much too limited to base accountability on competition, even if there were enough providers to compete. There are tensions in all accountability strategies that need to be accommodated, and this accommodation means continuous collaboration among the full range of organizations representing providers, employers, patients, and governments. Moreover, providers are more likely to be open to new accountability processes if employment is secure. The Complexities of Planning for Care Social, economic, political, and demographic factors establish contexts for health care needs and demand. So do assumptions about the nature of care and how it should be delivered, because these assumptions shape what kind of care providers are required to deliver.

140 Pat Armstrong and Hugh Armstrong

If the emphasis is on the specificity of care, then collaborative teams based on the recognition of complementary skills will be promoted, although multiple models will be contemplated in order to accommodate the quite different needs of different populations and the variety of practice strategies. This means continued emphasis on specialized skills and scopes of practice, while allowing for more occupations such as midwives or paramedics and more overlaps in scope of practice. Emerging occupations such as pharmacy assistants and new assignments such as telehealth would require regulation and specific training. Those working in long-term care facilities and home care would participate in upgrading programs to meet new demands, and bridging programs would be encouraged. Attention would be paid to training support workers and unpaid providers. It would also mean the return of some middle managers who are trained in the specific fields they direct, and less emphasis on physician care in some areas. Equally important, providers and the organizations representing them would participate in decision making about change and play an important role in developing accountability mechanisms. If the emphasis is on business practices and costs, then teams based on much more flexible scopes of practice and limited training for specific tasks would be the norm both inside and outside facilities. Accountability would emphasize data collection, financial factors, and managerial control. Management training would focus on business skills, not health care skills. Patient participation would be encouraged, and so would public measurement of outcomes as means of promoting competition and results. Contracts would be the way to specify care, rather than relying on scope of practice and self-regulating professions. In addition to making assumptions about care, it is also necessary to decide about the role various levels of government play in planning, in education, in accountability, and in employment. The federal government now influences human resource planning through data collection, support for research and teaching, the principles governing the financing of most public health care, the provision of Aboriginal health care and health protection measures, and immigration policies. International agreements, signed by the federal government, as well as other pressures such as security concerns following the events of 11 September 2001, place some restrictions on the government's right to decide who enters and leaves the country, and what kinds of services are provided within the country. Other federal initiatives, such as support for primary care reform, also influence resources. But these differ-

Planning for Care 141

ent aspects of federal involvement are not well coordinated with each other, and there is little systematic planning for resources at this level. Although provinces and territories are more directly involved in human resource planning, they, too, lack systematic and coordinated planning mechanisms. Indeed, the current Ontario and British Columbia governments have eliminated their health care labour adjustment agencies. And reforms have reduced the control they did have in some areas, as have federal reductions in funds and changes in funding. 'Since almost all jurisdictions have decentralized health-care delivery by regionalizing or by implementing similar measures of devolution, at least in so far as care delivery by non-physician providers is concerned, government no longer possesses as many direct policy levers that can be applied to employment and/or deployment issues' (Kazanjian 2000, 10). The move to allow more private educational institutions and health services also further limits government planning, as do the continuing right of the regulated professions to control educational requirements, the absence of regulation for other health care providers, and international trade agreements. Currently, there are only limited common data on providers across the country, and even more limited centralized information on education, qualifications, and admissions to practice. A variety of reports have suggested that planning must begin with the systematic collection of such information and a means of tracking where individual providers go (e.g., Backman 2000; Kazanjian 2000; CNAC 2002). With such data, there would at least be a better understanding of current resources, their allocation, and skill mix. These reports have also recommended a national organization to 'identify strategies and methods of supporting better human resource planning' (Backman 2000, 81). This could include support for more consistency across jurisdictions in terms of education for and admission into practice. The 1995 Agreement on Internal Trade (AIT) promises both greater freedom of movement for members of regulated professions and a movement towards more common standards for entry and practice. Other occupations in health are not part of the plan. However, provinces and territories wish to retain at least some control, as do provider organizations, not only in order to protect their own interests but also to respond to local needs and resources. Moreover, if standards are the same and movement is easy, each jurisdiction will compete with the others for labour. At the same time, though, the female domination of the labour force may contribute to making providers less mobile.

142 Pat Armstrong and Hugh Armstrong

There are basically three sources of provider supply. One is the existing supply, including both persons currently employed, full- or parttime, and those unemployed, or those trained in health care but employed in other areas. Is there a shortage of skilled providers or a shortage of good jobs for skilled providers? The high injury rates, combined with the large numbers of persons not employed full-time in their profession, suggest the latter. U.S. research indicates that physicians' job satisfaction and willingness to practise decline as their control is reduced (Williams, et al. 2001). The same relationship almost certainly holds in Canada. Recall that a major survey of hospital nurses in several countries, including Canada, links high levels of job dissatisfaction to the current shortage (Aiken, et al. 2001). Appropriate working and educational conditions, as well as secure employment and trust, are essential ingredients to attract providers back into health care. Another source of supply is new recruits. There does not appear to be a shortage of people willing to enter education programs, although rapidly rising tuition fees seem to be changing who is entering programs and what kind of work they seek when they leave. This is especially the case for medical students, who are increasingly likely to come from households with high socio-economic status (Kwong, et al. 2002). Students in Ontario medical schools, where tuition fees have recently climbed steeply to become the highest in the country, also anticipate higher indebtedness upon graduation, at $80,000 for those entering in 2000 as against $57,000 for those who entered in 1997 (Dhalla, et al. 2002, Table 3). Meanwhile, graduates are flocking to residency openings in surgery, leaving 109 openings in the less lucrative field of family medicine unfilled in the initial 'matching' exercise (Sullivan 2002). There does seem to be a shortage of those willing to work in areas now requiring new recruits and a shortage of good jobs. Secure employment would undoubtedly encourage retention in these areas. So would the creation of more employer-sponsored training opportunities. Research on physicians in Ontario demonstrates that market solutions are unlikely to solve the location problem (Blomqvist and Tissaaratchy 2002). The high proportion of women, combined with changing male attitudes towards work and family responsibilities, means that more organizational accommodation to family needs is also required. The third source of supply is foreign-trained providers. This raises critical questions about appropriate skills and about relying on other countries' resources. There is little systematic assessment of this

Planning for Care 143

source, so it is difficult to suggest strategies beyond research on consequences and options. One advantage for Canada in employing foreigntrained providers is that they bring needed language skills and cultural sensitivity to serving segments of our diverse population. Demand is linked not only to the higher education of the population and access to the Internet but also to advertising designed to increase demand, especially for prescription drugs. The growing proportions of elderly, of people with chronic diseases and disabilities, and of people with capacity in neither French nor English also affect demand, as does the shift to long-term care facilities and home care. This means a changed emphasis in education programs, along with more integration of education into workplace activities. To sum up, federal and provincial governments have played important roles in shaping both supply and demand. These roles are diminishing, however, with international trade agreements, regionalization, funding reductions and instability, and privatization. There are few ways to track changes in human resources at the national, provincial, or local levels. Before more effective planning for change can be undertaken, decisions must be made about contributions at each level, and better data must be collected. And fine balances between standards and standardization, and between planning and choices, must be set. Conclusion There is currently an array of proposals for human resource planning in health care. And more proposals are being developed for the Commission. The challenge is to develop the means to assess them, rather than to add another one to the heap. This paper is intended to set out the kinds of questions that need to be asked about such work. While these questions are raised throughout the paper in more detail, they can be summarized as follows: • What model of health care guides the analysis and recommendations? Is health care understood as a business like any other or is it understood as different, with its own specific kinds of values, personnel, relations, work organization, decision-making structures, and demands? • How are the global, national, regional, local, and institutional contexts taken into account? • What social, economic, cultural, and other demographic compo-

144 Pat Armstrong and Hugh Armstrong

•

•

•

•

nents are included in the understanding of the changing demands for care and how will the proposed strategies balance or otherwise address these demands? Is the entire range of paid and unpaid providers considered, and are their current work structures recognized and the relationships among providers, as well as their specific characteristics, addressed? How is education understood and does this understanding encompass formal and informal learning for the full range of paid and unpaid providers? Is knowledge sharing seen as a critical aspect of daily practices, and continuing education as an integral part of workplace culture? How is accountability defined and does this definition extend beyond finances, scopes of practice, and certification regulations for some providers? Who is accountable to whom, for what, through what processes, for what time period, and on the basis of what evidence? Are the tensions between managers/governments and providers, between patients and providers, and among providers considered? Is employment security considered alongside flexibility? How much influence do providers affected by decisions have over their content and timing? Does planning for care recognize the tensions, diverse approaches, and complexity inherent in health care and the need to balance conflicting pressures in different ways, in different places, and at different times? Does it recognize, and make explicit, the roles of various levels of government, of providers, and of citizens?

Planning for health care will always be a somewhat messy and unfinished project. This paper describes some ingredients of the mess, and some basic issues and approaches to planning for what will always be a work in progress. NOTES 1 Care pathways or protocols are the detailed specifications of timed steps for treating identified, discrete diseases and injuries. 2 Elsewhere (e.g., Armstrong, Armstrong, and Connelly 1997) we have used the term 'privatization' to label not only the development of private sector funding and delivery, but also the transfer of work to private households

Planning for Care 145 and the adoption within the public sector of private-sector practices. This latter form of privatization, which is consistent with the care-as-anotherbusiness model, is often termed 'reinventing government' (see, for example, Osborne and Gaebler 1993; Treblicock 1994) or 'new public management' (see, for example, Aucoin 1995; Shields and Evans 1998). 3 Only those with the relevant licence are allowed to perform a given task, e.g., prescribe medications. Only those who are certified can use a specific occupational title, e.g., physiotherapist. Whether or not under a licensure or certification framework, a government agency or professional association may register members. This registration may be voluntary or compulsory for those in the occupational group (CIHI 2002, p. 24).

REFERENCES

Aiken, Linda H, et al. 2001. Nurses report on hospital care in five countries. Health Affairs 20 (3): 43-53. Akyeampong, Ernest B. 2001. Fact-sheet on unionization. Perspectives on Labour and Income 13 (3): 46-54. Anonymous. 2002. Learning how and learning that: Different patterns of performance improvement for tacit and explicit knowledge. Available at http://kuuc.chair.ulaval/english/master.php?url=bulletin.php. Appelbaum, Steven H., Andrea Everard, and Loretta T.S. Hung. 1999. Strategic downsizing: Critical success factors. Management Decision 37 (7): 535-52. Armstrong, Hugh, Pat Armstrong, and M. Patricia Connelly. 1997. The many forms of privatization. Studies in Political Economy 53 (summer): 3-9. Armstrong, Pat, et al., eds. 2001. Exposing privatization: Women and health reform in Canada. Aurora:, Ont: Garamond. Armstrong, Pat, et al. 1997. Medical alert: New work organizations in health care. Toronto: Garamond. Armstrong, Pat, and Olga Kits. 2001. One hundred years ofcaregiving in Canada. Ottawa: Law Commission of Canada. Aucoin, Peter. 1995. The new public management: Canada in comparative perspective. Montreal: Institute for Research on Public Policy. Backman, Allen. 2000. Job satisfaction, retention, recruitment and skill mix for a sustainable health care system. Report to the Deputy Minister of Health for Saskatchewan. Baumann, Andrea, and Linda O'Brien-Pallas. 2001. The status of the nursing workforce in Ontario: The numbers and the worklife issues in November 2001. Report submitted to the Ontario Nurses' Association.

146 Pat Armstrong and Hugh Armstrong Benner, Patricia. 1984. From novice to expert: Excellence and power in clinical nursing practice. Menlo Park, Calif: Addison-Wesley. Blishen, Bernard. 1991. Doctors in Canada. Toronto: University of Toronto Press. Blomqvist, Ake, and Piyanjali Tissaaratchy. 2002. Patterns of physician service utilization and supply: Recent evidence from Ontario. Currently under review. Burke, Ronald J.; and Esther R. Greenglass. 2000. Effects of hospital restructuring on full time and part time nursing staff in Ontario. International Journal of Nursing Studies 37:163-71. Canadian Health Services Research Foundation. 2001. Myth: The aging population will overwhelm the healthcare system. Links 4 (4), available at www.chsrf.ca. Canadian Practical Nurses Association. 2000. Primary health care: What is it and where do licensed practical nurses (LPNs)fit? Ottawa: Canadian Practical Nurses Association. Carrier, Betsy, and Jennifer Tolbert. 1998. The health care workforce at safety net hospitals and health systems: Future directions and strategies for change. Washington: National Association of Public Hospitals and Health Systems. CIHI [Canadian Institute for Health Information]. 1999. Supply and distribution of registered nurses in Canada. Ottawa: CIHI. - 2002. Canada's health care providers. Ottawa: CIHI. CNAC [Canadian Nursing Advisory Committee]. 2002. Our health, our future: Creating quality workplaces for Canadian nurses. Ottawa: Health Canada for the Federal/Provincial/Territorial Ministers of Health Advisory Committee on Health Human Resources. Cohen, Marjorie Griffin. 2001. Do comparisons between hospital support workers and hospitality workers make sense? Report prepared for the Hospital Employees Union [CUPE]. Coulter, Angela. 2001. Citizens and patients as partners in health services - an overview of various methodologies to include citizens and patients in decision making, in European health forum Gastein 2000: Information and communication in health. Bad Hofgastein, Austria: Gunther Leiner, 155-62, available at www.efhg.org. Dhalla, Irfan, et al. 2002. Characteristics of first-year students in Canadian medical schools. CMAJ166 (8): 1029-35. Exworthy, Mark, and Susan Halford. 1999. Professionals and managers in a changing public sector: Conflict, compromise and collaboration? in Exworthy and Halford, ed., Professions and the new managerialism in the public sector. Buckingham, U.K: Open University Press, 1-17. Flood, Colleen, and Tracey Epps. 2001. Can a patients' bill of rights address

Planning for Care 147 concerns about waiting lists? Draft working paper, Health Law Group, Faculty of Law, University of Toronto. Gold, E. Richard. 2002. Health care reform and international trade, in Timothy A. Caulfield and Barbara von Tigerstrom, ed., Health care reform and the law in Canada. Edmonton: University of Alberta Press, 223-44. Gottlieb, Laurie N. 2000. Editorial: shortage of nurses, shortage of nursing. Canadian Journal of Nursing Research, 32 (3): 3-5. Hassen, Philip. 1993. Rxfor hospitals: New hope for medicare in the nineties. Toronto: Stoddart. Hollander, Marcus, and Angela Tessaro. 2001. Evaluation of the maintenance and preventive model of home care: Final report. Victoria: Hollander Analytical Services. Hospital Employees Union. 1999a. Briefing note on health human resources research and planning: Focus on practical nursing. [Burnaby, B.C.]: Hospital Employees Union. - Vancouver/Richmond Health Board, and Healthcare Labour Adjustment Agency. 1999b. Executive summary report on research into information technology staffing. Johnson, Terrence J. 1972. Professions and power. London: Macmillan. Kazanjian, Arminee. 2000. Nursing workforce study, Volume V: Changes in the nursing workforce and policy implications. Vancouver: Health Human Resources Unit, Centre for Health Services and Policy Research, University of British Columbia. Koehoorn, Mieke, et al. 2002. Creating high-quality health care workplaces. CPRN Discussion Paper no. W14. Ottawa: Canadian Policy Research Networks. Krumholtz, Harlan M., et al. 2002. Evaluation of a consumer-oriented Internet health care report card. JAMA, 287 (10): 1277-87. Kwong, Jeff C, et al. 2002. Effects of rising tuition fees on medical school class composition. CMAJ 166 (8): 1023-28. Lam, Helen, and Yonatan Reshef. 1999. Are quality improvement and downsizing compatible? Relations industrielles/Industrial Relations 54 (4): 727-44. Lattimer, Val, et al. 2000. Cost analysis of nurse telephone consultation in out of hours primary care: Evidence from a randomised controlled trial. BMJ 320 (15 April): 1053-7. Levine, David. 2000. A letter from David Levine, in The merger decade: What have we learned from Canadian health care mergers in the 1990s? A report on the Conference on Health Care Mergers in Canada organized by the Ottawa Hospital and the Association of Canadian Teaching Hospitals. Ottawa: Canadian Health Services Research Foundation.

148 Pat Armstrong and Hugh Armstrong Markham, B., and Jonathan Lomas. 1995. A review of the multi-hospital arrangements literature: Benefits, disadvantages and lessons for implementation. Health Care Management Forum 8 (3): 24-35. Mechanic, David. 2002. Socio-cultural implications of changing organizational technologies in the provision of care. Social Science and Medicine 54: 459-67. Messing, Karen. 1998. Hospital trash: Cleaners speak of their role in disease prevention. Medical Anthropology Quarterly 12 (2): 168-87. Mintzes, Barbara, Norns Barer, Arminee Kazanjian, et al. 2002. An assessment of the health system impacts of direct-to-consumer advertising of prescription medicines (DTCA). Volume I: Executive summary. Vancouver: Health Policy Research Unit, Centre for Health Services and Policy Research, University of British Columbia. Morris, Marika. 2001. Gender-sensitive home and community care and caregiving research: A synthesis paper. Final report to the Women's Health Bureau, Health Canada. National Forum on Health. 1997. Values Working Group Synthesis Report, in Canada health action: Building on the legacy. Volume II: Synthesis reports and issues papers. Ottawa: Minister of Public Works and Government Services. Naylor, C. David, 2002. Editorial: Public profiling of clinical performance. JAMA 287 (10): 1323-5. Osborne, David, and Ted Gaebler. 1993. Reinventing government: How the entrepreneurial spirit is transforming the public sector. New York: Penguin. Page, Stewart, and Ken Cramer. 2001. Maclean's rankings of health care indices in Canadian communities, 2000: Comparisons and statistical contrivance. Canadian Journal of Public Health 92 (4): 295-8. Peartree Solutions. 2001. A statistical analysis of human resource issues in the pharmacy profession in Canada: Detailed report. [Ottawa]: Canadian Pharmacists Association for Human Resources Development Canada. Petrie, Keith J., and Simon Wessely. 2002. Editorial: modern worries, new technology, and medicine. BM] 324 (23 March): 690-1. Pollock, Allyson M., and David Price. 2001. Rewriting the regulations: How the World Trade Organization could accelerate privatisation in health-care systems. The Lancet 356 (9 December): 1995-2000. Rosen, Per. 2001. Citizens and patients as partners in health services - Is it necessary to include patients and citizens in the discussion on priority setting? in European Health Forum Gastein 2000: Information and communication in health. Bad Hofgastein, Austria: Gunther Leiner, 162-5, available at www.ehfg.org. Sanger, Matthew. 2001. Reckless abandon: Canada, the GATS and the future of health care. Ottawa: Canadian Centre for Policy Alternatives.

Planning for Care 149 Schetagne, Sylvain. 2001. Building bridges across generations in the workplace: A response to aging of the workforce. Vancouver: Columbia Foundation. Shamian, Judith, and Donna Thomson. 1999. The effectiveness and efficiency of nursing care: Cost and quality. Ottawa: Canadian Federation of Nurses' Unions. Shields, John, and B. Mitchell Evans. 1998. Shrinking the state: Globalization and public administration 'reform.' Halifax: Fernwood. Sicotte, Claude, Jean-Louis Denis, and Pascale Lehoux. 1998. The computerbased patient record: A strategic issue in process innovation. Journal of Medical Systems 22 (6): 431^3. Stein, Janet Gross. 2001. The cult of efficiency. Toronto: Anansi. Sullivan, Patrick. 2002. Medical students grab surgery slots, ignore FT posts. CMAJ, 166 (8): 1077. Swift, Jamie. 1999. Saving the corporate soul. Canadian Forum 78 (June): 16-21. Trebilcock, Michael J. 1994. The prospects for reinventing government. Montreal: C.D. Howe Institute. Wagar, Terry H., and Kent V. Rondeau. 2000. Reducing the workforce: Examining its consequences in health care organizations. Leadership in Health Services 13 (3): i-viii. White, Jerry P. 1990. Hospital strike! Women, unions and public sector conflict. Toronto: Thompson Educational Press. Williams, Eric S., et al. 2001. Understanding physicians' intentions to withdraw from practice: The role of jo"b satisfaction, job stress, mental and physical health. Health Care Management Review 26 (1): 7-19. Woodward, Cristel A., et al. 2000. Changes over time in attitudes towards health care policy options. Research Working Paper 00-01, Centre for Health Economics and Policy Analysis, McMaster University.

5 How Do Human Resources Policies and Practices Inhibit Change in Health Care? A Plan for the Future GAIL TOMBLIN MURPHY AND LINDA O'BRIEN-PALLAS

Canada's health care system has undergone enormous change over the past few years through reform and economic rationalization. These initiatives have created new challenges for what is known as 'health human resource planning' (HHRP). Every health profession has been affected in terms of levels of demand for their services, roles played, and skills required. Positions have been downsized, restructured, and in some cases eliminated. The nature and scope of practice within and among health professions have been altered, often dramatically. Effective HHRP has become critically important in this environment of change. However, current work in this area is intermittent and there is little evidence that current planning practices consider trends in society, determinants of health, needs of consumers, and the unique and shared knowledge and skills of providers. Typically, planning has been based on utilization patterns in the health care system and/or current fiscal commitment. There is growing consensus in the literature that the continuous cycles of over- and under-supply of human resources in health care world-wide reflects the inadequate projection methods used to estimate future requirements for expanding health systems and/or the failure to consider the evidence supplied by ongoing labour-market trends. HHRP in most countries has been poorly conceptualized, intermittent, varying in quality, profession-specific, and without adequate vision or data upon which to base sound decisions. In Canada, the growing recognition of the serious consequences of poor HHRP practices has led the Canadian Institutes for Health Research (CIHR) and the Canadian Health Services Research Founda-

How Do Human Resources Policies Inhibit Change? 151

tion (CHSRF) to identify HHRP as the dominant health policy issue for the next five years (Birch, et al. 2002). To ensure system efficiency and effectiveness, planning activities should be needs-based, responsive to a changing system, and outcome-directed (O'Brien-Pallas 2002). Introduction The three specific questions posed by the Commission on the Future of Health Care in Canada regarding HHRP can be seen to intersect around a single common theme: that human resources in health care must be matched as closely as possible to the health needs of the population. It is, however, precisely in this regard that historical approaches to HHRP have failed. The recurrent cycles of over- and under-supply that continue to plague Canada and other OECD countries can in part be traced to the fact that, while the stated goal of HHRP is to match human resources to needs, the theoretical models and empirical tools used to inform allocation decisions in health care have concentrated primarily on measuring demand for services (as measured by utilization rates), thus making them poorly suited to the creation of a useful predictive framework. It is being increasingly recognized that factors which affect the matching of supply to needs (rather than simply matching supply to demand) are many and varied. Recognition of this fact suggests the need for the development, testing, and utilization by decision makers of a dynamic model that adequately accounts for both the number of influencing factors and how each factor is likely to influence others. Accordingly, we structure this around a 'Health Human Resources (HHR) Conceptual Framework' which we and colleagues have developed and which has been adopted recently for use by the Canadian Institute for Health Information (CIHI) and adapted for the World Health Organization (WHO). Our purpose in employing this conceptual framework here is not to promote the framework per se but rather to use it as a reference point in our discussion of the commission's questions and to show how the various factors that influence HHRP cohere. As with any other journey, unless we have a road map, the odds are small that we will take the most direct route. According to Hall (1993), the human resource process comprises three interrelated steps: planning, production, and management. The focus on one component at the expense of the others does little to ensure an effective and efficient health care system (O'Brien-Pallas, et al. 2000). Yet a review of HHRP work done to date reveals that these conceptual linkages have

152 Gail Tomblin Murphy and Linda O'Brien-Pallas Figure 5.1 HHR conceptual framework

Source: O'Brien-Pallas, Tomblin Murphy et al., 2001.

not been made. Our dynamic, open-system-based framework considers these linkages by defining the key elements of human resource planning (O'Brien-Pallas 2002) (see Figure 5.1). We have found the framework to be an invaluable aid in navigating our way through the varied and rapidly changing terrain of our own HHRP research and modelling activities; therefore, it seems quite natural that we should adopt it as the road map to guide us in responding to the important questions posed by the commission. Although we will touch on various aspects of the model throughout the paper, it will be helpful at this point to define briefly some of its key elements in order to orient the reader to the complexity and dynamic nature of HHRP and to indicate to which questions particular elements of the framework are most relevant. • Population health needs reflect the multivariate characteristics of individuals in the population that create the demand for curative as well

How Do Human Resources Policies Inhibit Change? 153

•

•

• •

•

•

as preventive health services; addressing the health needs of the population provides the motive, context, and justification for HHRP practices. (Questions 1, 2, and 3) The production element of the framework highlights that, in order to ensure future capacity to meet population health needs, these needs must be considered in setting production targets for health education and training programs. (Question 1) The supply element reflects the actual number, type, and geographic distribution of regulated and unregulated providers; it also recognizes that supply is fluid and is related to production elements as well as to factors such as recruitment/retention, licensing, regulation, and scope of practice. (Questions 1 and 2) Planning and forecasting reflect the varieties of available HHRP practices and models as well as their assumptions, methods, data requirements, and limitations. (Question 3) Health, provider, and system outcomes refer to establishing the effectiveness and quality of human-resource practices by examining their effect on population health, provider health, job satisfaction, and system costs and efficiencies. (Question 2) Efficient mix of human and non-human resources (e.g., fiscal resources, physical plant, space, supplies, equipment, and technology) reflects the number and type of resources that must be developed in order to achieve the best population, provider, and system outcomes. Context elements (represented in the outer band of the framework) speak to the need to recognize factors outside the health care system that influence population health, the health system, and the HHRP process. (Questions 1, 2 and 3)

Using this conceptual framework as a heuristic device will allow us to locate more precisely and better articulate the strengths and limitations of various approaches to HHRP. Employing the framework will also force us to keep in mind the subtle interplay of variables in a way that will help us predict the direct and indirect consequences of proposed or anticipated change in any one factor. We will proceed by drawing on the existing academic literature, the 'grey' (unpublished) literature, and the experiences of jurisdictions in Canada and other OECD countries to describe, analyse, evaluate, and offer our opinions (i.e., recommendations) concerning HHRP practices. Specifically, for each question we will: • describe the current state of practice and knowledge about HHRP;

154 Gail Tomblin Murphy and Linda O'Brien-Pallas

• compare and contrast the common approaches to HHRP; • discuss the relationships among the various factors affecting HHRP (with emphasis on those factors that relate directly to the commission's questions); • articulate the strengths and limitations of the various approaches to HHRP as these relate to the issues and choices facing the commission; • suggest ways that knowledge acquisition, dissemination, and 'uptake' (the incorporation of knowledge into the decision making policy process) can be enhanced; • suggest steps that can be taken in both the short and the long term to enhance HHRP research; and • urge the commission to make recommendations that promote and support HHRP models, practices, and strategies that are needsbased and explicitly recognize and embrace the complexity and dynamic nature of human resource allocations in health care. Question 1 Do current HHR educational programs produce a workforce with the capacities and perspectives to meet the needs of the population in an environment of a rapidly changing and increasingly integrated health care system? If not, what changes should be made? Are policies and practices governing scope of practice, certification, and licensure of health care occupational groups up-todate and consistent with contemporary health system goals and directions? If not, how might they be changed to promote desirable innovation and change while at the same time preserving quality and protecting the public interest?

These questions relate directly to both the production and the supply elements of HHRP. Recall from the HHR Conceptual Framework that, in order to meet population health needs, these needs must be considered in setting production targets for health education and training programs. Typically, however, production targets have been established simply on the basis of employer projections, which are in turn typically based on historical utilization rates together with the number of workers historically designated to service those rates. Although, in some cases, demographic change has been used to inform these projections, in most cases this has not gone beyond consideration of simple age and sex distribution. Furthermore, the actual health needs of the population have not been likewise accounted for. While it is clear that both demographic and epidemiological data are useful in determining the num-

How Do Human Resources Policies Inhibit Change? 155

ber of learners required to create an adequate future workforce, these factors, together with the context elements associated with changes in the funding, values, and delivery of health care, also have a role to play in the way the future workforce is educated. Specifically, the political and geographical factors of the HHR Conceptual Framework are key components that influence all aspects of HHRP. Decisions about HHR are not always rationally based. To the contrary, decisions regarding education and recruitment are often taken in order to quell community outrage and/or to offer political support to particular communities. Recall further that the supply element of HHRP reflects the actual number, type, and geographic distribution of regulated and unregulated providers; it also recognizes that supply is fluid and related to production elements as well as to factors such as recruitment/retention, licensing, regulation, and scope of practice. Again, it must be emphasized that the supply element, like the production element, is both influenced by and has an influence on population health needs. Much like the production element, the supply element is also greatly influenced by a variety of contextual factors. It is critical to have the right providers working with Canadians to meet their health needs. We need to consistently ask ourselves if we have the right (appropriately educated and trained) health care providers working in the right location to meet the health needs of the population. While there is no unambiguous right number and mix of health professionals, all health care professionals need to practise to their fullest capacity, within their scope of practice. In a paper recently published by the Commission on the Future of the Health Care in Canada (2002), it is pointed out that there are more than thirty 'regulated' health professions in Canada and that their roles, while said to be unique, often overlap. Changing scopes of practice based on the needs of the population is a challenge since there are tensions between and among health professions that are complicated by union contracts and issues of self-regulation. However, the evidence is mounting regarding positive health, system, and provider outcomes associated with multiskilled workers, higher ratios of registered nurses (RNS), nurses working in advanced-practice roles, and baccalaureate preparation (Horrocks, Anderson, and Salisbury 2002; Needleman, et al. 2002). There is currently a great deal of debate (more so in political circles and hospital corridors than in the literature) about the increasing length of educational programs and the growing number of health

156 Gail Tomblin Murphy and Linda O'Brien-Pallas

occupations now requiring a four-year university degree for entry to practice. The common perception is that this lengthens the time it takes to get new providers into the workforce and is counterproductive in meeting the demands of a sector starved for human resources. A common solution being proposed is to shorten educational programs, make them more task-specific, and introduce locally determined competency requirements and certification processes that would be under substantial control of jurisdictional governments and local health care facilities. While it is easy to understand why such an approach has gained political cachet, there is no evidence to suggest that it would provide anything more than an extremely short-term fix for the longstanding problems of episodic over- and under-supply that are known to be associated with poor planning. Furthermore, such approaches fail to recognize that human resources differ in kind from non-human resources. Human resources are unique insofar as people have preferences and exercise these preferences in ways that make them less predictable, and therefore less amenable to distribution, than non-human resources. Human resources are likely to resist and undermine distribution schemes that do not accord well with their preferences or that they find otherwise unacceptable (Hubert 2002). We know, for example, that a sense of autonomy and control over decisions that affect individuals and their work leads to enhanced job satisfaction, which in turn has a positive impact on retention (as well as on a number of work-performance indicators). What is clear, however, is that, given the rapid rate of change in the way health care is delivered, preparing graduates only for present conditions is in fact preparing them for the past. The focus of health professions' education must be on preparing graduates for a future that remains largely unknown. This means providing them with knowledge and skills that are easily transferable and that will allow them to manage uncertainty and adapt readily to change. According to the American Council of Medical Colleges (2001), a well-rounded health professional demonstrates the following attributes: literacy, clinical competence, life-long learning skills, evidence-based practice, interdisciplinary teamwork, balance between health care and disease-prevention/health promotion, professional values and ethical behaviour in practice, optimal use of resources, and consciousness of well-being of self and colleagues. The answer to whether providers with these attributes are in fact being educated in the current system will depend to a significant degree

How Do Human Resources Policies Inhibit Change? 157

on which professional group we are talking about. Since various occupational groups typically have different roles within the system, their education and training is likely to vary accordingly. Furthermore, while task-specific learning (i.e., training) can be accomplished equally well for most health occupations in a variety of settings and via a number of program delivery models (e.g., colleges, hospitals, technical schools, universities, etc.), others among the above-described attributes are best learned within the academy (i.e., universities). We have learned from the experience of nursing that one of the advantages of a broad-based education in health care is that it tends to prepare graduates better for change. Role expansion, changes in practice venue, and additional taskspecific learning are easier to introduce and to adapt to if built on a strong platform of undergraduate education. According to O'Brien-Pallas, Doran, et al. (2002), clients cared for by degree-prepared nurses in the community demonstrated improved knowledge and behaviour scores. Similarly, visits performed by degree-prepared nurses resulted in fewer total visits and improved RN perceptions of visit adequacy (O'Brien-Pallas, Doran, et al. 2001). Education in Primary Health Care Initiatives The delivery of primary health care is continually touted as being a high priority for policy makers and health care leaders in every jurisdiction throughout the country. Yet, while significant efficiencies and gains in population health have been reported in OECD countries that have moved towards a primary health care model, there is little evidence in Canada that current strategies to redesign care settings have incorporated the principles of primary health care. This is due to a number of factors. First, there is a lack of understanding about what primary health care actually means; there continues to be a misperception that it should be physician-centred. Second, the requisite shift in policy as well as a change in thinking among health care workers, bureaucrats, politicians, and the public have not yet taken hold. For example, while university schools of nursing across the country have for several years now been structuring their curricula around a focus on primary health care, medical faculties have been slow to do likewise. We believe that the tardiness to respond to the need for change in educational programs based on the principles of primary health and the needs of people speaks to the continued domination of the physician model of health care.

158 Gail Tomblin Murphy and Linda O'Brien-Pallas

Much of the resistance to adopting a primary health care model has to do with the continued limitations placed by physicians on the scope of practice of other health professionals - this, despite the fact that we know outcomes for people have been positively associated with nurses working in advanced-practice roles. For instance, nurses have proven to be efficient and effective care providers who need to be utilized according to evidence that shows where they can have the greatest impact on population health. Numerous studies have demonstrated the effectiveness and efficiency of utilizing RNs in a variety of settings and roles supportive of direct clinical services and population-based approaches (Prescott 1993; Brown 1994; Shamian and Chalmers 1996). Nursing interventions have been associated with positive health outcomes involving personal health knowledge, treatment adherence, health maintenance, follow-up return, accessibility of care, and patient satisfaction (Brown, 1994; Shamian and Chalmers 1996). Nursing actions have also been associated with increased capacity to function well in daily life (Prescott 1993; Doran, et al. 2001). Data show that advanced-practice nurses facilitate continuity of care, access to care in rural and other settings, and the provision of care to underserved populations (Shamian and Chalmers 1996). Advanced-practice nursing results in enhanced health outcomes such as reduced rates of smoking and alcohol use (Green and Simons-Morton 1984), shorter hospital stays (Brooten, et al. 1989; Cohen 1991; Ethridge 1991; Ethridge and Lamb 1989; Lamb and Huggins 1990; Marchette and Holloman 1986; McKenzie, Torkelson, and Holt 1989; Naylor 1990), decreased hospital admissions, and more appropriate office visits (Prescott 1993; Shamian and Chalmers 1996). It has also been shown that better health outcomes occur when the advanced-practice nurse and physician work together (Schultz, et al. 1977). Such nursing interventions enable patients to maintain their health and thus reduce the costs associated with more expensive medical and emergency services (Brown 1994; Denton, et al. 1983). More recently, a systematic review of whether nurse practitioners working in primary health care can provide equivalent care to doctors concluded that increasing the availability of nurse practitioners in primary health care is likely to lead to high levels of patient satisfaction and high-quality care (Horrocks, Anderson, and Salisbury 2002). Studies designed to determine the impact of the role of other health professionals on health outcomes are not apparent in the literature. It appears clear that, at least in the case of nursing, the current system of education has served the population well and prepared nurses for

How Do Human Resources Policies Inhibit Change? 159

changing roles within the health care system. This is not to say that alterations in the way that we educate nurses (i.e., movement towards baccalaureate education) are not required as health systems and population health needs are better understood. However, policy makers need to focus their immediate attention on why it remains the case that nurses are underutilized, and why forms of health care delivery that are known to enhance their capacity to affect population health positively (i.e., primary health care) are being resisted. Commissioner Romanow will recall that when, during a recent public consultation, he asked a group of Ontario nurses why they had not had a greater impact on making primary health care a reality, the powerful one-word answer given by a single nurse was 'power/ Educating for Teamwork Human resources do not provide care in isolation; instead, they use their knowledge and skills in combination with other human and nonhuman resources (e.g., fiscal resources, physical plant, space, supplies, equipment, and technology) to provide services to meet the needs of the population (Birch et al. 2002). It has long been recognized that both institutional health care and primary health care require a high degree of teamwork. Educational programs for all health care personnel need to do a better job emphasizing and developing the skills, attributes and, most important, dispositions required for working effectively on complex teams. This is important since we know that teamwork reflects the reality of much of modern health care; is associated with increased efficiencies and effectiveness; enhances inter-professional cooperation and decreases turf wars; reduces errors and increases patient safety; and has a positive impact on working relationships and job satisfaction. However, other than within a few progressive HHR education programs (and in even fewer workplaces), there is little in the way of meaningful opportunities to create well-functioning health care teams. The way health care professionals combine knowledge and skills for the shared benefit of patients has direct implications for the health care production function (i.e., the relationship between the levels and mix of resources used to produce health care services and the quantity and quality of services produced). The challenges in providing team-oriented educational programs are particularly daunting in rural areas. However, with the advent of technologies (e.g., Web Course Tool, other distance-delivery innovations), creative teaching strategies may facili-

160 Gail Tomblin Murphy and Linda O'Brien-Pallas

tate ongoing dialogue between and among the health professions during their participation in educational programs. It is also essential that team teaching and learning experiences are carried through to, and made available in, the clinical setting. It is important to re-emphasize that all HHR educational programs must be planned and delivered based on the health needs of the population. Health, social, political, economic, and technological advances must be considered in the planning, implementation, and evaluation phases of educational programs. Meeting the needs of populations requires an integrated approach to educational programming for health professionals. Certification, Licensure, and Scope of Practice A common theme appears to be surfacing in many health care jurisdictions, namely, that existing policies and practices related to certification, licensure, and regulation (particularly self-regulation) present unreasonable obstacles to positive change, innovation, and better health care. There seems to be a growing consensus among bureaucrats and health care CEOs that they would be better able to improve health care and ensure adequate supplies of the right type and mix of health care workers if only the traditional professional boundaries could be dissolved and replaced with jurisdictional regulation and bureaucratically mandated scopes of practice. There is nothing in the form of empirical evidence to support this claim. It is, however, easy to see why policy makers interested in finding a quick fix for pressing HHR shortages would be attracted to the notion of cutting the length of educational programs and putting scope of practice under their own control. However, in the absence of clear empirical evidence that current educational, competency, or certification requirements create a real barrier to meeting the health needs of the population, strategies that involve wholesale change to current policies and practices need to be introduced only as the outcome of widespread consensus among relevant stakeholders in the context of open and respectful dialogue. Given that there are a variety of higher-priority (and less politically troublesome) issues at hand, we suggest that the discussion of such strategies is premature. Furthermore, there has not yet been reasonable expert or public/political consensus reached about what one has in mind when speaking of the 'desirable innovation and change' that current policies and practices are ostensibly blocking.

How Do Human Resources Policies Inhibit Change? 161

This is not to say, however, that significant positive change may not come about as a result of alterations in the scope of practice and flexibility in establishing a wider range of shared competencies among various health professions. In fact, the philosophy of primary health care to which we subscribe depends on just such a willingness to be flexible about the boundaries of each profession. Whether primary health care is a success will, in large part, be a function of the degree to which physicians are willing to relinquish authority and blur scope of practice boundaries. This, however, does not require abandoning professional self-regulation, particularly if this is done in favour of centralized bureaucratic regulation. Such a move would likely diminish rather than enhance public confidence and trust in the health care system. Health care professionals (especially nurses) have higher public trust ratings than other occupations (including the clergy) and certainly higher than politicians and bureaucratic mechanisms (Pollara 2000). In addition, the costs involved in replacing existing licensing and regulatory structures for all health disciplines that currently regulate their own activities at their own expense would add an additional financial burden to the health care system, with the possible consequence that health services would be further jeopardized. We want to make a further point about the role of self-regulation, certification, and licensure. Messages from a variety of stakeholders suggest that the collective-bargaining voice of health occupations is drowning out other reasoned voices and has, to some degree, marginalized the voices of those who have as their mandate public safety (i.e., professional bodies). In the case of nursing, for example, the media focuses on labour strife and demands for higher pay, and those who have an interest in preserving the status quo claim that if nurses are allowed to provide services historically provided by physicians, they will soon be demanding similar levels of compensation, with the consequence that any intended savings to the health care system will be lost to new wage demands. What the media and others typically do not capture is how the scope of practice of all professionals could be realigned to better meet the health needs of the population more effectively. Unlike labour unions, which are unapologetically intended to serve first the interests of the group, professional associations and similar governing bodies are ostensibly designed to protect the public and enhance service through mechanisms that promote and enforce a commitment to standards and values. The downside of professionalization is that exclusive territories are created in which a particular group is

162 Gail Tomblin Murphy and Linda O'Brien-Pallas

allowed to operate without undue interference from governments or the encroachment of other professions. While created in the name of public protection, the line between public interest and professional self-interest is, by its nature, an extremely thin one. Care must be taken to ensure that public safety is maintained without having the quality of care diminished through the misappropriation of professional structures by factions more concerned with self-promotion and gain. Question 2 What are the main strengths and weaknesses of current approaches to HHRP nationally, provincially, and territorially? What changes should be made in the approach to HHRP to ensure an adequate supply of qualified personnel to meet the needs of the population? What impact would alterations in current policies and practices governing scope of practice, certification, and licensure of health care occupational groups have on the future health care workforce? How will we balance the losses in this regard with the gains?

The bulk of HHRP in Canada and its jurisdictions continues to rely on outdated approaches. Most jurisdictions employ approaches based on supply, given that in most cases these data are readily available and, in some cases, are the only ones available. In other cases, rates of service utilization are also used; however, such data are not generally available, and when they are available they are typically confined to physician utilization. Furthermore, current approaches to HHRP tend to be intermittent and based on faulty assumptions; most are also one-timeonly estimates focusing on single disciplines and often violating even the most basic measurement principles. These approaches have led to repeated cycles of shortage and surplus of providers. Simplistic approaches that ignore the complexity of the health care system have avoided consideration of why we provide health services and the impact of these services. It is critical that we improve our ability to plan HHR requirements (Birch et al. 2002). Birch has provided the standard delineation of HHRP methods, as well as their related questions and assumptions. His analysis isolates three distinct methods: Utilization-based (how many health professionals are required to serve the population in the same way as the current population?); needs-basedd (how many health professionals are required to meet the needs of the population?); and effective-demand-baseded (how many health professionals are required to support society's commit-

How Do Human Resources Policies Inhibit Change? 163

ment to health care?) (Lavis and Birch 1997; O'Brien-Pallas, Baumann, et al. 2000). Many approaches have been developed and tested to estimate the human resources required. Birch emphasizes that HHRP has largely been an exercise in demography based on implicit assumptions that population structure determines the service needs of the population, and that the age of providers determines the quantity of care provided (Birch et al. 2002)). The main limitation of such approaches is their failure to reflect the complex nature of the processes underlying the needs for services (population health) and the delivery of services (health care provision), as well as the effects of HHRP on population, provider, and system outcomes (Birch et al. 2002). Several reports have highlighted the limitations of these approaches, and recommendations have been made for improving the related processes (Lomas, Stoddart, and Barer 1985; Barer and Stoddart 1991; Birch et al. 2002). A focus on shortages and surpluses has not assisted in the determination of the efficiency of HHR levels. In fact, shortages may actually reflect problems of management such as work- and caredelivery models. Changing roles and demographics of a younger generation workforce seem to influence work participation. The changing work commitments among younger male and female physicians are thought by many physicians to be one of the root causes of the current shortage (e.g., choosing to work forty-hour instead of seventy-hour weeks) (Standing Senate Committee 2002). Over the last decade, major changes have occurred in the delivery of health care services. For example, there has been a major shift in the balance of care away from inpatient and institutional settings towards outpatient and community and home settings. Because these different approaches to service delivery involve different levels and mixes of resources, the changing balance of care will have implications on HHR requirements (Birch et al. (2002). Birch et al. (2002) argue that HHR requirements should be considered along with the relationship between the levels and mix of resources used to produce health care services and the quantity and quality of services produced. The efficient and effective mix of human resources is determined by the interaction of the several elements of the HHR Conceptual Framework. This provides an evidence base for exploring the implications of health care policies for planning future HHR. This includes policies associated with education, certification, licensure, and scope of practice. HHRP needs to be placed within the broader system in which health

164 Gail Tomblin Murphy and Linda O'Brien-Pallas

care services are provided, and it must take into account the impact of several important related factors. These include social, political, geographical, technological, and economic factors, and how these influence the efficient and effective mix of both human and non-human resources. Political and geographical factors cannot be overlooked in responsible HHRP. The issue of political will is an important one. The human-resource challenges we are facing today have evolved slowly over the last fifty years. Past mistakes cannot be overcome within the time-frame of a single or even a second political mandate. There needs to be sustained HHRP efforts by policy makers and key stakeholders, a challenge made all the more difficult by changing governments and political agendas. Policy makers and researchers must work in concert to keep the health policy issues relevant, easily understood, and practical. Geographical factors (urban and rural) must be considered in HHRP. With the appropriate data availability, the HHR Conceptual Framework can be applied to both urban and rural areas. However, ecological (i.e., transferring results from aggregates to individuals) and atomistic fallacy (i.e., researching exclusively at the individual level, thus failing to account for the context in which the individual action occurs) must be considered in understanding and planning for the geographical variations in needs. The issues of recruitment and retention of health providers are unique in rural Canada. HHRP cannot be performed effectively in isolation of broader policy processes (O'Brien-Pallas, Birch, and Murphy 2001; CIHI2001). In nursing, for example, care requirements associated with decreased lengths of stays have been the result of new care-delivery methods such as same-day procedures. This change in technology involves changes in the nurses' role in the delivery of care, which often includes an increase in work intensity. According to Birch et al. (2002), while planners can react to this by allocating fewer nurses to work with a patient population that we assume will pass through the hospital more quickly, this form of planning will perpetuate nursing shortages, if adequate consideration is not given to the concomitant increase in the level of work intensity. As a case in point, the decision made in the 1990s to reduce the number of hospital beds in Ontario had major implications for human-resource requirements. In the absence of proportional reductions in the number of episodes of all levels of severity, additional nursing inputs per episode were needed to support the technological innovations that were required to reduce hospital lengths of stay (Birch et al. 2002).

How Do Human Resources Policies Inhibit Change? 165 New Directions

The approach we are advocating begins by examining the dynamic nature of population health needs. Population characteristics related to health levels and risk reflect the varied characteristics of individuals in the population that create the demand for curative and preventative health services. Consideration must be given to people's responses to their environment, the economy, and the accessibility and quality of their health care system. Outcomes must drive the HHRP process. The outcomes of interest are related to the system, providers, and the health of Canadians (O'Brien-Pallas et al. 2002). HHR has been supply-driven, with little attention paid to demand factors and needs of the population. Demand factors are varied and may have on impact on the future demand for health resources. A comprehensive HHR model needs to address as many of these factors as possible. These demand factors may include population demographics, innovations or technology, availability of treatment and options, patient attributes, waiting lists, access to services, service utilization, incidence of disease (medical diagnosis) and health professionals' work patterns. At the most general level, demand factors interact with the supply of health personnel to create future requirements. The concept of 'demand' is a critical one in HHRP. There is, however, little consistency in the meaning and use of related terminology. Demand factors are really utilization factors. Research, policy, and practice in HHRP will need to go beyond demography to avoid endlessly replicating the cycles of boom and bust in the availability of human resources (Birch et al. 2002). HHRP also needs to be integrated. Integrated human-resource planning involves determining the number, mix, and distribution of health providers that will be required to meet population health needs at an identified future point in time. This requires integrating HHRP with service planning. HHRP should occur over a longer planning horizon than has been the practice to date. Short-term planning is aimed at ensuring that resources for health are allocated and managed in an efficient and effective manner. Service-planning activities are concerned with how many and what type of health resources will be allocated among different sectors and among human and physical capital such as technology, drugs, human resources, and infrastructure. The same basic principles that underpin good HHRP practice also underpin good service planning. If done properly, service planning and HHRP activities

166 Gail Tomblin Murphy and Linda O'Brien-Pallas

should be mutually supportive. Both should be seen as part of a continuous quality-improvement process which is updated at least biannually and where each activity informs the other. Both sets of activities should be based on evidence of best practice and current research. Where available, labour-market analysis is a useful tool for understanding the shortfalls of previous planning decisions, gaining insight into the current HHRP and planning context, and providing clues for future corrective action to be taken along the planning horizons (O'Brien-Pallas 2002). It is important, however, to emphasize that data and the management of data is not enough. A vision of the health care system is needed to guide and frame practices and research. This vision must be sensitive to the multiplicity of factors outlined in the HHR Conceptual Framework and provide simulation modelling opportunities to test new health care delivery strategies before implementation (e.g., studying the impact of program management on health, system, and provider outcomes prior to implementation). Because HHRP must determine how many health professionals are needed to meet population health needs, it must examine questions such as substitution and skill level within and between professional groups. In addition, the WHO has suggested that the relative price of different skill categories should be used to help guide decisions about the most efficient mix of workers in jurisdictions where labour markets are functioning. As suggested in response to Question 1, consideration of the skill level of providers must be made judiciously and based on the evidence from ongoing research. For example, some jurisdictions have used the ratio of other providers to the number of physicians in determining future requirements for integrated workforces. However, as health restructuring moves out of the hospital or clinical setting to non- traditional service venues, the number of other personnel might need to be unbundled from the ratio of physicians to ensure that full scopes of practice as defined in legislation work to the advantage of the population. Failure to do so could create inefficient substitutions which would lead to wasteful duplication of services. Focusing on the number of physicians as the starting point obfuscates the contribution of all other health team members and leaves us functioning in a physician model of health care. Furthermore, there is a direct correlation between the ratio of nurses to patients and the health outcomes of the patients (Canadian Nursing Advisory Committee 2002). Higher ratios of RNs have been linked to increased patient satisfaction, enhanced quality of life, fewer in-hospital complications, shorter hospital stays,

How Do Human Resources Policies Inhibit Change? 167

reduced costs (California Nurse 1999), and lower rates of urinary-tract infections, upper gastrointestinal bleeding, pneumonia, shock, and cardiac arrest (Needleman et al. 2002). Higher proportions of RNs and Licensed Practical Nurses (LPNs) in the staff mix are associated with better patient satisfaction and health outcomes, including lower rates of medication errors and wound infections (McGillis Hall et al. 2001). However, we do not advocate standardized nurse-to-patient ratios until further research is available to inform this process. A report on the roles and utilization of LPNs and care aides (2000) confirms that British Columbia has the lowest LPN-to-RN ratio in Canada and that LPNs are not utilized to the full scope of their competencies. Other factors identified for the low utilization of LPNs were jurisdictional conflict with RNs and the inability to recruit qualified LPNs. Outside nursing, there is little evidence regarding the impact of province-to-patient ratios and staff mix on outcomes. To move us into the twenty-first century, we need to make a concerted effort to move away from old and safe approaches and embrace conceptual and analytical complexity. This includes a focus on outcomes and integrated planning in order to provide efficient and effective health services for future generations. Certification, Licensure, and Scope of Practice Although breaking down inter-professional borders and allowing for greater overlap in scopes of practice are likely to enhance equity in the division of health care labour, create the conditions necessary for the exploration of more flexible/efficient delivery models, and contribute positively to patient outcomes, decision makers need to be cautious in selecting the appropriate mechanisms and processes for working towards this goal. Historically, professional spheres of responsibility and influence have been closely guarded and vehemently defended. Certification, licensure, and self-regulation have been the traditional weapons with which professional territories have been defended and new territories captured. Although doctors are arguably the most sophisticated and successful defenders of their territory, other groups, such as nurses, have also used legal and political mechanisms, ostensibly designed to protect the public, as instruments of self-interest by which to gain professional advantage. Calls for self-regulation by other occupational groups and the move from diploma training to baccalau-

168 Gail Tomblin Murphy and Linda O'Brien-Pallas

reate entry-to-practice requirements are interpreted by many as simply a status-enhancing exercise unlikely to contribute positively to health outcomes. Employers, however, are increasingly demanding greater employee preparation in areas such as communication skills, workplace management, coping skills, and general knowledge of the health system. These topics are not components of traditional diploma programs but are visible in university-based delivery models. However, as suggested in Question 1, since public protection is the mandate of licensing/professional bodies, careful consideration needs to be given to where these points of view both converge and diverge from those of the labour unions. Furthermore, it is doubtful that public support could be garnered for initiatives that would usurp the authority and scope of existing professional bodies, which are, with good reason, perceived to be performing well in their public-protection mandate. Attention does need to be directed towards issues of recruitment and retention and the role of work environments in the production function. Redesigning Current Work Environments We need to redesign current work environments for all health professionals in order to make them valued places to work, where each professional works to his/her fullest scope of practice. Work environments vary across jurisdictions and especially between rural and urban Canada. There are numerous reports that outline the problem, but few set directions for the future. In July 2002 the report of the Canadian Nursing Advisory Committee was released to the deputy ministers of health. The recommendations address the main issues identified as barriers to a quality workplace for Canadian nurses, focusing on, among other things, the need to increase the number of nurses, the need to improve education and maximize the scope of practice of nurses, and the need to improve the working conditions of nurses. Furthermore, factors influencing nursing job satisfaction and retention are highlighted with a clear message that 'the more important variable ... is the political will (at all levels) to implement the changes and strategies we know are central to creating high quality workplaces (Canadian Nursing Advisory Committee 2002, 28). This report not only identifies evidence-based strategies but also is currently determining the costs associated with implementation of these strategies across Canada. While focused on nursing, it is likely that the report will also inform

How Do Human Resources Policies Inhibit Change? 169

decision making in regard to other health professions. Given that more than half of nurses and some other providers in this country will be eligible to retire in the next five to ten years, time is of the essence. In a report prepared by the Canadian Medical Association (CMA) in 2002, physician-workforce issues are identified as being of paramount importance. The CMA has been involved in a number of initiatives related to the physician workforce, and the result has been a series of formal policies addressing physician-resource planning, rural and remote practice issues, physician health and well-being, and physician compensation. Similar to nurses and physicians, allied health care providers report difficulty in recruiting students, aging workforces, and high levels of fatigue and burnout. In addition, they are predicting nation-wide shortages in the next five to fifteen years (Standing Senate Committee 2002). According to the WHO (2001), integrated, multidisciplinary, collaborative, and cooperative models are required for workforce planning. Governments are establishing action plans to deal with workforce issues. Common features include better utilization and distribution of health professionals, increased cooperation between providers, enhanced education for health professionals, and the development of guidelines for safe staffing. HHRP processes need to focus not only on how many and what type of providers we need but also on the incentives required to keep our current supply in the workforce for as long as possible and to ensure that each practitioner is able to make his or her full contribution (O'Brien-Pallas 2002). Question 3 What are the assumptions associated with different approaches to HHRP and what are their related strengths and limitations? What is the best approach to ensure that we develop a plan with meaningful numbers, types, and mix of professionals to best meet the needs of the population?

Canada and other OECD countries are experiencing shortages of nurses, doctors, and other professionals in the health sector. In this contexts, policy makers, health care leaders, and researchers are interested in developing a meaningful, long-term approach to HHRP. This will require collaborative efforts to develop effective mechanisms and policies for establishing, monitoring, and predicting professional requirements to meet the health needs of the population (Sechrist,

170 Gail Tomblin Murphy and Linda O'Brien-Pallas

Lewis, and Rutledge 1999). It is critical that resource allocations be based on the best possible evidence. HHRP models attempt to provide such evidence in the context of determining requirements for health care workers by approximating future planning requirements based on a variety of factors specific to the model being employed. Regardless of which model is used, however, sound methods of study design and data analysis must be used if the evidence is to be meaningful. To complicate matters, HHRP research is a nascent discipline for which a firm scientific basis has not yet been fully worked out and tested. Methods for predicting human-resource requirements are few in number and those available for use continue to be plagued with methodological and conceptual difficulties (O'Brien-Pallas, Birch, Baumann, et al. 2001). In order to move beyond current practice, it is important at the outset to gain some familiarity with the three available approaches to HHRP (utilization-based, needs-based, and effective-demand based) and their related assumptions, questions, and methods. Proceeding in this fashion will help make clear that the needs-based approach we are advocating has evolved from the experience gained working with utilizationbased frameworks. The Table 5.1 presents the basic assumptions, questions, and methods of each approach in a way that is designed to highlight the evolution towards a needs-based approach. It is a supplement to the text descriptions that follow. It must be stressed that factors such as supply and utilization rates are important pieces of the HHRP puzzle; therefore, these data play an important role in the overall structure of a needs-based model. The intent is not to jettison what is valuable in current practice, but rather to build an approach that better reflects the complexity and dynamic nature of the health care system. In utilization-based models, the quantity, mix, and population distribution of current health care resources are adopted as a baseline for estimates of future requirements. As shifts are predicted in the basic demographic characteristics of the population (i.e., age and sex), these are compared to current baseline factors in order to derive predictions about future requirements. This approach, however, is limited by the fact that utilization rates are dramatically affected by factors other than the population characteristics typically included in utilization-based models (Lavis and Birch 1997; O'Brien-Pallas, Baumann, et al. 2000). As the name suggests, needs-based planning models attempt to approximate future human-resource requirements based on an empirical assessment of the levels of health need in the relevant population. Unlike utilization-based approaches, however, needs-based planning

Table 5.1 Approaches to health human resource planning2 Approach

Assumptions

Question

Utilization-based

Current level, mix, distribution of services is appropriate.

How many health professionals are required to serve the future population in the same way as the current population?

Appropriate level, mix, and distribution remain constant. Estimated future demographic profile is accurate. Needs-based

(All) health needs can and should be met.

Cost-effective methods can be identified and implemented.

How many health professionals are required to meet the health care needs of the population?

Unmet needs are caused (only) by inadequate supply.

Method Uses population-based utilization rates as a baseline. Applies rates to demographic profile of future population.

Population-based rates of health problems replace population-based use of services. Identifies human-resource requirements for solving population's health problems. Applies rates to demographic profile of future population.

Non-needs/non-cost-effective use of resources can be eliminated. Effective-demand-based

Health needs are only one of a set of social priorities. Definitions of need are less than precise. There are clear possibilities for resource trade-offs.

How many health professionals are required to support society's commitment to health care?

Estimates the size of the economy supporting health care. Estimates the proportion of the economy devoted to health care. Estimates the proportion of health care expenditures allocated to health services. Estimates the number of health professionals that could be employed using these resources. Provides a fiscal-resource context for needs- or utilizationbased methods.

Source: (Tomblin Murphy 2002 54-5; Birch 2001). The table presents a general overview of the three methods. Alternative approaches address different questions that arise from the basic assumptions of each model and that ultimately determine the methods. Each approach is informed by experience gained with the previous one.

a

172 Gail Tomblin Murphy and Linda O'Brien-Pallas

begins from the premise that the current distribution of health professionals and health services is not necessarily optimal for addressing population health needs and that human resources must be redistributed if health needs are to be met (Markham and Birch 1997; Roos, et al. 1995; Evans and Mustard 1995). According to Eyles, Birch, and Newbold (1995), there are two broad approaches to needs-based planning: the absolute-needs approach and the relative-needs approach. The absolute-needs approach attempts to quantify the health care resources required to produce a target level of health status in the general population. The relative-needs approach considers the variation in health status among communities together with available resources in order to meet the needs of communities. The downside of the former approach is the risk that resources will be allocated based on the demand for services (rather than actual needs), thus placing too great a claim on social resources. It is primarily because of this concern that the relative-needs approach is preferable. Data obtained through the relative-needs approach can better serve to plan for the distribution of the health workforce in such a way that the overall cost of increasing supply is explicitly acknowledged. HHRP must therefore consider the relative needs of communities rather than simply the demand for services as expressed as a function of the utilization of services. Otherwise, the use of resources will in fact be based on demands rather than needs, with the consequence that HHRP will hopelessly continue endeavouring to provide a forever escalating number of workers. Both approaches, however, require an assessment of the health needs of populations and variations among communities in order to devise effective resource allocations (Tomblin Murphy 2002). It is critical that, in a needs-based approach, geographic variation in needs is considered in HHRP. For instance, the health needs of Canadians living in rural areas are different from those living in urban centres. The CMA (1999) surveyed, 7000 health care professionals, including 2000 rural physicians, pharmacists and a range of nursing professionals (RNs, LPNs, and nurse practitioners) who practise in rural or remote areas of Canada. Findings from the survey inform the national 'framework of rurality' regarding a variety of factors, including the most important characteristics in defining rural practice, the characteristics of health care communities in rural Canada, and nurses', pharmacists', and physicians' satisfaction levels with their personal and professional lives in rural and remote communities. The type of needs-based model we are advocating treats human

How Do Human Resources Policies Inhibit Change? 173

resource requirements as independent of current service utilization, while at the same time recognizing they are interdependent with the requirements for other human resources and various other factors. Moreover, to the degree that current needs are not all met, unmet needs will be included in the estimation process. Thus, the main virtue of this approach lies in its ability to avoid perpetuating existing inequities and inefficiencies in the deployment of services (O'Brien-Pallas 2002). Effective-demand-based approaches to HHRP attempt to incorporate economic considerations into the principles of the needs-based approach. The starting point is to estimate the future size of the economy from which the particular human resource and competing services will be funded. This allows an estimation of the proportion of total resources that might be allocated to health care and the estimated share that will be devoted to health human resources in health. Notwithstanding the welcome inclusion of general economic considerations brought to the mix by effective-demand-based approaches, it remains clear that information about the level and distribution of needs in the population, and about the role that human resources play in meeting those needs, will continue to be at the heart of any model of HHRP. Furthermore, measuring population health needs will help identify when these needs can reasonably be satisfied through the distribution of health care resources and when they should be addressed through various other social mechanisms. Although the health care system may not have the resources to affect directly all the determinants of health, it is nonetheless important to understand that the health care system itself is just one of the determinants of health and that it is often left to deal with the consequences of more general social failures to address these needs. It is therefore imperative that the measurement and understanding of population health needs be an integral part of HHRP. In contrast to traditional supply-driven approaches, models based on population health needs provide more meaningful information about how to deploy human resources effectively. Consequently, they generate opportunity for more relevant decisions about legislation, policy development, and the production and utilization of health care providers. To build a workforce designed to meet the health needs of our community, it must be based on comprehensive information that promotes sufficient numbers of relevant health providers who can work in ways that maximize health outcomes. To be optimally effective, needs-based models of HHRP must rely on data from varied

174 Gail Tomblin Murphy and Linda O'Brien-Pallas

sources. From health care consumers, data are needed about the changing demographics of society, changing health and illness patterns, changing determinants of health, and changing behaviour of consumers. From the health system, it is essential to have knowledge about factors such as those that influence the effectiveness and efficiency of the organization, management and delivery of services, and the capacity to achieve enhanced health outcomes. From governments, data are needed about the known/anticipated effects of existing or proposed policies. From universities, it is critical to have knowledge pertaining to factors that influence the effectiveness, efficiency, productivity, and relevance of educational programs. Given the massive restructuring under way as part of health reform, other types of information are needed as well. For example, data regarding shifts from institutional to community-based care and the integration of community services, primary health care, and institutional care must be monitored. It is essential to track changes in governance (e.g., devolution to regional health boards, mergers, etc.), methods of funding services and reimbursing professional services, cost-shifting, and changes in burden sharing. It is also critical to follow trends in the redefinition and reconfiguration of professional roles. Tracking the effects of these reforms, however, will require a significant shift away from historical database practices. As the WHO (1978, 2000) argues, now is the time to embark on HHRP using an approach that is policy-relevant, integrated, health-needs-based, consumer-focused, goal-and-capacity driven, adequately resourced, evidence-based, accountable, and reflective of the unique and shared competencies of all stakeholders. To begin this enterprise, if we are to build a truly effective approach to HHRP in Canada, it is essential that key stakeholders join together in identifying common research needs, existing resources for meeting those needs, and available opportunities and obstacles. High-Quality Data Acquisition To accomplish this vision, we need good-quality data that is comparable, comprehensive, and available to researchers in a timely fashion. There is a pressing need for mechanisms that will enable meaningful and efficient partnerships to develop between researchers and data-gathering/holding facilities (e.g., Statistics Canada). To ensure system efficiency and effectiveness, HHRP activities should be both needs-based and outcome-directed. Furthermore, planning at all levels requires

How Do Human Resources Policies Inhibit Change? 175

good-quality data. In describing the approaches to modelling or service planning, we make the assumption that the data that form the basis of resource planning are currently available and of good quality; that is, they are consistently reported (reliable) and they actually measure the key variables to estimate human-resource requirements (valid). This is an assumption that, regrettably, does not reflect current reality (O'BrienPallas 2002; O'Brien-Pallas and Tomblin Murphy 2002). Needs-based approaches, in which resource requirements are based on estimated populations' health needs, create greater data demands than those required for supply/utilization-based planning. The requirement to link to outcomes will initially create greater data challenges. To plan services and/or to model human-resources requirements without highquality data will lead only to unreliable estimates of future human resource needs and erroneous service-planning models. Planning should be conducted when planners are confident that the data that underpins the estimates are of good quality (Tomblin Murphy 2002). Modelling and Simulation Our conceptual model is a dynamic system-based framework which considers population characteristics related to health levels and risks (needs-based factors); service-utilization patterns, nurse-deployment patterns, and the deployment patterns of others who provide similar or identical services (utilization-based); economic, social, contextual, and political factors that can influence health spending (effective-demandbased); and health status and system outcomes resulting from the different types of nurse- and other health-provider utilization. This model incorporates each of the three methodological approaches outlined earlier but places these approaches in the context of an assessment of needs and outcomes of service provision. Simulations of the health system are used to provide needs-based estimates that are aimed at optimizing outcomes. The practical applications of this model are currently being tested by members of our research team, which, as a result of our work on the framework, has expanded to include philosophers, political scientists, labour economists, demographers, geographers, epidemiologists, physicians, health economists, and psychologists. Simulation is a powerful technique. Hall (2000) suggests that it allows planners to explore consequences of alternative policies, facilitates input and output sensitivity analysis, and makes it easier to involve stakeholders throughout the process. Simulations are a means

176 Gail Tomblin Murphy and Linda O'Brien-Pallas

to assist planners to make decisions; they are not an end in themselves or a substitute for careful planning. The extent to which simulation provides useful scenarios for consideration is dependent upon the quality of the data used in the model and the extent to which the variables modelled reflect the system as a whole. Simulation probably offers the most useful tools for assessing substitution across and within professions and for addressing issues such as geographic distribution of health personnel. Whatever method used, O'Brien-Pallas, Baumann, Dormer, et al. (1998), Song and Rathwell (1994), and Eyles, Birch, and Newbold (1993) suggest that estimates for requirements will not be exact numbers but instead a range of numbers. Until further development of models occurs, sensitivity analysis will allow policy makers and planners to have different estimates of required resources from which to plan their service needs and HHRP. The importance of continuously updating estimates cannot be overstated. Networking, Knowledge Transfer, and Uptake We need to plan HHR in an integrated way. Planning for nurses, doctors, physiotherapists, pharmacists, and others must be done at the same time, each being driven by the health needs of Canadians. To do this, there needs to be a significant investment in creating and maintaining databases for planning: databases that allow us to compare differences between and across jurisdictions and that allow us to understand the needs and to determine whether the system is working in effective and efficient ways to meet those needs. The CIHI is the logical partner for development, storage, and upkeep of databases. The other partners are researchers and policy makers who are committed to needs-based and outcome-driven analysis that embrace the complexity of the health care system. We need to establish a standing committee on integrated HHR composed of consumers, expert researchers, and policy makers. This standing committee should be charged with the responsibility of creating next-generation approaches to forecast and plan human resources, conducting the actual analyses, and working with both management and policy makers in implementing the results of these activities. Furthermore, a virtual national and international network of HHR key stakeholders (consumers, policy makers, managers, researchers) must be developed to facilitate a more comprehensive understanding of HHR, to foster interdisciplinary mentorship, and to help build a critical mass of expertise in HHRP.

How Do Human Resources Policies Inhibit Change? 177 There is a clear need for increased knowledge transfer and uptake between the research and policy sectors. This can be facilitated through the development of partnerships among consumers, researchers, and policy makers that involve direct regular interaction at each step of the research, implementation, and assessment phases of HHRP. Building solid relationships and strengthening the links between both the research and policy communities will move HHRP policy-relevant research forward and enhance the use of findings in policy making. There needs to be effective and ongoing coordination of the interaction among government, research, and administrative stakeholders through advisory, research, and communication infrastructures. Serious consideration, therefore, should be given to the establishment of several standing HHRP research/policy institutes linked nationally through a permanent virtual institute. This, together with an enhanced commitment to ongoing HHRP research/policy infrastructure support, will help forge the links necessary to ensure optimal HHR planning for the future. HHRP has become a priority for Canadian researchers, policy makers, and decision makers. As social, economic, and technological developments propel health care into the information age, Internet based access to HHRP-related information is rapidly assuming greater significance. Convenient access to such information is important for current and future HHRP and will continue to be a priority as the area develops and responds to new challenges (Tomblin Murphy and Barrath 2002). Evidence for decision makers comes in a variety of forms. Policy institutes, research units, governments and their agencies, professional associations and unions, think-tanks, universities, and not-for-profit organizations release a variety of reports that are seldom integrated into conventional literature vehicles (such as journals or bibliographic databases). The Internet sites of these organizations frequently provide access to this unpublished or 'grey' literature. The grey literature must be accessible to Canadian consumers, policy makers, and researchers. It includes academic papers, scientific protocols, white papers, preprints, committee reports, proceedings, conference papers, research reports, standards, discussion papers, technical reports, dissertations, theses, government reports, journals, newsletters, working papers, essays, and electronic columns. Grey publications are valued as sources of comprehensive, concrete, and up-to-date information on research findings and investigations (System for Information on Grey Literature 2002). They provide valuable and often unique information that can rapidly inform specific scientific communities, funding bodies, and policy makers about the results of research projects and other initiatives.

178 Gail Tomblin Murphy and Linda O'Brien-Pallas As users gain confidence in searching and accessing Internet-based grey literature, the demand for more sophisticated and user-friendly sites will continue to grow. Technical innovations such as these promise to expand knowledge transfer and uptake in a way that will ultimately enhance research productivity and have a great impact on HHRP internationally.

NOTE The authors would like to acknowledge the Canadian Health Services Foundation, the Ontario Ministry of Health and Long Term Care, the Nova Scotia Health Research Foundation, and the large team of researchers and policy makers who continue to inform and graciously support our ongoing work in this complex area of inquiry. We would also like to thank the anonymous reviewers whose thoughtful suggestions helped strengthen the finished work. Moreover, we would like to acknowledge the guidance of the Commission staff for their support. Finally, we wish to acknowledge those special people behind the scenes who gave so willingly during the development of this paper.

REFERENCES

American Council of Medical Colleges. 2001. http://www.nlm.nih.gov/ siteindex.html Barer, M., and G. Stoddart. 1991. Toward integrated medical resource policies for Canada. Report to the federal/provincial/territorial conference of deputy ministers of health. Hamilton, Ont: McMaster University. Birch, S. September. 2001 Health human resource planning. Presented at the Nova Scotia Health Research Foundation Workshop, Halifax, N.S. Birch, S., L. O'Brien-Pallas, C. Alksnis, G. Tomblin Murphy, and D. Thomson, April 2002. Beyond demographic change in health human resources planning: An extended framework and application to nursing. Paper presented at the 5th International Conference on Strategic Issues in Health Care Management, St Andres, U.K. Brooten, D., S. Jumar, L. Brown, P. Butts, S. Finkler, S. Blackwell-Sach, A. Gibbons, and M. Papadopoudolis, 1989. A randomized clinical treatment of early hospital discharge and home follow-up of very low birth weight infants. New England Journal of Medicine 315(1): 934-9. Brown, G. 1994. Nurses make the difference in delivering quality cost-effective

How Do Human Resources Policies Inhibit Change? 179 health care. Paper presented at the Canadian Nurses Association, 1994 biennial convention, Winnipeg, Man. California Nurse. 1999. Studies link RN staffing to patient safety. California Nurse, 95(6): 7. Available at http://www.calnurse.org/cal/aug99/ 7cnaug99.html (18 September 2002). Canadian Institute for Health Information. 2001. Future development of information to support the management of nursing resources: Recommendations. Ottawa: Author. Canadian Medical Association. 1999. Rural community development tools from the medical perspective: A national framework of rurality and projections of physician workforce supply in rural and remote areas of Canada. Ottawa: Author. - 2002. Physician resource planning in Canada. Ottawa: Available at http:// www.cma.ca/cma/menu/displayMenu. do?skin=130&pMenu!d=2&pSubMenuId=6&pageId=/staticContent/HTML/NO/12/working_on/physicianworkforce.htm Canadian Nursing Advisory Committee. 2002. Our health, our future: Creating quality workplaces for Canadian nurses. Ottawa: Author. Cohen, E.L. 1991. Nursing case management: Does it pay? Journal of Nursing Administration 21(4): 20-5. Commission on the Future of the Health Care in Canada. 2002. Issue/survey paper: Health human resources in Canada's healthcare system. Ottawa: Canadian Health Services Research Foundation. Denton, F.T., A. Garni, B.C. Spencer, and G.L. Stoddard, 1983. Potential savings from the adoption of nurse practitioner technology in the Canadian health care system. Socio Economic Planning Sciences 17(4): 199-209. Doran, D., L. McGillis Hall, S. Sidani, L. O'Brien-Pallas, G. Dormer, G.R. Baker, and G. Pink. 2001. Nursing staff mix and patient outcome achievement: The mediating role of nurse communication. International Perspectives 1(2-3): 74-83. Ethridge, P. 1991. A nursing HMO: Carondelet St. Mary's experience. Nursing Management 22(7): 22-6. Ethridge, P. and G.S. Lamb. 1989. Professional nursing case management improves quality, access and costs. Nursing Management 20(3): 30-5. Evans, R., and J. Mustard. 1995. Medical Care 33(12): DS5. Eyles, J., S. Birch, and K.B. Newbold. 1993. Equity and health care: Analysis of the relationship between need for health care and the utilization of nursing services in Canada. Canadian Journal of Nursing Research 25(4): 27-45. - 1995. Delivering the goods? Access to family physician services in Canada: A comparison of 1985 and 1991. Journal of Health and Social Behavior 36(4): 322-32.

180 Gail Tomblin Murphy and Linda O'Brien-Pallas Green, W., and B. Simons-Morton. 1984. Introduction to health education. New York: Macmillan. Hall, T. 1993. Human resource for health: Models for projecting workforce supply and requirements. Geneva: World Health Organization. - WHO simulation models for intermediate-term health workforce planning. Tool kit for planning, training and management, 2000. Available at http:// projects.forumone.com/toolkit/models/. Horrocks, S., E. Anderson, and C. Salisbury. 2002. Systematic review of whether nurse practitioners working in primary care can provide equivalent care to doctors. BMJ 324(7341): 819-23. Hubert, J. 2002. Health human resource planning: Challenging the past, creating the future (project 1: The health care needs of the population). Discussion paper on ethical issues in the design and application of a needs-based health human resource planning model. Lamb, G., and D. Huggins. 1990. The professional nursing network, in G. Mayer, M.J. Madden, and E. Lowrenz, eds, Patient care delivery models. Rockville, Md: Aspen. Lavis, J.N., and S. Birch. 1997. The answer is ... now what was the question? Applying alternative approaches to estimating nurse requirements. Canadian Journal of Nursing Administration 10(1): 24-44. Licensed Practical Nurses and Care Aides in B.C.: Research on Roles and Utilization. 2000. Available at http://www.heu.org/members/lpncareaidefront.pdf. Lomas, J., G.L. Stoddart, and M.L. Barer. 1985. Supply projections as planning: A critical review of forecasting net physician requirements in Canada. Social Science & Medicine 20(4): 411-24. McGillis Hall, L., et al. 2001. A study of the impact of nursing staff models and organizational change strategies on patient, system, and nurse outcomes. Toronto: Faculty of Nursing, University of Toronto and Canadian Health Services Research Foundation/Ontario Council of Teaching Hospitals. McKenzie, C.B., N.G. Torkelson, and M.A. Holt. 1989. Care and cost: Nursing case management improves both. Journal of Nursing Administration 20(10): 30-4. Marchette, L., and F. Holloman. 1986. Length of stay. Significant variables. Journal of Nursing Administration 16(3): 12-9. Markham, B., and S. Birch. 1997. Back to the future: A framework for estimating health-care human resource requirements. Canadian Journal of Nursing Administration 10(1): 7-23. Naylor, M. 1990. A comprehensive discharge planning for hospitalized elderly: A pilot study. Nursing Research 30(1): 1565-611.

How Do Human Resources Policies Inhibit Change? 181 Needleman,}., et al. 2002. Nurse-staffing levels and the quality of care in hospitals. New England Journal of Medicine 346(22): 1715-22. O'Brien-Pallas, L. 2002a. Presentation to the Romanow Commission: Health human resources in Canada. 31 May 2002 Colony Hotel, Toronto. - 2002b. Where to from here? (Editorial and discourse). Canadian Journal of Nursing Research 33(4): 3-14. O'Brien-Pallas, L., A. Baumann, S. Birch, and G. Tomblin Murphy, 2000. Health human resource planning in home care: How to approach it - that is the question. Healthcare Papers 1(4): 53-9. O'Brien-Pallas, L., A. Baumann, G. Donner, J. Lochhass Gerlach, M. Luba, L. Lakats, Y. Armasi, and C. Mallette. 1998. Health human resources: An analysis of forecasting models. Ottawa: Canadian Nurses Association. O'Brien-Pallas, L., S. Birch, A. Baumann, and G. Tomblin Murphy, 2001. Integrating workforce planning, human resources, and service planning. Workshop on global health workforce strategy (Annecy, France, 9-12 December 2000). Geneva: World Health Organization. O'Brien-Pallas, L., S. Birch, and G.T. Murphy. 2001. Workforce planning and workplace management. International Nursing Perspectives 1(2-3): 55-65. O'Brien-Pallas, L., D.I. Doran, M. Murray, R. Cockerill, S. Sidani, B. LaurieShaw, and J. Lochhaas-Gerlach. 2001. Evaluation of a client care delivery model, part 1: Variability in nursing utilization in community home nursing. Nursing Economic$ 19(6): 267-76. O'Brien-Pallas, L., D.I. Doran, M. Murray, R. Cockerill, S. Sidani, B. LaurieShaw, and J. Lochhaas-Gerlach. 2002. Evaluation of a client care delivery model, part 2: Variability in client outcomes in community home nursing. Nursing Economic$ 20(1): 13-36. O'Brien-Pallas, L., and G. Tomblin Murphy. September 2002. The case for increasing the numbers of healthcare workers. Presented to the Romanow Commission. Saint John's, Nlfd. O'Brien-Pallas, L., G. Tomblin Murphy, A. Baumann, and S. Birch. 2001. Framework for analyzing health human resources. In Canadian Institute for Health Information: Future development of information to support the management of nursing resources: Recommendations. Ottawa: CIHI. Pollara. 14 January, 2000. Pollara Public Trust Index. Available at http:// www.pollara.ca/new/POLLARA_NET.html (6 June 2002). Prescott, P. A. 1993. Nursing: An important component of hospital survival under a reformed health care system. Nursing Economic$ 11(4): 192-9. Roos, N.P., C.D. Black, N. Frohlich, C. Decoster, M.M. Cohen, DJ. Tataryn, C.A. Mustard, F. Toll, K.C. Carriere, C.A. Burchill, L. MacWilliam, and B.

182 Gail Tomblin Murphy and Linda O'Brien-Pallas Bogdanovic. 1995. A population-based health information system. Medical Care 33(12 suppl). DS13-20. Schultz, P., F. McGlone, E. Kinderknecht, L. Morton, S. Eylar, S. Monley. 1977. Primary health care to the elderly: An evaluation of two health manpower patterns. Final Report. Denver: Medical Care Research Foundation. Sechrist, K.R., E.M. Lewis, and D.N. Rutledge. 1999. Data collection for nursing work force strategic planning in California. Journal of Nursing Administration 29(6): 9-11,29. Shamian, ]., and B. Chalmers. 1996. Nurse Effectiveness: Health and cost effective nursing services. Geneva: World Health Organization. Song, F., and T. Rathwell. 1994. Stochastic simulation and sensitivity analysis: Estimating future demand for health resources in China. World Health Statistics Quarterly 47:149-56. Standing Senate Committee on Social Affairs, Science and Technology. 2002. The health of Canadians - the federal role. Interim report. Volume two - current trends and future challenges. Available at http://www.parl.gc.ca/37/ l/parlbus/commbus/senate/com-e/SOCI-E/rep-e/repjan01vol2-e.htm (18 September 2002). System for Information on Grey Literature. 2002. The role of grey literature. The European Association for Grey Literature Exploitation. Available at http://www.kb.nl/infolev/eagle/frames.htm (4 January 2002). Tomblin Murphy, G. 2002. Methodological issues in health human resource planning: Cataloguing assumptions and controlling for variables in needsbased modelling. Canadian Journal of Nursing Research 33(4): 51-70. Tomblin Murphy, G., and D. Barrath. 2002. A survey of web-based health human resource planning activities in Canada. Canadian Journal of Nursing Research 33(4): 123-37. World Health Organization. 1978. Report of the international conference on primary health care, Alma-Ata, USSR. Geneva: World Health Organization. - 2000. The world health report 2000. Health Systems: Improving performance. Geneva: World Health Organization.

6 Changes and a Few Paradoxes: Some Thoughts on Health System Personnel CLEMENCE DALLAIRE AND SONIA NORMAND

In recent years, a succession of changes in the health system have had a significant impact on personnel.1 Yet it seems that neither new investments nor new technologies nor administrative controls have resulted in real progress in making the system more effective and efficient, especially with regard to the public's health. Health system reforms should have provided an opportunity to identify the place and role of personnel and introduce changes to it. Infact, however, the changes implemented seem to have considered the financial dimension (i.e., the escalation of health system budgets and the ways to counter it) more than the primary mission (i.e., services provided, direct care, and the personnel needed to fulfil this mission). Although it is true that personnel represents a significant expense in the health system, it is nevertheless an essential element. The Commission on the Future of Health Care in Canada has examined the way in which the nature and quality of personnel-management policies and practices have helped or hindered change in the health system by asking one general question and three more specific questions. Several recent reports from specialists in the field of health system personnel have already informed the commission on the current situation, the changes implemented, and the main issues involved. Our goal in this essay is to answer the commission's questions about changes in the professional workforce by referring to a selected literature to support a personal vision based on years of practice, teaching, research, and work in nursing and by using paradoxes in order to guide the analysis on personnel-management policies and practices. The evidence reveals several findings and paradoxes and it may seem point-

184 Clemence Dallaire and Sonia Normand

less to mention them. However, a careful examination shows that they appear to play a critical role in the present situation. This approach will allow us to convey what seems essential and, it is hoped, contribute to further reflection on the topic.

General Question To what extent was the desired change helped or hindered by the nature and quality of human-resources management policies and practices in health, for example, education, laws and regulations, certification, licensure, accountability, etc. 1

The answer to this question depends on the prior definition of the term 'change/ which can have several meanings. Above all, a change is a break, an adjustment, and a process; it is not a goal. Should the various alternatives proposed for the current system be described as changes? We have a medical care and hospitalization system, and this has not changed. We have changed the way of doing things and the roles of professionals and interventions, without calling into question the curative orientation of the system. Yet there is a lot of evidence regarding the limitations of such a system in improving the public's health and, consequently, there are valid reasons for calling the vision of the system into question. Or should the term change perhaps be reserved for a fundamental questioning of the Canadian health system and the means it is using to improve health, such as those found in Canadian law (i.e., calling into question the medical care and hospitalization system)? This study prefers the second alternative, where the desired change first addresses the vision of the health system without challenging the principles underlying the Canadian system, in particular, universality. It is our belief that, before examining the changes in personnel, a re-examination of the vision of the health system and its goals is in order, taking into consideration the means used to achieve those goals, for which the contribution of personnel is essential. This choice implies that, generally, minor modifications within the current system will not be described as changes in this brief study. Paradox 1: Need for a Vision of the Health System, but the Difficulty of Proposing One Calling the current system into question is particularly difficult to do given the number of actors and competing goals that often succeed in

Changes and a Few Paradoxes 185 focusing attention on their own interests. The goal of a health system should or must be better health, meeting the expectations of the public, and fair contribution to the financing of the health system (Biscoe 2001). These goals must resolve the conflicts between several objectives, and that is why it is difficult to propose a vision. Among the conflicts, some pertain to society, such as the fair distribution of health care regardless of socio-economic status; others have to do with professionals who wish to preserve clinical autonomy by organizing care in their own way and by receiving compensation that reflects the perception of their own value; and still another involves conflict within the system between the exercise of economic and budgetary control over health costs and the ability of government to finance the system for years to come (Biscoe 2001). Moreover, these conflicts are determined by social decisions about the level of resources allocated to health, the organization of services, and the financing of health and service programs, as well as the quantity and type of care provided (O'BrienPallas, Birch, Baumann, and Murphy 2001). Such a situation should prompt us to examine carefully the ideal perspective in this quest for financial balance and fiscal responsibility. Inevitably, choosing among priorities implies that other objectives are considered less important. The political nature of health organizations, where informal alliances and negotiations over various objectives allow for compromises, may make politicians hesitant to reveal the trade-offs reached (Bach, 2001) in order to sustain the operational system. These elements help to perpetuate the impression of the system's inertia in the midst of the turmoil created by a rapid succession of narrowly focused changes in the short term. At present, our health system is a medical care and hospitalization system, which already represents a form of compromise. This system favours medical practice as a means to achieve better health for the public, and conflicts are thus channelled into a medical-practice subdivision exercise. For example, the proposal to allow nurses to replace doctors, which would be less costly in many cases, would mean dividing up tasks. In such a system, the pursuit of financial balance and fiscal responsibility - a source of conflict between the health system and society - could be addressed by controlling costs. However, the unsuccessful efforts of recent years in the area of financing should prompt us to re-examine the vision at the heart of the current Canadian system and the conflict between the interests of that system and better health for the public. Furthermore, the current financial situation and the rise in diseases associated with lifestyle further underline the conflicting

186 Clemence Dallaire and Sonia Normand

viewpoints and call for a careful examination of the curative care that medicine provides. Expensive technology that allows more precise diagnoses is not a treatment; thus, it is difficult to improve health in this manner (Fournier 1999), although the area may be of interest. Investing in a precise diagnosis without available treatment in cases where the problem is incurable perhaps does not constitute a priority investment, except in the area of research and for some of the actors involved. Yet this type of expenses will probably not attract attention in the curative context, even though an effort is under way to control costs. A variety of visions and suggestions to redirect the health system have emerged in recent years. As many have suggested, a major redefinition of the health system could mean a larger primary care system associated with a reduced hospital system, which would be more efficient and better equipped (Smith and Preker 2001). The report from Alberta, A Framework for Reform, suggests concrete ways to direct the Alberta health system towards health and away from illness. Many studies show the limitations of curative interventions and their sometimes exorbitant costs and limited results in terms of improving the public's health. A change involving a different mix of curative, preventative, and promotion interventions provided by the system would undoubtedly have an impact on personnel. For example, attempting to achieve a balance between the two opposed visions of the health system, that is, offering curative services and at the same time investing more in the prevention of problems which our knowledge allows us to act on, seems difficult. A medical care and hospitalization system presents contradictory curative objectives and promotion/prevention objectives because different professionals usually strive towards these goals and compete with each other for resources. While it is true that society will not be in a position to finance the growth of the curative medical care and hospitalization system, the mission of the health system will have to be specified accordingly. To date, the health system has not succeeded in proposing a vision that reconciles the objectives and roles of professionals with those of society, as shown by the budgetary debate. The health system and society should have the maturity to debate priorities, make their decisions explicit, and extend the debate beyond financial and medical considerations, which seem to monopolize the majority of current discussions. This debate is crucial to any examination of change in personnel, since it means either several modifications in the fragmentation under medi-

Changes and a Few Paradoxes 187

cal control or a redefinition of the links between different professions and, for some, even running the risk of losing their autonomy and influence. Only then will it be possible to specify the nature of the work assigned to personnel and consider long-term changes, rather than carrying out adjustments, modifications, and the division of tasks within the current curative system. This aspect of the paradox underlines the difficulty of making decisions, mainly on account of conflicting viewpoints, especially if the disruptions caused are politically difficult. Paradox 2: Need for Change, but Low Priority Given to Change This paradox entails reflecting on the fact that there is a constant issue of change, but paradoxically, it seems to be given low priority, as demonstrated by the lack of concern for some of the factors that foster or thwart change. Among these factors, we mention three: the difficulty of talking about change in terms of financial gains and losses for the system, the monitoring of change over a sufficiently long period, and the resistance to changing structures. A major factor of the political process is the promotion of change that generates greater benefits than costs (Smith and Preker 2001), suggesting that in any change there are winners and losers (Bach 2001; Smith and Preker 2001). An important lesson learned from reform in several countries is that many influential actors believe, rightly or wrongly, that reform has a detrimental effect on their status, their working conditions, and their compensation (Bach 2001). Thus, resistance to change from personnel is probably based on a desire to avoid losses. An approach that exposes gains and losses has the advantage of not requiring the approval of everyone as a precondition. Employees are able to compromise and tolerate personal losses, especially if they find answers to the following questions: in the name of what, which principle, and what vision? Human beings are often thought of as seeking only to maximize their profits, whereas concrete examples and theoretical models suggest that motives are much more complex. At the same time, it is important to ensure that the costs of change to improve the system are not assumed by individuals (Smith and Preker 2001), particularly when this involves a career change, relocation, training, or losses in compensation or status. Rather than pretending that everyone wins, what is missing in the changes that have been carried out and those that

188 Clemence Dallaire and Sonia Normand

are planned is a recognition of the losses suffered and a justification of them, as well as compensation mechanisms for those losses. Incentives have to allow professionals to focus on service to individuals and the desired changes to undertake, with the understanding that their personal losses will be compensated. Such mechanisms already exist in collective agreements, but there is a tendency to view them from an administrative standpoint as constraints, thus preventing a fair understanding of their role in managing change. Moreover, change may produce savings, and efforts should be made to increase their visibility, which may be limited if these gains are spread among personnel. In short, giving priority to change involves highlighting savings and losses, using incentives to facilitate decisions when there are losses, evaluating existing mechanisms, compensating losses, and making savings visible for personnel. Another aspect that would draw attention to the priority given to change would be monitoring it for a sufficiently long period. In Great Britain, radical attempts to change personnel policies in the early 1990s failed because the minister of health did not provide concrete support and did not offer guidance to hospitals, which were supposed to implement personnel changes. This made the local managers wary, since they believed that they were receiving ambiguous messages about the priorities set by government for personnel reform (Bach 2001). General policies are thus not enough to specify the priority given to a particular change. This priority is better perceived if it is accompanied by implementation guidelines and, in particular, support to make the necessary adjustments, to experiment with unexpected effects, and to readjust accordingly. It should not be assumed that an initiated change will not raise problems until it is fully implemented. Furthermore, a rapid succession of changes makes their introduction difficult and does not reflect their importance. Thus, giving priority to a change means monitoring the progress of implementation until it has been stabilized and fully integrated into existing procedures before initiating another change. While everyone agrees that changes are necessary and can identify those that are being considered, there is an impression that a low priority is actually given to these changes. Some are even talking about the inertia of administrators and clinicians, referring to them as two 'silos' that operate with different information and make little effort to reconcile their viewpoints. The perception that a low priority is given to change may only be a false impression, resulting from the opposition of stakeholders who are in a position to pro-

Changes and a Few Paradoxes 189

mote the status quo (Smith and Preker 2001). In short, the importance of a change will be better perceived if it is carried out in conditions that ensure its implementation. Change and the priority given to it can be considered from another perspective, that is, the resistance of structures. In this category, unions and legislation can be given as examples of institutions created to reduce the scope of change (Smith and Preker 2001), as can the health system, which must maintain stability so individuals who need care and related services can rely on it. More specifically, any change in the organization threatens its stability and, for this reason, structural factors will cause resistance to change: 'Quantum change, that is, a complete reorientation, disrupts the existing organizational categories, on which planning depends. Consequently, a change of this nature tends to give rise to resistance or is often ignored' (Mintzberg 1994, 187) [Translation]. Integrating structural incentives to counter this organizational inertia would be one way to give priority to change. The paradox between a desired change and a perception of low priority affects the health system's capacity to initiate changes. Priority given to changes would draw attention to the winners and losers, would monitor change over a sufficiently long period, and would take into account structural resistance. Paradox 3: Need for Personnel, but Lack of Consideration of Personnel This third paradox reflects the fact that the health system depends on its personnel to fulfil its mission, while at the same time the system seems to consider personnel as a hindrance. Recent literature on this topic includes four key reports from Quebec (Clair 2001), Ontario (Sinclair 2000), Alberta (Mazankowski 2002) and Saskatchewan (Fyke 2001). The human resource perspective evident in these reports seems to be based on one key finding: there is a widespread climate of gloom and exhaustion within the health system, which is caused mainly by the work overload of professionals and, more specifically, by management's short-term decisions. Any change considered within a budget-restructuring perspective will necessarily come up against the paradox between the personnel's needs and the decision to cut staff, a significant source of 'expenditures,' often perceived by employees as a lack of consideration. For example, when governments decrease financing, the health system cuts

190 Clemence Dallaire and Sonia Normand

jobs and requires personnel to be more flexible in order to control budgets. There is no doubt that the Canadian health system relies on the work and competence of its professional staff, of which nursing represents nearly 65 per cent (Statistics Canada 1997). The performance of the health system therefore depends on the knowledge, skills, and motivation of the individuals responsible for providing services (World Health Organization [WHO], 2000). Besides these characteristics, recent studies show that the number of professionals involved in providing care and services has a significant impact, especially on the mortality rate. This has been brought to the fore in the case of nurses, on whom the majority of studies have focused (Aiken, Lake, and Smith 1994; Fagin 2001). Any decision to cut personnel or change their functions may be perceived as challenging their usefulness, thereby contributing to the sense of gloom and exhaustion. To date, a rapid succession of targeted, short-term changes have taken place. Meanwhile, the expected surplus of personnel has turned into a shortage, in an overall context where numerous other problems need be dealt with. Employees have suffered disruptions, and this has made the health professions less appealing. The idea that the health system is a source of employment offering competitive benefits likely to attract a sufficient workforce must now be abandoned. This loss of appeal has been revealed by a public debate and by policies in the field of personnel management, which seem to focus more on the difficulties involved in following the path of compulsory retirement and job cuts than on planning changes in personnel and on creating pleasant work environments. The overall impression is that physical resources are more important than individuals. It can be said that this is an important element of the current paradox in which 'the administration of things is substituted for the management of people' (Enriquez 1997). According to Koehoorn, Lowe, and Schellenberg (2002), senior managers who make decisions to modify personnel are often more concerned with cost-cutting and short-term efficiency than with the quality of life at work or the effects on patients (WHO 2000). Managers should obviously be concerned with the management of resources and their use, but they should also give adequate attention to the services offered by the health system and the individuals involved: management includes management of people. Adding financial concerns to personnel concerns may indeed complicate management, since what management is trying to accomplish affects the staff. Losing sight of employees may therefore be another way of adding problems that could have been pre-

Changes and a Few Paradoxes 191

vented. Fiscal constraints have led to all kinds of unexpected adjustments in health system personnel, which are apparent in the fact that professionals consider their working conditions difficult, whereas the public perceives a threat to the survival of the health system. We will need to change the thinking of managers and the organizational culture in order to acknowledge and give long-term support to health sector personnel, rather than carry out changes that do not emphasize the importance given to the staff. The focus on the difficulties of health system personnel will in turn make it necessary to put forward a new approach to personnel management in order to convey clearly the attention given to employees in implementing changes. In short, the personnel is the best instrument to facilitate change in the system with maximum effect when it is strategically linked to appropriate management, relevant policies, and equivalent financing (Biscoe 2001). The paradox between the importance and the lack of consideration of personnel has other dimensions. Considering employees does not simply mean being concerned with salaries and working conditions, since they are not the only things that motivate health professionals. Several strategies are recommended to ensure that the system adequately considers employees and, in particular, re-evaluates personnel-management functions and redefines its mission (Biscoe 2001). Other measures may also be considered in order to increase the validity of this paradox, notably recognizing personnel as a strategic asset, adopting a more realistic understanding of human beings, encouraging participation in decision making, and proposing changes that make sense to the professionals, as discussed in the following sections. Taking personnel into consideration means recognizing that the complexity of the health system resides in individuals and not in structures. In the changes currently being made and contemplated, that which is complex, that is, individual is simplified by reducing them to rational beings who will react to change in a linear way and in the desired manner, based on policies, if they are given accurate information; on the other hand, what is simple, that is, the structure, is made more complex. Recent models in the field of behavioural change suggest that such change is possible if the individual has the intention to change, but only with the necessary resources in the environment and factors facilitating decision making (Godin 2002). In this model, the complexity of the elements involved is emphasized. Godin suggests that the intention is defined by eight variables grouped into three categories, whereas the development of an individual is best described as a

192 Clemence Dallaire and Sonia Normand

spiral process with five stages, taking into account individual and environmental characteristics. In short, changes in organizations should be based on a notion of the individual that goes beyond that of a rational being with predictable behaviour, by taking into account recent knowledge in the area of human-behaviour patterning, which introduces a more complex model where the environment plays a vital role. Taking personnel into consideration entails recognizing it as a strategic asset that is carefully nurtured and demonstrating concern about the quality of life at work, the career development of professionals, and the quality of services provided. According to Smith and Preker (2001), concern should also be shown for the relationship between the individual and the organization (culture, policy, and structure). A better consideration of employees and better relationships between them, managers, and government would certainly help to develop a shared vision of the importance of personnel, including the importance of managers, who also need to be re-evaluated, as suggested in the Clair Report (2001). Furthermore, while problems are at the heart of the debate, the understanding of what employees accomplish on a daily basis is still not clear. There is thus an urgent need for a clear vision of the health system and the role of its personnel, since, without qualified, motivated, and committed staff, the state, through the health system, cannot play its role in the current context (Koehoorn, Lowe, and Schellenberg 2002). Leadership needs to be exercised in order to refocus changes on the fact that they are mainly carried out through health system personnel and to consider employees as an asset for the system. Taking personnel into consideration involves proposing changes that make sense to them and encouraging them to participate in change, based on what was mentioned earlier. Moreover, the resistance of professionals to change is understandable, since the change required is not experienced as a necessity. These professionals are trained to focus on the clinical aspect of a case, whereas changes are justified from a financial perspective, a subject with which they are generally not familiar and on which they have little information. Moreover, although they have the necessary information concerning the financial reasons, they are probably not concerned about this issue, they are more interested in the clinical aspect. Thus, they should be asked to be flexible in order to adjust and improve care and service rather than cut expenses, which is often the exclusive concern of managers. Justifying changes in a way that makes sense to the individuals who are being asked to make the changes appears to be quite simple.

Changes and a Few Paradoxes

193

Yet, associated with other measures, this nevertheless shows consideration for health personnel. Taking personnel into consideration also means allowing them to participate in decision making (Smith and Preker 2001). This participation is so critical that decentralization, which increases the role of professionals in the system, seems to be favoured (Bach 2001) and their increased commitment is expected. Decision making needs both management and clinical data that personnel holds, which illustrates the importance of participating in decision making. In the health system, differences in knowledge, values, and attitudes between various categories of professionals, between professionals and individuals who need their care and services, and between professionals and the government should be considered when changes are carried out (Biscoe 2001). Differences of this nature make it difficult to communicate information in an egalitarian and rational way when participating in decision making and subsequently developing policies. Taking into account the variety of viewpoints shows consideration for the individuals involved in the change process. Rather than encouraging participation in decision making, the focus is often on the quality of communication, even though the decision is conveyed to the staff after it has been made by managers and politicians. A principle often mentioned is the degree of transparency and simplicity of proposed changes, which has a significant impact on implementation (Bach 2001). From this perspective, a policy will succeed if it can ensure that employees understand what their role is in the overall change. Many believe that this simple communication task has not been executed or has failed, which would explain the lack of success of some changes. This lack of success probably has other causes too. For example, the very principle of transparency in conveying planned changes stems from a simplification of human beings, whereas the resistance to change is directly linked with the degree to which communication is effective (Smith and Preker 2001). While this position may be valid, effective communication is not a guarantee of success. It is not necessary to simplify personnel's attitude towards change to such a point that a stimuli-response type situation is created, whereby good policy would be well received. Individuals do not change their behaviour based on this type of model, since numerous other factors play a role in encouraging genuine participation in decision making, with all the efforts that this requires from the personnel, in addition to the efforts already made in providing care and services.

194 Clemence Dallaire and Sonia Normand

In short, change occurs through personnel, but the strategic importance of personnel in the health system is not generally mentioned in change projects. This paradox is a reminder that changes should be carried out by employees who are viewed as an asset, to whom the proposed changes are justified based on reasons related to care, and who participate in decision making.

Question \ Overall, has the planning of human resources in health been a success or a failure? What are the main strengths and weaknesses of the current approaches on a national, provincial and territorial scale? What changes should be made to the planning of human resources in the health sector?

Workforce planning aims to balance the number of workers and their location in the organization, that is, the right person at the right place at the right time. To achieve this type of balance, Egger, Lipson, and Adams (2000) suggest that there must also be coordination with the health system's mission. The very orientation of the system would thus be the fundamental element of personnel planning, as discussed in the first paradox. Until very recently, the growth in professional staff was controlled by training, which adjusted supply to needs. Such a concept was based on the idea that a continuous flow of professionals could be expected. In this context, planning was not necessary and was often not done, since it was thought that planning had little or no connection with health policies and the needs of the public (O'Brien-Pallas et al. 2001). It is important to remember that surpluses of health professionals were expected following the changes undertaken in the health system (Dussault, et al., 2001; Mitchell 1998; Pew Commission 1995). It now seems obvious that other elements linked with these changes are instead announcing shortages. In this context, planning the number of specialized personnel by controlling admissions in professional programs becomes problematic owing to the rather long time it takes to adjust to the needs. Moreover, this type of training requires significant efforts from individuals and creates incentives to preserve the investment already made. Consequently, training prevents rapid fluctuations in the response to demand. In other words, there is an element of irreversibility in planning based on the admission of students in a professional program that must be taken into consideration when plan-

Changes and a Few Paradoxes 195

ning the rapid changes that now appear to be required in the health system. According to Markham and Birch (1997), planning can be done on the basis of three approaches: needs, workforce demand, and use. The first approach, planning based on needs, is unusable because it must be carried out using data that cannot currently be obtained. As regards planning based on workforce demand, it is difficult to carry out because of a discrepancy between the number of professionals reported in government-planning documents and the actual number of professionals currently found in the care environment. The main challenge with this type of planning is dealing with clinical and administrative databases that do not provide information on the needs of the health system, the outcomes of the care provided by the staff, or the costs (O'Brien-Pallas, Birch, Baumann, and Murphy 2001). At present, this second form of planning is complicated. The third form of planning, that based on personnel-management use, has yielded satisfactory results in the context of shortage when it has used three types of strategies: more efficient use of available personnel through better geographic distribution; when applicable, increased use of multi-skilled individuals; and better matching of skills and functions (WHO 2000). Planning based on management allows effective tasks to be taken into consideration, the division and subdivision of each person's tasks in order to deliver several types of unrelated services, and their assignment to individuals based on their skills. Planning of this nature raises several questions: Who determines the tasks? Do the tasks primarily serve the needs of the institution and professions and subsequently the needs of the individuals under care? Is the matching of these tasks with the goals of the organization and those of the health system being monitored? These are important questions, since professionals complain that their skills and knowledge are not being fully used. This is a source of waste that has a demoralizing effect on personnel. Furthermore, these questions indicate a concern for quality and continuity of care. In fact, the division of duties leaves patients to deal alone with the problems posed by a lack of continuity between various professionals in the system. It is therefore extremely important to examine the work styles in order to gain a better understanding and to ensure a better match between tasks and needs (Gauthier 2000). The preceding paragraph underlines both the distinction between and the complementary nature of clinical and organizational aspects of

196 Clemence Dallaire and Sonia Normand and

workforce planning. The clinical aspect deals with the work content of professionals and the time required, whereas the administrative aspect deals with the number of individuals, in view of the clinical requirements, and with resources, costs, and organization of services. Planning that does not encompass these two aspects will certainly prove to be problematic. For example, the large number of female doctors has exacerbated shortages because they envision their career differently (for example, some opt to work less hours per week than their male colleages or to have periods of inactivity during a given year). Also, the responsibility of professionals within care and interdisciplinary teams is changing and, as a result, new roles are emerging (OIIQ 2001). Planning should be able to take these changes into account. In sum, this brief overview of planning, which has examined various methods, does not allow us to conclude on success or failure. Rather, it brings to the fore the paradox between various ways of planning that meet different objectives: planning based on training seems to promote the professional's vision of the health system, whereas planning aimed at meeting demand does not have the required data; planning based on effective practice must come to terms with an ambiguous definition of current tasks, which are often divided, leaving patients to deal alone with various professionals. This brings us to a new paradox. Paradox 4: Group-based Planning, but Increased Need for Mobility and Flexibility The preceding section clearly underlined that planning for health sector personnel suffers from a lack of conceptualization, that is, numbers are strung together without clearly knowing what they represent in reality. Planning generally done by groups of professionals is often referred to as 'silo planning.' This type of planning is also based on a hierarchical vision of health personnel, in which one group determines the required number of professionals from other groups. This has several ramifications. First, the focus of the personnel-planning analysis tends to be a profession, specifically, doctors and nurses (Bach 2001), with the medical profession dominating the health system despite the fact that it makes up only about 10 per cent of the total complement of professionals. Such a situation causes fragmentation into sub occupations and planning that assumes relatively permanent roles for different groups. To all intents and purposes, planning is based on medical needs and the number of doctors. Planning of this type could be effec-

Changes and a Few Paradoxes 197

tive provided that medicine standardizes its practice, thereby facilitating workforce planning. Considerable efforts are currently being made in this regard, particularly with the trend towards basing clinical decisions on credible data and practice guidelines. In the first part of the paradox concerning professional groups, the current difficulties in planning may be a consequence only of the model used, that is, planning based on medical needs and the fragmentation into suboccupations of other professions in a system that is almost exclusively curative, rather than an actual planning failure. Moreover, the health system is based on competencies, behavioural indicators required of professionals in order to increase the ability of the workforce to meet the demand (Bach 2001). By attaching importance to the skills rather than to the individuals, it is assumed that the flexibility offered by decompartmentalization and interdisciplinarity will provide a solution to the problems of personnel shortages (Clair 2001; Fyke 2001; Mazankowski 2002). In a curative system, the decompartmentalization of professions other than medicine seems to be favoured in order to achieve this objective, with implications for planning. According to O'Brien-Pallas, Birch, Baumann, and Murphy (2001), it is done in an intermittent way, varies in quality, and is focused on a particular profession. Another way of planning, one in which medicine would be a group among others, would be to consider each group independently from one another. This radically different approach assumes the autonomy of each profession. In addition, this type of planning requires an examination of the use of various types of professionals within the health system. To differentiate between what currently exists, it would entail direct access of the public to various professionals, rather than an almost complete reliance on medical practice, prescription, and hospitalization. The shift to ambulatory care that has led the practice in less traditional contexts, such as the community or in the home, may benefit from this type of planning. Personnel planning would change while remaining focused on forecasting the number of professionals required. The main difficulty in this scenario would be to specify what these various professionals would do and, accordingly, estimate the number required. A qualitative aspect should be added. What is needed is probably profession-based planning, which could take into account the flexibility required and collaboration with other professionals. Since this would likely offer many different variations, it is difficult to predict accurately the number of professionals needed. This last strategy would

198 Clemence Dallaire and Sonia Normand

include a concern for employees' skills, whereas the methods are concerned only with the quantity of personnel. The next section deals with some aspects inherent to this type of planning. In sum, the limitations of the existing models are better understood, but planning is a complex activity that must come to terms with several paradoxes and that does not provide an answer to the fundamental question related to personnel planning, especially since it is concerned more with the number of employees than with what they are supposed to do. Question 2 Are the policies and practices regulating the scope of practice, certification and licensure of professional health groups up-to-date and in accordance with the goals of a modern health system? If not, how could they be modified in order to promote innovation and the desired change while preserving quality and protecting public interest?

There seems to be agreement that several factors combine to create a working environment that is more complicated and much different than in the past (OIIQ 2001). For example, our society is experiencing significant demographic changes, such as population aging, increased life expectancy, the onset of new health problems, and growing social problems, that may result in a larger and more diverse health system clientele. In addition, budget constraints are pervasive, a point we have discussed in the context of conflicting objectives, as is the implementation of new technologies, which have an impact on the organization of care. The professional system does not seem to be in question, even though licensing requirements are viewed by many as a vestige of the past century that encourages misuse of resources, errors, duplication of work, and a lack of collaboration. Despite these disadvantages, the professional system seems to be achieving its objective of protecting the public, while changes in policies and practices governing the scope of professional practice are being planned. Moreover, the notion of competence is increasingly used to try to address the various limitations of the professional system and to facilitate planning. In this context, addressing the issue raised by the policies and practices regulating the scope of professional practice may entail examining the regulations from the perspective of a paradox between regulating a field of practice and promoting skills.

Changes and a Few Paradoxes 199 Paradox 5: Field of Professional Practice and Competence The paradox lies in the desire to preserve the professional structure and, at the same time, introduce the notion of competence, a goal that, although not the opposite, of the first, is nonetheless of a different nature. The paradox arises from efforts to increase the clinical independence of professionals based on a specific field of practice and the increased flexibility expected of them from skill-sharing. The notion of competence has been suggested as a way out of the current impasse resulting from the division of fields of practice, although many consider that the notion of collaboration and interdependence has become necessary in the integrated health system, in which the interdisciplinary team takes charge of patient care. New ways of managing and organizing services and care, as well as new structures to support this decompartmentalized model, are anticipated. Participative management, which promotes dialogue and discussion between partners, would facilitate collaboration on abolishing hierarchical structures and replacing them with horizontal systems. In this context, a certain flexibility would be expected in defining roles, whereby skills would serve as indicators and as a way to reduce conflicts between professionals. The appeal of the competence-based approach is that it would provide a foundation to describe and convey professional practices that are often perceived as a group of disparate activities without much coherence (Bach 2001). The competence-based approach, which entails focusing on behaviours rather than on the qualifications and fields of professional practice, could bring down the barriers between occupations and encourage interdisciplinary work and flexibility on the part of professionals. There is an impression that assignments based on the notion of competence could resolve the compartmentalization of fields of professional practice. In short, the paradox lies in the juxtaposition of a specific professional field with a group of skills that allows professionals to be flexible. The competence-based approach is often viewed with mistrust by professionals, who want to preserve their monopoly on expertise. Many also mention the need to protect against a risky sharing of activities requiring a unique and complex level of skills. For them, the barriers offers protection against the confusion that the competence-based approach could cause. Thus, the notion of competence is not without its own problems, since it is secondary to a reflection on the role that generally arises from

200 Clemence Dallaire and Sonia Normand

the field of practice, once again emphasizing the paradox at the centre of the two approaches. This approach assumes that a certain distinction persists between the roles. Moreover, the concrete implementation of interdisciplinarity requires a clear definition of the roles, functions, and the contribution of every individual, as well as a good understanding of the objectives of partnership. According to the Pew Commission (1995), changes in the fields of professional practice must aim to reduce conflicts between professionals and the confusion among the public. Collaboration and interdisciplinarity will not be without conflicts. Conflict-resolution mechanisms and procedures for sharing fields of practice between two or several professions will be important, perhaps more so than the notion of competence. As for promoting innovation, the shortage of personnel offers an excellent opportunity to transform care and services. It offers an opportunity to refocus on health before illness occurs, when this is possible, because health requires fewer personnel resources. Innovation could consist in offering curative services and simultaneously making a greater effort to promote health and prevent problems, using knowledge to improve the public's health. It could also be an opportunity to offer the public direct access to various professionals without having to go through medicine or hospitalization. When all is said and done, the current shortage is an ideal opportunity to re-examine the procedures and ways of delivering care and the determination of fields of practice, and to review the public's access to various professionals currently under the control of medicine in the public system. In sum, professional regulations are still seen as the best way to ensure the public's protection. Adding the notion of competence creates a paradox when the two are considered as interchangeable. In fact, competencies are based on fields of practice and professional qualifications and must remain secondary to a definition of the professional role and mission of the health system. To increase flexibility, conflictresolution mechanisms linked with interdisciplinary collaboration are desired instead. Question 3 Are educational programs producing a workforce that has the capacity and necessary perspectives to meet the needs of a changing and increasingly integrated health environment? If not, what changes should be made?

It is quite difficult to offer a definitive answer to this question, which

Changes and a Few Paradoxes 201

raises several other issues. Training is intended to meet the new requirements of the health system and changes in the system. What exactly does this mean? • Is this a call for change in the more general context of the hierarchical organization of the health professions? If so, should the profession at the top of the hierarchy (medicine) be the only one adding new elements to its training? Will it delegate to others some of its current prerogatives, in particular those that give it the impression of being under utilized? The context of desired interdisciplinarity, present in the debate but very marginal in reality (Schmitt 2002), seems here to provide a good example. • Or is it perhaps a new way of providing care and services based on an approach where the entry point into the system is not necessarily medicine? Could it be an interdisciplinary practice that the public could directly access, and that would require much greater clinical independence for all professionals, even though they would collaborate on interdisciplinarity by upgrading skills? If so, all professions would have to make changes to their training. • Or does it mean widespread changes in the practice environments, or even a requirement of mobility between care and service environments which, by providing less supervision and a reduced presence of professionals, demand greater clinical independence of all professionals and, consequently, changes in training? • Or - and it is necessary to mention the possibility - is it a change brought up in the debate even though the practice environments have barely changed or the adjustments made have already been incorporated into the training; for instance, 93 per cent of the programs in U.S. nursing and medical schools already use the competencies suggested by the Pew commission (Brady, et al. 2001). Another important element is the perception of professionals, who consider that their knowledge and skills are currently under utilized. For example, the Saskatchewan Report (Fyke 2001) indicates that many complain that their competencies and skills are not fully used and that, if they were, better patient outcomes and system savings would be achieved. These questions are not easy to answer, especially since they are mutually exclusive: answering in the affirmative to one automatically means answering in the negative to another. And several other questions may be raised when training changes are discussed. One focuses

202 Clemence Dallaire and Sonia Normand

on when changes should be made in training programs: Should they precede or follow changes in professional practice? Implementing changes in training before changes in practice would likely create some uneasy graduates, who would be faced with practice environments that do not reflect their expectations. On the other hand, if training changes follow changes in professional practices, other problems would arise, given the fairly long delays before graduates with the necessary training would come on the market. Furthermore, insisting on the necessity of changing training could suggest that it is in fact a strategy to force the system to change, by making new professionals individually responsible for change. The overhauling of training programs is certainly justified and may serve as a catalyst for change in the health system, but it may put the burden of systemic changes on beginners. These changes come up against great resistance from existing personnel for reasons that are obvious and others that are not so clearly understood, and that seem to stem from the paradoxes discussed in this study. Intentions of this nature regarding training should be carefully considered and support mechanisms for graduates should be planned if the need arises. Moreover, changes in training often emanate from those in the practice environment who complain that new graduates are not able to function autonomously or are even practically incompetent. In light of the repeated criticism heard from the practice environment regarding the training environment, an initial question is whether or not apprenticeship or professional training is required. Apprenticeship training prepares for the execution of standardized tasks and seems at first to be the best way of meeting the needs both of the practice and of a certain form of management of this practice. Furthermore, professional training does not allow beginners to master perfectly the skills and make the links between the knowledge acquired and practice situations. Requiring complete mastery would entail longer training and gradual integration into the service environment. These criticisms are becoming more relevant on account of changes in the health system and shortages of workers. Consequently, the clinical environment has difficulty supporting beginners. The latter need specific support in order to develop professionally and become autonomous, and the practice environment is the only one able to support this professional maturation. Demanding changes in training does not seem to address the real problem. In addition, practical training adds another fundamental problem to

Changes and a Few Paradoxes 203

the current situation, especially when it is suggested that professionals should be better prepared to practise in diverse clinical contexts. The health system has traditionally trained professionals in hospitals but is now headed towards out-of-hospital care. This environment does not offer better conditions for hosting and supervising students. There, professionals often work alone or in small groups, and may find it difficult to get away from their duties to provide supervision. The decrease in financing for training associated with changes in the health system limits the ability of the training environment to offer practical training to all students because of staff cuts. It therefore becomes necessary to examine the role of everyone in practical training, as well as in situations where the system has not succeeded in ensuring an internship environment that reflects the work environment, and to implement transition procedures for new graduates. The health system should examine its commitment to practical training and promote effective integration of training and practice in order to benefit from a new generation of competent, well-trained workers who will not decide prematurely to leave their profession because the integration conditions are too difficult (Baumann, et al. 2001; Lindeman 2000). Moreover, new approaches to care and services provision increasingly aim at ensuring the continuity of care supervised by professionals. Proceeding in this way adds to training requirements. Training for continuity of care may mean training personnel to be mobile, in order to monitor patients over the whole-care episode, that is, in the pre-, intra-, and post-hospitalization phases. Professionals should also improve their organizational knowledge of the health system in order to ensure continuity of care across various establishments. However, this type of training will require structural changes in the health system, in particular changes in the employment contracts of several professional groups, which make it difficult to have increased mobility among the workforce. Continuity once again raises the question of when to introduce training; meanwhile, structural constraints are still in place. Conclusion The succession of changes in the health system over recent years has had a significant impact on personnel. Several recent reports by specialists on health system personnel have already informed the commission about the present situation, the changes made, the alternatives, and the main issues.

204 Clemence Dallaire and Sonia Normand Several findings and paradoxes that arise from the evidence have been examined. This brief study has shown that the debate about change in the health system is widespread and that it would be advantageous to consider employees as the focal point of care to proceed from a collective vision of the health system. The regulation of professionals seems adequate, although it does not clearly show how it can be reconciled with interdisciplinarity, cooperation, and practice in a variety of environments. The training of professionals raises important issues regarding the appropriate time to introduce changes in a way that would allow professionals to feel comfortable when the mission of the health system is precisely defined and the nature of the personnel's work and training is subsequently clarified.

NOTE 1 The word 'personnel' will be used here instead of the expression 'human resources' in order to highlight the difference between individuals and financial and material resources.

REFERENCES

Aiken, L.H., H.L. Smith, and T.E. Lake. 1994. Lower medicare mortality among a set of hospitals known for good nursing care. Medical Care 32: 771-87. Bach, Stephen. 2001. Human resources and new approaches to public sector management. Improving health resources management capacity. Geneva: World Health Organization. Baumann, A., L. O'Brien-Pallas, M. Armstrong-Strassen, J. Blythe, R. Bourbonnais, S. Cameron, D.I. Doran, M. Kerr, M. Vezina, M. Butt, and L. Ryan. 2001. Commitment and care: The benefits of a healthy workplace for nurses, their patients and the system. Ottawa: Canadian Health Services Research Foundation. Brady, M., J.D. Leuner, J.P. Bellack, R.S. Loquist, P.P. Cipriano, and E.H. O'Neill. 2001. A proposed framework for differentiating the 21 Pew competencies by level of nursing education. Nursing and Health Care Perspective 22 (1): 30-5. Clair, M. 2001. Les solutions emergentes. Report of the Commission d'etude sur les services de sante et les services sociaux. Quebec City: Ministere de la Sante et des Services Sociaux [MSSS]. Dussault, G., M.A. Fournier, M.S. Zanchetta, S. Kerouac, J.L. Denis, L. Bojan-

Changes and a Few Paradoxes 205 owski, M. Carpentier, and M. Grossman. 2001. Le marche du travail en soins infirmiers au Canada: revue de la litterature. Report presented at the Table ronde des intervenants en soins infirmiers, Universite de Montreal, Groupe de recherche interdisciplinaire en sante. Egger, D., D. Lipson, and O. Adam. 2000. Human resources for health. Issues in health services delivery. Discussion Paper no. 2. Geneva: World Health Organization. Enriquez, E. 1997. Les jeux de pouvoir et du desir dans I'entreprise. Paris: Desclee de Brouwer. Fagin, C.M. 2001. When care becomes a burden: Diminishing access to adequate nursing. Report, Milbank Memorial Fund. Fournier, L. 1999. Pourquoi le virage ambulatoire? in O. Goulet and C. Dallaire, Soins infirmiers et societe. Boucherville, Que.: Gaetan Morin. Fyke, J.K. 2001. Caring for medicare: Sustaining a quality system, Commission on Medicare. Saskatoon: Government of Saskatchewan. Gauthier, G. 2000. Rapport du Forum sur la planification de la main-d'ceuvre. Quebec: Ministere de la sante et des services sociaux. Godin, G. 2002. Le changement des comportements de sante, in G.F. Fishter, ed., Traite de psychologie de la sante Paris: Dunod. Koehoorn, M., G.S. Lowe, K.V. Rondeau, and G. Schellenberg. 2002. Creating high-quality health care workplaces. Ottawa: Canadian Policy Research Networks discussion paper. Lindeman, C.A. 2000. The future of nursing education. Journal of Nursing Education 39(1): 5-12. Markham, B., and S. Birch. 1997. Back to the future: A framework for estimating health care human resource requirement. Canadian Journal of Nursing Administration 10(1): 7-23. Mazankowski, D. 2001. A Framework for reform - Report of the Premier's Advisory Council on Health. Edmonton: Government of Alberta. Mintzberg, H. 1994. Grandeur et decadence de la planification strategique. Paris: Dunod. Mitchell, J.G. 1998. Recreating health professional practice for a new century. Fourth Report of the Pew Health Profession Commission. San Francisco: UCFS [University of California at San Francisco] Center for the Health Professions. O'Brien-Pallas, L., S. Birch, A. Baumann, and G.T. Murphy. 2001. Integrating workforce planning, human resources and service planning. Geneva: World Health Organization. Ordre des infirmieres et infirmiers du Quebec (OIIQ). 2001. La vision contemporaine de I'exercice infirmier. Montreal: OIIQ. Pew Health Profession Commission. 1995. Critical challenges: Revitalizing the

206 Clemence Dallaire and Sonia Normand health profession for the twenty-first century. Third Report of the Pew Health Profession Commission. Schmitt, M.H. 2002. Seizing the moment: an opportune time to study the outcomes of interprofessional education and health care delivery. Editorial, Research in Nursing & Health 24: iii-v. Sinclair, G. Duncan. 2000. Looking back, looking forward. Ontario Health Services Restructuring Commission. Toronto: Government of Ontario. Smith, A. Douglas, and Alexander S. Preker 2000. Labour market adustment in health systems. Geneva: World Health Organization. Statistics Canada (Demography Division). 1988-97. Health personnel in Canada. Available at http://www.cihi.ca/english/fmedrls/juneltablesf/ tableau_l.shtml (1988 to 1997). World Health Organization. 2000. Why do health systems matter? World Health Report 2000, chapter, 3-22. Available at www.who.int/whr/2000/en/ report.htm

7 The Influence of Physician-Payment Methods on the Efficiency of the Health Care System MICHEL GRIGNON, VALERIE PARIS, AND DOMINIQUE POLTON, with the cooperation of AGNES COUFFINHAL AND BERTRAND PIERRARD

The efficiency of the health care system affects the public economy and economic policy because health costs are always met from the pooled contributions of individuals and corporations, if not from various forms of taxation. The question of efficiency has long been confined to controlling the rising costs of health care, but it is now being raised in a more complex fashion: how to reconcile reasonable spending with the requirement for quality and fairness. This question has generated a great deal of economic literature, which analyses the characteristics of the medical care market and the behaviours of producers and consumers of care in order to suggest optimum contracts to the regulator for satisfying the three contradictory objectives of expenditure containment, quality, and fairness. Focused on latent behaviours within the payer/patient/physician triad, the theoretical literature has given rise to unresolved empirical questions, particularly as regard the extent of the physician's influence over the patient. Our interest here is in the segment of this literature that addresses the impact of the method of physician payment on the volume of services the physician provides or prescribes for the patient, the quality of those services, and inequalities of access to care. The method of paying the physician is a true strategic choice, and there is no 'natural' mode of payment. Because the physician is subject to an obligation of means and not of outcome (for reasons of uncertainty), he is not in fact a classic producer who sells an outcome or end-product and takes the risk that this end-product may be worth less than its production cost. Besides,

208 Michel Grignon, Valerie Paris, and Dominique Polton

we are talking about the remuneration of physicians, which is what makes them factors of production, and not the retail price of medical services. And yet the physician is not a mere factor of production, because his degree of autonomy in organizing the elementary actions that contribute to the production of his services is a component of the quality of the services provided. After a brief presentation of the different methods of physician remuneration, the first part of this paper, The Economic Literature/ will summarize the economic literature's analysis of the impact of these methods on health care costs. We explain the theoretical issues behind the various payment mechanisms and examine the extent to which the empirical studies conducted in various national contexts confirm these theoretical postulates. Economists hypothesize that, while the physician's profit is not the sole motive behind a medical decision, it is an element of it. This hypothesis is shared and supported by medical sociology, which shows that when the consequences of a medical decision for a patient's well-being are not clear, the trade-offs for the physician in terms of leisure, reputation, or income take on greater importance. It is then no longer certain that the patient is receiving the exact amount of care that his treatment requires. Economic analysis implicitly raises the following question, among other: Given the motivations of physicians, what impact does the method of payment have on medical expenditure? Such an attitude does not suggest that motives other than profit (professional standards, respect of peers, or the patient's interest) have no significance, but rather that methods have no impact on them. In reality, however, this hypothesis is certainly erroneous, and experiences of changing physicians' modes of payment can show us why the mechanisms adopted can depart significantly from economic recommendations, particularly because, as explained in the second part of this paper, 'In Search of the Optimum Method of Payment: Lessons from a Variety of Experiences,' the payment method can interfere with these other motivations of the physician. The Various Mechanisms of Physician Payment It is customary to distinguish three main methods of physician remuneration: • fee-for-service (FFS);

Physician-Payment Methods and Efficiency 209

• capitation, which is a lump-sum payment per patient managed for a given period of time; and • salary. Different countries use one or another of these three forms of compensation for general practitioners (GPs). Specialists generally receive a fee-for-service or are on salary. This traditional typology raises certain questions. First of all, the overall remuneration of a physician can combine a number of methods of payment: these mixed or blended payments are the practice in some health care systems, as we shall see in the second part of this paper. Next, while it is convenient to play these three mechanisms in opposition to each other, they can also be considered on a rising scale of inclusiveness, in which intermediate situations can be imagined: • At one extreme, FFS is the most specific and least inclusive form of payment for a medical service. Yet it still constitutes a fixed price: when a consultation is paid at a flat rate, as in France for example, it makes no difference whether it is long or short, whether a clinical examination is performed or a prescription is simply renewed. Further, the degrees of inclusiveness of the fee schedules used in different countries for pricing services vary, and a comparison of those fee schedules could be one way of describing different types of FFS. • Capitation adds another level of inclusiveness, since the physician is paid the same amount for a patient regardless of the quantity of services provided. The physician's income varies only according to the number of patients taken. • Salary is the all-inclusive extreme form of payment, where income does not change regardless of the physician's level of activity - in terms of the number of patients in care or the quantity of services provided per patient. On this scale, one can imagine intermediate situations, for example, between FFS and comprehensive capitation: a lump-sum payment per episode of care, or a lump-sum payment for monitoring a chronic pathology over a given period of times. A few experiments with these types of mechanisms will be described in the paper's second part. In summary, even though the three 'archetypal' methods of payment are traditionally opposed to each other - a practice we shall adopt in our review of the economic literature below - there is no reason why intermediate methods cannot be devised.

210 Michel Grignon, Valerie Paris, and Dominique Polton

Finally, as regard the forms of payment of health professionals, we are witnessing the emergence of mechanisms that do not fit any of the three models mentioned above but are similar to results-sharing mechanisms, whether those results are financial in nature (cost containment) or target the quality of the medical service. This type of mechanism, of which we provide examples in the paper's second part, might be considered a fourth mode of compensation, as an alternative to the three previous ones. That is, in fact, the approach taken by some authors who are now distinguishing between four methods, often labelling the fourth 'target payments.' In practice, however, this type of results-sharing is only complementary and peripheral to the main method of payment. It is thus difficult to consider it as an alternative - it is, rather, an adjunct to the three main methods of payment. In future, it could possibly become a greater factor in physicians' remuneration. The Economic Literature We present here a review of what the economic literature has to offer on the impact of the method of payment of physicians' services on the total health care budget. The Facts and the Simple Theoretical Postulates Based on aggregated data, a country's total health expenditure is affected by supply variables in general and the method of physician payment in particular. In nineteen OECD countries, Gerdtham et al. (1992) find that, all other things being equal, 'physician fee-for-service' (i.e., a variable indicating that this is the majority payment method in the country) increases health spending by 11 per cent. A more recent study employing panel data (for twenty-two OECD countries between 1970 and 1991) finds that FFS per se has not a direct but rather an indirect effect: medical density increases expenditure when physicians are paid on a FFS basis (Gerdtham et al. 1998). However, these relations, observed on aggregate data, are difficult to interpret. It is possible that total national spending and specific institutional arrangements such as FFS are both positively correlated to a third variable, such as the fact that physicians practise in isolation. The economic literature thus attempts to describe the micro-economic process by which the method of physician payment might have an influence on medical spending; to formalize this process, a payment

Physician-Payment Methods and Efficiency 211

method is characterized by its degree of inclusiveness. In the rest of this section, we will focus on the opposition between systems of inclusive or lump-sum payment (capitation and salary) and FFS schemes. We can identify how payment methods are expected to affect health care costs by isolating the other objectives of the health system (quality and fairness). To do so, let us briefly describe the physician's activities. To care for patients, the physician uses certain resources. Resources such as an office are necessary whether there are patients or not; others are used only if there are clients. In reality, the physician does not set his price himself: he receives it from a payer. The resources used in production are of two types: quantifiable material resources and nonmaterial resources that are difficult to quantify. An example of the former would be the consumables of the medical practice; examples of the latter would be the time the physician devotes to each patient, his competence (a function of his continuing training, the effort he makes to keep up to date), his ability to listen to the patient, and his mental effort. In economics, the level of non-material resources is usually referred to as the 'physician's effort.' An important difference between material and non-material resources is that the payer can take the former into account in the payment scheme, whereas doing so for the latter is more problematic. It is difficult for a payer to compensate a resource when there is no proof that the resource was actually used. With this simplifying model in hand, what can be said about the methods of payment? In FFS, the price takes into account fixed costs and variable costs (if not, the physician refuses to work) and so is higher than the average cost of the variable resources alone: thus, caring for an extra patient brings the physician a financial profit. If, in addition, the physician has some power over the patient's demand, FFS creates favourable conditions for a larger quantity (and hence expenditure) of services than the patient may wish to receive, or than may have been medically necessary. Under this model - the physician has some power over the patient - if the payer reduces the unit price, the physician will respond by increasing quantity1 to maintain his profit; this explains why a reduction in the price of care leads to an increase in the supply of services. Under a capitation formula, on the other hand, the physician is not reimbursed proportionally to the resources committed and so has no interest in multiplying his services, since he would then lose money or waste efforts. He prefers to treat the patient using the least possible amount of resources to achieve a given level of quality.

212 Michel Grignon, Valerie Paris, and Dominique Polton

The Cost-Quality-Fairness Trade-off: Status of the Problem If the question were limited to cost containment, one might wonder why all regulators have not opted for some form of lump-sum payment. Obviously, there are historical reasons that support FFS in spite of the evident superiority of the lump-sum payment as an instrument of budget control (see, e.g., Emery, Auld, and Lu 1999). However, in the culturally homogeneous United States, one notes that different payers have adopted different arrangements: some prefer a form of capitation, while others prefer FFS; the latter are surviving, and even seem to be recapturing market share lately. So there are certainly reasons of cost-effectiveness that also explain why the lump-sum payment method is not universally adopted: • An initial reason is related to the quality of the medical service provided: while lump-sum payments offer a structural means of resolving the problem of increased volumes of services delivered by physicians, it raises the risk that the physician will provide too few services (consultations, examinations) and too few non-material resources (shorter consultations, less attention) to the patient. Of course, the regulator can impose a minimum level of services per patient, but it will be difficult to control the level of non-material resources. If the physician is paid to manage a patient and is not sanctioned in any way,2 he has an incentive to collect the lump-sum payment and not provide the corresponding services and effort. The economic literature tries to shed light on this point and to define the circumstances where the lump-sum payment method can jeopardize medical quality. More generally, lump-sum payments can be seen as a delegation to the physician of responsibility for the optimum organization of services that are factored into health care, whereas, with FFS payment, this role is shared between physician and regulator. Where there are multiple payers, FFS raises an additional problem of inconsistency among pricing regimes, but to our knowledge this problem has been studied only for hospital services (Tai-Seale, Rice, and Stearns 1998). • This quality problem can also have consequences for spending: the physician is not only a producer of services, he is also a prescriber. Capitation is generally accompanied by a referral responsibility for the primary care physician (PCP): he makes decisions concerning quantities of analyses, medications, and specialized or hospital con-

Physician-Payment Methods and Efficiency 213

sultations. If the physician reduces the intensity of his effort because he is paid in a lump-sum fashion, even as he remains concerned for his patient's health, he can very easily generate a high volume of health care costs by increasing his prescription rate. • Finally, there is the issue of universal access to care. The lump-sum formula always entails a financial risk for the physician: if doctors are averse to uncertainty, they will refuse to take patients who incur highly variable ex ante costs. If doctors have better selection skills than payers, they will make money by treating individuals paying a high lump-sum and refusing to provide care to others (e.g., see Mulligan 2002). Ultimately, then, the question is: What form of payment can best reconcile the objectives of expenditure control, quality of medical service, and universality of access to care? We will first look at the theoretical models that explain the physician's discretionary power over the patient's consumption, models that conflict over the potential ability of the physician to 'induce' patient demand (Evans 1974). This ability has some important consequences for our study, since it increases the risk of lump-sum payments leading to a reduction in the medical services received by patients. The discussion concludes with an analysis of the physician's behaviour as a prescriber of medical services that he does not provide himself. The Cost-Quality-Fairness Trade-off: Discretionary Power of the Physician Theoretical Approach: We saw earlier that the physician will profit from FFS if he is able to impose on the patient more services than the latter wants or needs. How is this made possible? There seems to be broad agreement that, in interacting with his patients, a physician does indeed have a superiority that allows him to influence the consumption of his services. One reason for the physician's superiority is that he has a knowledge advantage over his patient. There is abundant theoretical literature debating the fact that superior knowledge is sufficient for an expert to impose an agency relationship, that is, a relationship in which the layperson puts himself entirely in the expert's hands. This literature concludes that, in a competitive context among experts, if the lay person can get a fairly accurate idea of the link between the expert's effort and his results, then

214 Michel Grignon, Valerie Paris, and Dominique Polton

superiority of knowledge is not enough to establish an agency relationship. The expert is selling so-called 'experience' goods, the quality of which will ultimately be determined by the buyer. On the other hand, if some fundamental uncertainty prevents the buyer from forming a notion of how the expert's effort affects the result he obtains, then an agency relationship can exist. The expert is then selling 'confidence' goods, since the lay buyer has to have absolute trust in him. What about medical goods? Rice (1993a, b) advances the radical idea that there simply is no such thing as informed and independent demand for care on the part of the patient. He bases his argument on the fact that, according to the findings of Rand researchers, the additional care obtained by the best-insured patients is not, in the view of independent experts, really appropriate given their health status. The idea that there is no independent patient demand for care is no doubt excessive (we all make health decisions independently of our physicians), but, on the other hand, it is true that part of the medical services provided by the physician is not observable. This difficulty in assessing the medical services provided is undoubtedly one of the reasons why physicians in developed countries are not generally authorized to own pharmacies or sell medications: if they could increase their earnings in this way, they might abuse their expert's position by persuading patients that it is in their interest to consume drugs. One can also make a few a priori predictions. For example, it is certainly more difficult to observe the quality of a specialist's services whom we rarely see: to resolve the asymmetry of information, the regulator will prefer to have the quality of this specialist's services observed by another physician through referral3 by a PCP. However, empirical proof of a degree of inducement on the part of physicians is difficult to establish: even if it is determined that different treatments were provided by two doctors to two identical patients, and that a connection is found between this variation in practice and the economic interest of the physician, it is still necessary to ascertain whether the variation is not perceived by the patient as a deterioration of services received. In fact, the agency relationship assumes not only that the physician imposes his views on the patient, but also that the patient does not realize it (he trusts the expert). As McGuire (2000) demonstrates, however, one can easily imagine a physician imposing his views even if the patient realizes that the physician is not doing exactly what the patient wants, or what the physician would do for himself. In technical terms,

Physician-Payment Methods and Efficiency 215

an agency relationship is not necessary for the physician to hold some discretionary power over the patient: it is enough for the physician to be in a situation of imperfect competition with patients for the latter to accept a service delivery that is not totally satisfactory. One good reason explaining this imperfect competition between physician and patient is the latter's attachment to certain characteristics of his doctor, which results in the patient not seeking out another doctor. The patient's attachment to 'his' doctor (family doctor, the usual source of care) is abundantly documented and is explained by the fact that long-term familiarity allows the physician to specialize in the patient. Obviously, this attachment of the patient to his doctor applies more to the GP (or PCP) or chronic care specialist. As a result of this attachment, physicians are not perfect substitutes for each other: each doctor more or less has a free hand with his patient and can impose different choices upon him than those the patient would spontaneously select if he were able to replace his physician with no adverse effects. In economic terms, having this freedom translates into 'having market power' and behaving like a monopoly. The physician is precisely in a situation of monopolistic competition with his patient: like a monopoly, the physician can take optimum advantage of his patient's willingness to pay by playing on prices and quantities (whereas, in a situation of perfect competition, there is only one price, that of the marketplace); but for all that, if the physician goes too far (strays too much from his patient's expectations or needs), the patient can leave him for another doctor. Attachment and agency can even be said to complement each other in the way they operate: when the patient sees the doctor only rarely, he has little attachment to him but also has more chance of being in an agency relationship and having to bestow blind trust; on the other hand, when the patient is in a long-term relationship with his doctor, it is easier for him to observe the quality of services received but more difficult to act in a competitive manner and oppose services that leave him dissatisfied. What impact might the nature of this discretionary power of the physician have on the problem that concerns us here? Schematically, the closer one gets to a situation of monopolistic competition, the more the physician has to take into account the risk of the patient's departure: with FFS, that threat limits the physician's power to provide too many services the patient does not want; with capitation, it limits the

216 Michel Grignon, Valerie Paris, and Dominique Polton

physician's power to reduce his volume of services excessively and, above all, the effort he devotes to his patient. If an individual's attachment to a given doctor is a constant variable, the physician arbitrates between the length of his client roster and the benefit he derives from each client listed on it. The physician is then said to influence the patient's consumption. On the other hand, the closer we get to an agency relationship, the less likely the patient is to pose a threat of departure for the physician, who can provide too many services if he is paid on a per-service basis or expand insufficient effort if he is receiving a lump-sum payment. In this case, the physician is said to induce demand. A large body of literature attempts to prove empirically, in an indirect way, that the physician has absolute discretionary power, since the patient is not in control. A classic test is to measure the response of physicians to a variation in medical density in a FFS context. Remember that, in the case of inducement, the physician has no competition: if medical density increases, the physician will have on average a lower number of patients (where inducement is involved, the quality of the physician has little impact on his attractiveness, for he is not in competition) and so will react to preserve his income by increasing the volume of services used by his patients. But in a context of monopolistic competition, higher medical density raises the risk of the patient leaving the physician (a break being less costly to the patient), so the physician will not create a wider gap between the consumption wanted by the patient and the services provided. Apart from the empirical issues (the truly exogenous nature of the increase in density, the difficulty of observing major variations), this test is not perfect; it is based on the arguable hypothesis that, in monopolistic competition, an increase in medical density facilitates the departure of unhappy patients; in fact, if a patient's attachment is due to personal characteristics of the physician that are not affected by medical density, one can find an increase in services following an increase in medical density in a context of influence. Pauly and Satterthwaite (1981) raise a similar objection: a higher density of physicians increases the capacity of each physician to influence demand via a process of medicalization of public expectations. Overall, it does not seem possible to decide on the degree of inducement on the basis of measurements of physician activity. Direct surveys of patients' level of satisfaction and their motives for choosing or leaving a physician would doubtless be of great interest. But it must be

Physician-Payment Methods and Efficiency 217

acknowledged that there has been no measurement of the degree of inducement based on such direct statements by patients. An invaluable indication is provided by the fact that, in the United States, the sickest persons insured by Medicare prefer to remain in FFS plans and retain the freedom to access the physician of their choice instead of opting for the more extensive coverage proposed by Health Maintenance Organizations (HMOs). This is a reason to suspect that the physician's latitude to persuade the patient that what he needs is precisely what will maximize the physician's income may increase with the patient's degree of morbidity. Empirical Approaches: Faced with varying medical practices, it is impossible to know beforehand whether a given practice is excessive or whether another is restrictive. Practices can be compared after the fact to some technical standard, such as a recommendation, or to a statistical norm, such as an average practice, but there is nothing to validate such comparisons. Hence, we can measure the impact of the method of physician payment only on the volume of services consumed by the patient, quality, and universality of access. To measure this impact, researchers look for comparisons where all other things are held equal except the payment method. Scott and Hall (1995) have found seven empirical studies published in English on this theme, noting that the small number is linked to the fact that these are 'opportunistic' studies which make use of data compiled for other purposes. Gosden et al. (2001) list another similar study. For comparisons to be meaningful, we must first be sure that the contexts are comparable: comparing the impact of different compensation systems in areas that are far removed from each other geographically or socially does not make much sense. Therefore, one must find homogeneous contexts with sufficient variability of payment methods. A second problem with this type of comparison is that it is difficult to ensure that all things apart from the payment method are in fact equal: the fact that a patient chooses a physician who is paid on a per-service basis or by capitation can be correlated with some unobservable characteristics that are themselves related to the use of the health care system. Similarly, physicians can specialize by type of contract (Glied 1998), again for reasons not unrelated to their work methods and the volume of services they provide. Finally, it is difficult to be sure that the level of payment is the same from one method to the next, and that

218 Michel Grignon, Valerie Paris, and Dominique Polton

one is not measuring the combined effect of a difference in payment method and level. Hickson et al. (1987) conducted an experiment controlling for the first two problems: fifteen pediatricians, members of the same clinic, were randomly allocated a mode of payment (salary or FFS). The study compares the volume of preventive visits to physicians on salary and those paid on a per-service basis: with controlled observable characteristics of physicians and patients, the pediatricians paid on a per-service basis delivered significantly more visits. This additional volume also reflects an ambiguous variation in quality: patients treated by physicians paid per service say they received better follow-up, but the volume of preventive visits suggested by these physicians departs more from the recommendations of the American Academy of Pediatrics than the volume observed for salaried physicians. According to Gosden et al. (2001), this average difference between salaried physicians and those paid on a per-service basis involves two different phenomena: visits in excess of the recommended number are fewer for patients of salaried physicians than for those of FFS physicians, but the recommended visits missed by the former are more numerous than those missed by the latter. Another study with physicians randomly assigned to payment methods is that of Davidson et al. (1992) on eighty physicians treating children under the U.S. Medicaid program. The study monitors three groups: a control group, paid on a per-service basis; a FFS group receiving higher rates; and a group paid by capitation. The volume of primary care sessions increased for the two tests groups, underscoring the impact of the level of remuneration regardless of payment method: however, it increases more in the FFS group (+0.8-0.9) than in the capitation group (+0.5-0.6). The price to be paid for these methodological gains (random assignment of physicians to payment methods) is obviously the sample's weakness and results in findings that cannot be easily generalized. We therefore turn to some 'before/after' comparisons, in which a payer changes the payment method of a single group of physicians and patients. In this case, we can assume a real exogenous change in payment method and observe its effect free of any bias. Furthermore, in implementing reforms, efforts are made to maintain physicians' income at a constant level, so the effect of payment method can be measured separately from the level effect. An initial study of this type was done by Stearns et al. (1992). It

Physician-Payment Methods and Efficiency 219

reached the conclusion that capitation has a weak impact on volume of services provided: in a health plan for Wisconsin government employees, PCPs moved from FFS to individual capitation. At the same time, the unit price of specialists' services were greatly reduced. Paradoxically, there was an increase in the number of ambulatory visits, but it was visits to specialists that increased, and there is reason to believe that specialists increased their self-referral to compensate for the drop in service rates. The volume of visits to GPs remained unchanged, so the transition to capitation had no effect. It is possible, however, that the experiment was disrupted by another factor: at the same time, the GPs were allowed to share in the profits from reduced hospital expenditures. It is thus possible that they were motivated by this profit-sharing and stepped up their effort in primary care medicine in order to limit hospitalizations. The study in fact shows a perceptible drop in the hospitalization rate (though not in the length of stay). This study is not completely free of bias, because certain PCPs opted out of the plan or refused plan patients (or certain services). Some patients may also have left the plan, for, in addition to the change in physician remuneration, there was a transition to an exclusive HMO approach (an insured person is not managed if he consults outside the panel), whereas the plan previously functioned as a Preferred Provider Organization (PPO, which reduces management of an insured person who consults outside the panel). An experiment offering better control of these voluntary separations was conducted in Denmark (Krasnik 1990). General practitioners in the city of Copenhagen moved from pure capitation to mixed capitation and FFS, while regional/county physicians remained under capitation. The author was thus able to compare changes in volumes in the 'test' group (the Copenhagen GPs) and the 'control' group. There was a significant increase in the volume of services provided by the physicians in the test group with the transition to fee-for-service. This effect is not universal, however, and the Norwegian experience seems less conclusive on the impact of FFS. Using records of contacts by patient and by physician4 for Norway, Kristiansen and Holtedahl (1993) show that FFS increases the volume of home visits, all other things equal (although the patients' state of health was unknown), but that the effect of the payment method is weak compared with the influence of a patient's individual characteristics: the physician remains more motivated by health concerns than by his income.5 This precedence of the clinical over the economic in Norway seems to be con-

220 Michel Grignon, Valerie Paris, and Dominique Polton

firmed by Grytten and Sorensen (2001), who show that physicians paid on a per-service basis do not perform more examinations than salaried physicians, even when one controls for factors that lead physicians to choose a given type of contract. The authors claim that this primacy of clinical concerns is mainly due to strong peer control in Norway, which prevents the physician from taking advantage of his position of superiority over the patient. In addition to the damage caused in terms of average quality, inducement can also be damaging to the fairness of the system. Vulnerable segments of populations (children, seniors, the mentally ill, the poor, linguistic minorities) have more difficulty observing and tracking the quality of the physician's service, so it is tempting for the physician to use this ability to induce demand primarily on members of these groups. It is noteworthy that HMOs ration clinical quality for these groups, while leaving average quality unchanged (Miller and Luft 1997). The Cost-Quality-Fairness Trade-off: The Physician's Capacity to Increase Prescriptions Referrals and prescriptions are important components of the physician's practice and of total health care costs. In France, 33 per cent of visits to a physician are the result of a recommendation from another physician. In the United Kingdom, 5 per cent of PCP visits lead to a referral (Coulter 1992), and there is a large variation in physicians' practices which cannot be explained by patient traits (summarized by Scott 2000). A simple but sound theory of physician behaviour is that payment via capitation or salary encourages the physician to transfer his own work to others and increase the volume of prescriptions so as to cut down on unremunerated effort (time spent with the patient, listening, auscultation). This prediction has wide empirical confirmation. The study by Krasnik et al. (1990) shows that, when GPs changed to FFS, their rates of referral and prescription dropped substantially. This study uses a control sample whose payment method remained unchanged over the same period. International comparisons suggest that less time is spent in consultation when payment is by capitation: according to Sandier (1990), while the French GP, paid per service, devoted an average of fifteen minutes (in 1979) to each consultation, the English GP, paid by capitation, was

Physician-Payment Methods and Efficiency 221

content with eight minutes (in 1987). However, in a homogeneous context and using individual data, there is no confirmation that capitation or salary reduces consultation time: Kristiansen and Mooney (1993), using Norwegian data, even find that GPs paid per service spend less time in consultation than those on salary, although the difference is not significant. One way to prevent capitation from raising the cost of prescribed care is to make the prescriber responsible for a patient's total care budget. This is a form of vertical integration of care, assigned to the PCP; we will outline it below. In Search of the Optimum Method of Payment: Lessons from a Variety of Experiences The question of incentives associated with payment methods has received growing attention since health care systems have been faced with major issues of balancing cost, quality, and fairness. Through general reforms or local experiments, various solutions have been tried in an attempt to reconcile these potentially contradictory objectives by means of the methods of compensating professionals. These solutions will now be reviewed, with distinctions down among a number of approaches. One approach tries to combine various incentives within individual physician remuneration, using blended payment systems that include lump-sum, capitation, and FFS. Specific financial incentives may be incorporated into these, the most innovative being mechanisms for sharing in financial or health outcomes (performance-based payment). A second approach deals with different financial incentive systems at the individual and collective levels of the medical profession. The conflicts of interests resulting from this second approach are such that it lends itself more to smaller rather than large groups of professionals, and so, has led to a third approach, that of making physicians' groups responsible for reconciling financial and quality objectives. A final approach to the best cost-quality ratio is to combine financial incentives with incentives of other kinds, such as information provided to the physician, practice benchmarks, and different tools for the dissemination of these benchmarks. The collegial dimension of the medical profession can facilitate change in practices, and this affords a means of potentially gaining the interest of the groups or networks mentioned above.

222 Michel Grignon, Valerie Paris, and Dominique Polton

Incentive Combinations within Individual Physician Remuneration Before turning to mixed-payment systems, it should be noted that no method of payment has emerged as an overall alternative to the standard three: FFS, capitation, and salary. Episode-of-care payment, transposed from hospital payment to diagnostic-related groupings (DRG), was considered by medicare in the United States in the mid-1980s but seems to have had few concrete applications in ambulatory medicine. It can be envisaged for technical care in acute episodes, for example, a surgeon receiving comprehensive remuneration for pre-op, surgery, and post-op (Robinson 2001); and it has been studied in cardiology, for example, coronary bypasses (Showstack et al. 1987). But extending this mode of payment to all medical activities raises numerous technical problems. In terms of practical applications, therefore, innovation and inventiveness have mainly developed around the manner of organizing different elements of payment and using marginal incentives for a minor percentage of remuneration. One interesting example of mixed payment is that provided to British general practitioners, whose compensation includes (Red Book 2002) • a basic allowance to cover a fraction of operating costs, in the form of a sliding lump-sum per patient based on the number of patients. An additional allowance offsets extra costs of practising in a rural area. Top-ups are also paid based on the physician's experience ('seniority' fees). • a capitation per patient on the GP's roster, with three levels according to age (under sixty-five, sixty-five to seventy-four, and seventyfive and over); patients in disadvantaged areas also give rise to an additional capitation reflecting their greater needs. • FFS for night visits, vaccinations, emergency consultations for colleagues' patients, and minor surgery, as well as more comprehensive specific fees for monitoring contraception, child health, and maternity services. • target payments for cervical cancer screening (payment depends on the rate of women who have had a smear in the previous five years) and for immunization of two-year-old and pre-school children. The portion independent of level of service makes up 65 per cent (50 per cent for capitation and 15 per cent for basic allowances); FFS

Physician-Payment Methods and Efficiency 223

accounts for 25 per cent and target payments for 10 per cent. In addition, U.K. physicians working in poorly served rural areas can dispense medications to their patients and receive specific fees for this service. Have incentive payments had the results expected? In fact, it appears that activities paid on a per-service basis have increased most of the time, though not always (for example, the proportion of physicians agreeing to monitor pregnancies to term decreased from 1960 to 1980). In any case, it is not possible to apportion within this increased activity the role of financial incentives and that of other factors related to the general evolution of professional practices and patient demand (Whynes and Baines 1998). While in the National Health Service (NHS) certain productivity incentives have been added to what were originally lump-sum payments, we are now seeing the introduction of packages or lump-sums into FFS systems. For example: • in Quebec, for hospital specialists, mixed remuneration has taken the form of introducing a lump-sum portion and, in return, paying for specific services at a fraction of their rate and eliminating billing for related services. • In France, GPs who volunteer to be 'referring physicians' receive an annual lump-sum of forty-six Euros for general patient monitoring, in addition to billing per service (although this option has been chosen by only 10 per cent of GPs and 1 per cent of patients). The most innovative forms of blended payments incorporate elements of sharing in the outcomes (target payment). Those outcomes can relate to health matters (as in the British example above), but they can also be financial, that is, associated with cost control. Such incentives are used by 'managed care' organizations in the United States to reduce specialist referrals and use of hospitals. They are generally based on a formula that gives the physician an interest in the organization's financial performance. What assessment can be made of such outcome-sharing experiments? Interest in Financial Outcomes

The 1991 reform in the United Kingdom ('fundholding') gave British general practitioners a financial interest in medication and programmed-surgery expenditures for patients on their roster: the volunteer GP was assigned a budget at the start of a period for all his patients

224 Michel Grignon, Valerie Paris, and Dominique Polton

and could retain any year-end surpluses. Theoretically, such surpluses are not income, since the GP has to reinvest them in his practice, but in reality it is difficult to separate the physician's personal income from the revenue of his practice. Evaluation of this experiment has shown that physicians did indeed realize savings on their medication budgets, but not much on programmed surgery, one of the reasons for this being the quasi-monopoly held by British hospitals (Coulter 1995). A study by Debrock and Arnould (1992) measured the effect of financial incentives to the physician by controlling for existing group incentives. Their conclusions for the thirty-five Illinois managed care organizations studied between 1985 and 1987 were that individual financial incentives reduced the hospital admission rate by 16 per cent and cut the number of consultations per client by half. While giving physicians an interest in limiting expenditures seems an effective tactic, the conflict of interests it can raise between professional ethics and the financial interest of the practitioner is negatively perceived by the profession, and above all by the general public: • A survey of 766 U.S. physicians, 58 per cent of whom were subject to group or individual incentives, showed that 57 per cent of practitioners felt pressure to limit referrals and 17 per cent thought that this could negatively affect the quality of care management (Grumbach et al. 1998, quoted by Armour et al. 2001). This survey also gives an idea of the magnitude of such bonuses: according to the physicians surveyed, the bonus accounts for about 7 per cent of their median income. • Although not particularly sizeable, these financial incentives are raising concerns among patients, who fear that they may limit the professional autonomy of their physician. There are few empirical indications of the extent of this concern, apart from many legal actions concerning financial incentives of physicians. In 1996 U.S. legislation set certain limits on contracts between managed care organizations and physicians and made their publication mandatory. According to Hall et al. (2002), the publication of incentives has the effect of slightly increasing patient confidence if the physician is paid by capitation, but it has no effect if he is paid per service. Many authors, insurers, and HMOs now think that performance incentives must be used to improve quality by focusing on health as opposed to financial objectives. According to Rosser and Kasperski

Physician-Payment Methods and Efficiency 225

(2002), Ontario physicians say they are in favour of a transition to capitation provided that it is mitigated by fees for certain services such as being on call, and by premiums for attaining prevention objectives. Interest in Health Outcomes

This does not directly concern health objectives - a seductive but not practical idea - but rather the management of the population concerned. Prevention, both primary (immunization) and secondary (screening), is its most frequent domain of application. The example from the United Kingdom cited above seems to show that these 'target payments' have a positive impact: according to the British Department of Health (quoted by Scott and Hall 1995), the proportion of GPs reaching the objective set rose in five months, for vaccination, from 55 per cent to 70 per cent (objective: a vaccination rate of 90 per cent or 70 per cent) and for smears, from 53 per cent to 67 per cent (objective: for 80 per cent of women from twenty-five to sixty-four to have a smear within five years). Of course, this simple observation of before/after results cannot control for other possible explanations and categorically claim that the increase is due to financial incentive. That is the point made by Armour et al. (2001) regarding another experiment in Northern Ireland, involving childhood immunization: there again, while there was a substantial increase in the vaccination rate during this period, the pure effect of this measure is not distinguished from other concomitant effects. A bonus experiment to improve cancer screening, set up by a managed care organization for people on Medicaid, has been evaluated (Hillman et al. 1998). A six-month bonus ranging from 10 to 20 per cent of capitation for all female clients was paid to the six medical practices achieving the highest screening rate. Fifty-two practices were randomly assigned to an intervention group (with incentives) or a control group. The study revealed no significant difference in screening rates between the two groups. However, positive conclusions were drawn from an incentive experiment for flu vaccination for people over age sixty-five involving fiftyfour New York practices, which was evaluated via a controlled test (Kouides et al. 1998). In addition to remuneration for the vaccine ($8), the physicians received $0.80 per vaccine beyond a vaccination rate of 70 per cent for persons sixty-five years of age and over and $1.60 for a rate over 80 per cent. The immunization rate in the group receiving the bonus (69 per cent) was six points higher than that of the control

226 Michel Grignon, Valerie Paris, and Dominique Polton

group, and 52 per cent of the practices receiving the bonus had an immunization rate of 70 per cent, versus 44 per cent in the control group. An analysis controlling for a number of environmental factors suggests to the authors that the introduction of bonuses served to boost the immunization rate of the elderly by 7 per cent. A recent trend seems to be to tie the bonus to various quality indicators. In one experiment done by Aetna (Hanchak 1997), groups of contracting physicians received bonuses based on their score on a qualityof-care index incorporating a number of aspects, including patient satisfaction, appropriateness of treatment, and efficiency. The authors concluded that there was a positive impact on all aspects of quality (2 per cent decline in the Cesarean rate, 25 per cent decline in average length of hospital stay, 85 per cent increase in biopsies, and a certification rate approaching 95 per cent at the end of the program) except patient satisfaction, which does not seem to have improved. Armour et al. (2001) conclude from a review of the literature that the effects of incentives are contradictory and do not allow for any easy final verdict. They also note that studies rarely deal with volume effects and quality effects at the same time, and in this they see a recognition of some kind of trade-off between cost containment and quality. Different Incentive Combinations at the Individual and Group Levels A second approach is not to counterbalance different incentives within individual remuneration but to offset individual incentives with collective incentives. This is illustrated, for example, in the limiting of the overall envelope systems associated with FFS. Germany opted for such a system in the early 1980s, and other countries or regions have adapted the same principle, using varying methods suited to their own context. German physicians are paid per service but are bound by collective moderation: fees-for-service for all physicians in a land are capped, and the amount is distributed among practitioners by the physicians' unions proportionally to the activity of each, expressed in points on a list of services. In other words, the unit value of the point is not known in advance but is estimated after the fact by dividing the total envelope by the accumulated number of points of all physicians (Busse 2000). The direct remuneration of practitioners through the funds in the form of a cumulative payment for a given period (in Germany, per quarter) affects the operational nature of such a system. However, cer-

Physician-Payment Methods and Efficiency 227

tain systems in which these conditions are not met have also introduced expenditure caps (Quebec, France, the United States for Medicare): the difference is that the unit rate cannot be automatically adjusted in real time but must be decided - and negotiated - for the following year, based on the volume of services actually delivered. Obviously, this method is likely to be less effective in terms of strict adherence to the budget envelope. In theory, this type of mechanism leads to a conflict, for each physician, between the collective interest in moderating volumes (therefore achieving a higher point value and a level of income that is unchanged overall with a lower output of work) and the individual interest, which continues to be to maximize one's own volume of activity at a given unit price. Hence, it encourages 'opportunistic' behaviours, with each physician thinking that he will be able to profit personally from collective moderation if he raises his volume of activity, something that ultimately leads to inflated volumes, contrary to the target objective. Clearly, this is related to the dilution of responsibility within a large collective. Each physician on his own has no chance of affecting the unit value through virtuous behaviour; on the contrary, he has every interest in being this 'opportunistic' element in the great anonymous collective. We shall return to the question of the size of physicians' groups later. The theoretical postulates of systems of this type - ongoing inflation of volumes and a correlative decline in unit prices - seem indeed to be confirmed by the observations in Germany (even though one can debate the proportion of the increased volumes that is to be attributed to opportunistic physician behaviour or to other factors: demand, technological progress, etc.). So a complementary mechanism is needed to offset this incentive to individual behaviours that are too production-oriented. That mechanism can include: • control: this is the case in Germany, where individual activity is examined by the fund physicians. Incidentally, one might think that this control will be more effective since it is organized by the profession itself, but its feasibility depends on the profession's availability to co-manage financial constraints with the regulators, as demonstrated a contrario by the French example. • incentive: as in Quebec, with sliding-scale rates beyond a threshold of service volume; this has the effect of curtailing intensive activities.

228 Michel Grignon, Valerie Paris, and Dominique Polton

The Physicians' Group as Locus for Integrating Clinical and Economic Objectives In different countries we seem to be witnessing convergence towards a more collective practice of medicine, in various forms - networks, single-discipline, or multidisciplinary groups. Since the mid-1990s, this trend has emerged in such different national contexts as the British NHS and American managed care (Bocognano et al. 1998). In England, the 1997 reform instituting the Primary Care Groups or Trusts (PCG/Ts) made the first spontaneous associations of fundholder GPs a general fact of life; in the more competitive local contexts of the United States, physician groups have developed rapidly to contract with health insurers seeking to build networks, and in extreme cases even to incorporate the insurer's functions themselves. In both cases, the approach is to make these groups into highly structured networks with their own management infrastructure. These groups are sometimes given financial responsibility and delegated management. For example, the British PCTs are remunerated via comprehensive payments, on condition that they distribute this funding among their various constituent medical practices. Following up on the 'fundholder' experiment, budgets allocated include not only physicians' compensation but prescription and hospitalization budgets. The PCTs are thus at once responsible for meeting an overall budget, making decisions between different components of that budget, and distributing fees and prescription budgets among their members. To this end, they have to develop an expertise that allows them to optimize their utilization of resources. For example, in the area of drug prescription, all PCTs monitor their members' prescriptions by means of trend charts provided by the national level, almost all of them have a pharmaceutical consultant, 70 per cent use common prescription-reference systems, and 49 per cent have produced their own formulary (Wilkin, Gillam, and Coleman 2001). This example illustrates that the interest of physicians' groups lies in the ability to link financial accountability to clinical governance in a collective that is • managed by professionals, and so has both the technical competence to make the best decisions and a quality code of ethics; and • large enough to acquire a number of expert and assessment tools,

Physician-Payment Methods and Efficiency 229

but at the same time small enough to make each professional feel involved. The enthusiasm for physicians' groups, considered to be in the best position to implement efficient management of care processes, and thus to reconcile quality of care with economic use of resources, has also been strong in U.S. managed care programs. In the most competitive areas, the second half of the 1990s saw very fast development of such medical groups and their management and bargaining skills, both with insurers upstream and hospitals downstream. Within these networks, specific specialty groups can be assigned overall budgets, with the task of distributing them among the individual practitioners they manage. Practitioners can be paid per service, but the group sets up procedures (e.g., peer review, verification of the usefulness of care) which are more operational within a single specialty than in multidisciplinary settings (Robinson 2001b). Within this framework, capitation mechanisms called 'contact capitation' can be implemented - a fairly innovative approach for some specialists: once a patient is referred by a GP to a group physician, the latter receives a capitation for a period of twelve months, during which time he manages the patient as he sees fit. This principle of compensation is more adapted to specialties whose activity is relatively homogeneous (cardiology, gastro-enterology, etc.). It can be adjusted to account for the gravity of cases, for example, by balancing patient numbers for certain diagnoses or services, or by allowing FFS for certain procedures (Franck 1999). This payment method has been used in numerous HMOs, and managed care specialists recently predicted that it has a promising future (Carlson 1997). The Swedish experience seems to confirm that group incentives can have a strong impact, even if individual income is not affected by the change in mode of payment: some hospitals moved to one payment per pathology, while others stayed with the fixed annual envelope, but in both cases the physicians were salaried hospital staff. The result is that hospital physicians paid per pathology significantly reduced lengths of stay, even though the total cost in these hospitals did not decrease (Forsberg, Axelsson, and Arnetz 2001). The experiences described above entail formalized networks with substantial delegation of responsibility. But the idea of promoting a more collective practice of medicine in the form of networks or groups is also developing in other systems, in a less structured fashion (France,

230 Michel Grignon, Valerie Paris, and Dominique Polton

Germany, and some Canadian provinces with experimentations in family physician groups). Here the objective is rather to improve the coordination and continuity of care, something that an isolated FFS exercise does not spontaneously encourage. But the underlying idea is the same - that between the isolated physician and the profession as a whole, the locally established group of physicians, sharing skills and a common culture, can be a tool for changing practices in the desired direction. However, recent developments in the United States seem to reflect upon the excesses of this trend. California, where the medical group reigns supreme, has experienced many bankruptcies in recent years and is now seeing its groups unravel, with individual contracting picking up momentum. The decline of the groups is happening in tandem with the decline of capitation as a method of payment (Managed Care Outlook 2001): the percentage of groups with at least one capitation contract, which rose from 1998 to 1999 (66 per cent to 72 per cent), fell in 2000 (55 per cent). The factors behind the fragility of the groups and their lack of staying power (diseconomies of scale, difficulty of incorporating insurance perspectives, demand for free choice of patients, etc.) aptly illustrate that the reconciliation of cost containment and quality objectives through the virtue of collective 'clinical governance' has its limits, and that the context of extreme financial accountability is perhaps not the most conducive to making the best use of these collective organizations which, in spite of everything, seem to be the way forward in many systems. Non-financial Incentives If physicians' groups can coordinate cost containment and quality of care, it does not simply rests on the alchemy of group and individual financial incentives: it is because the latter are associated with other incentive mechanisms. These are of several types. The first is the development and distribution of clinical protocols and recommendations, found today in most public or private systems. These can be national or local and may provide guidance or be mandatory (e.g., medication formularies). There is abundant literature on the comparative efficacy of different methods of disseminating references (training, interviews, computerized reminders) for their implementation by physicians, following the recommendations of the Cochrane Collaboration on Effective Professional Practice. It is beyond our purview here to reproduce and comment on this literature, but we can highlight its main lessons,

Physician-Payment Methods and Efficiency 231

as summarized by Durieux (1999): no studies have found that distribution of standards via print material or ongoing medical training has a significant impact; however, distribution by opinion leaders or, better yet, by visits from colleagues does have a significant impact on physicians' practices. Consistent with this finding, the evaluation of individual practice, particularly in terms of these reference sources or compared with colleagues (release of information to the physician, peer review, etc.), is a tool for changing individual behaviours. The care network or physicians' group is a preferred forum for applying these incentives, for the reasons given above. This is especially true since these networks select their members based on their support for the group's values - clinical, economic and cultural. A further step can be taken in multiple-insurer systems with selective contracting (where the payer is able to manage only a limited number of professionals), which is obviously a powerful non-financial incentive. Conclusion: What Are the Lessons for the Canadian Context? Canada ranked fifth among OECD countries in 1998 for health spending as a percentage of GDP, and the interim report of the Commission on the Future of Health Care in Canada notes that the problem lies more with the unstable evolution of these expenditures than with their absolute level (OECD Health Data 2001; Commission 2002). Also in 1998, Canada was in seventh place in terms of per capita ambulatory-care expenditures, with a level ($333 in Purchasing Power Parity [PPP]) representing a third of that in the United States and 80 per cent of the German level but 40 per cent higher than the French level. The price trend for physician services in Canada over the 198596 decade matches that of other OECD countries (excluding the United States), and the increase in volumes, which was quite high at the outset compared to other countries, seems to have been curbed from 1995 onward. So Canada does not appear to have a serious problem of controlling health care costs.6 However, the commission's interim report underscores Canadians' great dissatisfaction with the accessibility and quality of their health care system. To resolve these problems, must FFS necessarily be called into question? We have seen that payment by capitation can pose a risk for quality of care and accessibility, particularly since the physician need not fear competition, should there be a shortage of doctors, for example.

232 Michel Grignon, Valerie Paris, and Dominique Polton

On the other hand, by modulating the physician's compensation according to the type of population served, capitation can be a means of addressing specific problems such as the uneven geographic distribution of physicians. The current debate on the future of the Canadian health care system certainly provides an opportunity for discussing payment methods. This is especially so given that the latest survey by the Canadian Medical Association reveals that, although 62 per cent of Canadian doctors derive almost all of their income from FFS, only 37 per cent prefer that method of payment to other alternatives (Martin 2000). Specific Questions What is the impact of physician-payment mechanisms on both the utilization of physician services and the overall utilization and costs of the health care system? How does FFS payment for physicians influence utilization and costs compared to other mechanisms, and where are the observed differences? There is no real alternative to the canonical payment systems of FFS, capitation, and salary. Recent innovations are more in the nature of marginal adaptations. Current knowledge of the impact of payment methods on health care costs is summarized, according to their degree of inclusiveness. FFS allows the physician to increase the volume of medical services provided to the patient, and so limits the scope of modes of regulating by unit prices. Conversely, lump-sum forms of payment (capitation and salary) limit volumes but can adversely affect the quality of care provided, limit access to care for vulnerable populations, and have unexpected consequences on the level of spending. The extent of the impact on quality and access is critically dependent on the patient's ability to observe the physician's effort, an ability upon which it is difficult to make any empirical statement. The unexpected effect on costs comes in the form of prescriptions (a lump-sum method of payment encourages the PCP to refer and prescribe more) and longterm effects (if the physician reduces service in the short term, the system can pay for the consequences of this in the long term). The concrete responses are designed to combine mechanisms so that weaknesses offset each other. Payments with a health outcome (intermediate objectives) serve to increase the capacity of payment by capitation to account for the diverse nature of the physician's work, and to combine public health objectives on the one hand with access and quality objectives on the other. Delegation of responsibility to the med-

Physician-Payment Methods and Efficiency 233

ical profession (the regulator sets an envelope which is then managed in a decentralized fashion) can come up against opportunistic behaviours if the decentralization is too weak and conflicts of interests if it is too strong (physician groups that are too small). After being described as the solution of the future, delegation to physician groups now seems to be marking time, to the benefit of non-financial incentives focused on patient and doctor information. What is the evidence on the impact of changes in payment mechanisms or incentives on utilization and costs? Where have the most dramatic changes taken place, and how?

Quantitative data are rare, opportunistic in nature (coming from experiences carried out for other purposes), and always biased. Noteworthy exogenous changes have been made in the United Kingdom, with the introduction of fundholding and PCTs, in Denmark (Copenhagen experiment), and in Germany (envelope and floating points). In the United States, there are numerous reduced-sample experiments (a clinic or a plan) which are difficult to generalize. The reform of Medicare rates has also been a learning ground, but the main problem is that it mixes level and method of remuneration. Empirical quantitative data suggest that the transition from capitation to FFS increases volumes and reduces prescriptions, although there is no real proof of a reciprocal effect. The quality assessment is more ambiguous, even though patients seem dissatisfied with the lump-sum payment method. The target-payment experiments seem moderately conclusive in terms of the impact of incentives on the desired outcome. In addition to quantitative evaluations, we have qualitative assessments centred on stakeholders' opinions. Tracking the evolution of physicians' groups in the United States and their internal arrangements is yielding numerous results, which, however, are difficult to generalize to single-payer contexts. What non-monetary incentives affect operate?

utilization and costs? How do they

Two main innovations are suggested. One is to assist physicians by providing them with validated and synthesized information about clinical protocols and efficiency, in particular, replacing promotion from the medical-products industry; the other is to help physicians and patients get some idea of variations in practices. This second type

234 Michel Grignon, Valerie Paris, and Dominique Polton Figure 7.1 Evolution of prices and volumes* of physician services, 1985-96

^Volumes correspond to ambulatory-care expenditures deflated by the corresponding price index. Source: OECD Health Data 2001.

Physician-Payment Methods and Efficiency 235 of incentive is aimed at strengthening patients' control of the physician's effort, which can either increase variability (adapting practices to demand) or decrease it (standardization). This is not simply a matter of finger pointing: physicians may also find it advantageous to compare their practice to that of their colleagues. The impact of these nonmonetary incentives is strongly context-dependent, particularly the level of organization of the profession and the ability of physicians to show respect for standards of good practice among their peers.

NOTES 1 Increasing quantity can mean changing the structure of services offered by increasing the proportion of better-paid services. Taking the example of Quebec between 1977 and 1983, Rochaix (1993) shows that freezing fees has led to a substantial increase in the proportion of elaborate consultations. 2 A salaried physician could be sanctioned by his employer. With capitation, he will be sanctioned by the patient if inferior quality is noticed and if professionals are in competition to attract clients. 3 Referral is the action whereby one professional refers a patient to another professional. 4 These data allow one to control bias in comparative studies by capturing the effects linked to each individual; however, because the data are complicated to obtain, samples are less often representative. 5 Another study on the same set of data appears to show a strong and significant impact of fee-for-service on the volume of urine tests (Kristiansen and Hjortdahl 1992). But Scott and Hall (1995) point out that this effect is measured without correcting for the bias related to the non-independence of the observations (many contacts per patient are observed). 6 This verdict is based on data available at the time this article was written (in 1996 or 1998, depending on the series). Canadian experts might possibly come to another conclusion with more recent data on trends in expenditures, prices, and volumes generated.

REFERENCES

Armour, Brian S., et al. 2001. The effect of explicit financial incentives on physician behaviour. Archives of Internal Medicine 161(10): 1261-6. Bocognano, Agnes, et al. 1998. Mise en concurrence des assurances dans le domaine

236 Michel Grignon, Valerie Paris, and Dominique Polton de la sante: Theorie et bilan des experiences etrangeres. Paris: CREDES-Institut national de la statistique et des etudes economiques. Brunelle, Yvon, and Howard Bergman. 1999. Capitation, efficience et rationnement. Ruptures 6(2): 270-88. Busse, Rheinardt. 2000. Health Systems in transition-Germany. London: European Observatory on Health Care Systems. Carlson, Bob. 1997. Capitation: State of the art. Available at http://www. managedcaremag.com/archives. Commission on the future of health care in Canada. February 2002. Shape the future of health care - Interim report. Coulter, Angela. 1995. Evaluating general practice fundholding in the United Kingdom. European Journal of Public Health, 5: 233-9. Davidson, Stephen, M. Manheim, S. Werner, et al. 1992. Prepayment with office-based physicians in publicly funded programs: Results from the children's Medicaid program. Pediatrics 89(4): 761-7. Debrock, Larry, and Richard, J. Arnould. 1992. Utilization control in HMOs. Quarterly Review of Economic Finance 32: 31-53. Durieux, Pierre. 1999. Actions de diffusion, d'information et de formation medicale continue, in Comment ameliorer les pratiques medicales? Approche comparee Internationale. Paris: Flammarion et Institut d'Etudes des Politiques de Sante. Emery, J.C. Herbert, Chris Auld, and Mingshan Lu. 1999. Paying for physician services in Canada: The institutional, historical and policy context. Working paper 99-6. Institute of Health Economics. Epp, Michael, et al. 2000. The impact of direct and extra billing for medical services: evidence from a natural experiment in British Columbia. Social Science & Medicine 51: 691-702. Escarce, Jose. 1996. Using physician practice cost functions in payment policy: The problem of endogeneity bias. Inquiry 33: 66-78. Evans, Robert G. 1974. Supplier induced demand: Some empirical evidence and implications, in M. Perlman, ed., The economics of health and medical care. Edinburgh: Macmillan, 162-73. - 1998. New bottles, old wine: right and wrong in physician supply. CAM] 158: 757-9. Ferrall, Christopher, Allan W. Gregory, and William G. Tholl. 1998. Endogenous work hours and practice patterns of Canadian physicians. Canadian Journal of Economics 31(1): 1-27. Forsberg, Ewa, Runo Axelsson, and Bengt Arnetz. 2001. Financial incentives in health care: The impact of performance-based reimbursement. Health Policy 58(3): 243-62.

Physician-Payment Methods and Efficiency

237

Franck, Clifford R. 1999. Can the pie be divided more fairly? Contact capitation for specialists 5(2). Available at http://www.cost-quality.com/5,2art3.hhnl. Gerdtham, Ulf-G, Jes Sogaard, Fredrik Andersson, and Bengt Jonsson. 1992. Econometric analysis of health expenditure: A cross-section study of the OECD countries. Journal of Health Economics, 11: 63-84. Gerdtham, Ulf-G, Bengt Jonsson, M. MacFarlan and H. Oxley. 1998. The determinants of health expenditure in OECD countries, in Peter Zweifel, ed., Health, the medical profession, and regulation. Boston: Klower Academic Publishers. Gillett, James, Brian Hutchison, and Stephen Birch. 2001. Capitation and primary care in Canada: Financial incentives and the evolution of health service organizations. International Journal of Health Services 31(3): 583-603. Glied, Sherry. 1998. Payment heterogeneity, physician practice, and access to care. American Economic Review 88(2): 127-31. Gosden, Toby, Frode Forland, Ivar Sombo Kristiansen, et al. 2001. Impact of payment method on behaviour of primary care physicians: A systematic review. Journal of Health Services Research and Policy 10(1): 44-55. Grumbach, Kevin, Dennis Osmond, Karen Vranizan, et al. 1998. Primary care physicians' experience of financial incentives in managed care systems. New England Journal of Medicine, 339:1516-21. Gryrten, Jostein and Rune Sorensen. 2001. Type of contract and supplierinduced demand for primary physicians in Norway. Journal of Health Economics 20: 379-93. Hall, Mark, A., Elizabeth Dugan, Rajesh Balkrishnan, et al. 2002. How disclosing HMO physician incentives affects trust. Health Affairs 21(2): 19-27. Hanchak, N.A. 1997. A performance-based compensation model for obstetricians/gynecologists. Clinical Obstetrics and Gynecology 40: £37-45. HCFA 2001, Study on pharmaceutical benefit management, PricewaterhouseCoopers. Available at: http://cms.hhs.gov/researchers/reports/2001/ cms.pdF. Health Canada. 2001. Canada Health Act, 2000-2001, Annual Report. Ottawa: Health Canada. Hemenway, David. 1995. Financial incentives for childhood immunization. Journal of Policy Analysis and Management 14:133-9. Hickson, B., A. Altemeier, and M. Perrin. 1987. Physician reimbursement by salary or fee-for-service: Effect on physician practice behaviour in a randomized prospective study. Pediatrics, 80: 344-50. Hillman, Alan L., Kimberly Ripley, Neil Goldfard, et al. 1998. Physician financial incentive and feedback: failure to increase cancer screening in Medicaid managed care. American Journal of Public Health 88:1699-1701.

238 Michel Grignon, Valerie Paris, and Dominique Polton Hutchinson, James M., and Robert N. Foley. 1999. Method of physician remuneration and rates of antibiotic prescription. CMAJ160:1013-117. Jelovac, Izabela. 2001. Physicians' payment contracts, treatment decisions and diagnosis accuracy. Health Economics 10: 9-25. Josseran, L., G. King, and J. Chaperon. 2000. Les Primary Care physicians americains. Journal d'economie medicale 18(6): 375-87. Kouides, R.W., N.M. Benett, B. Lewis, J.D. Cappuccio, W.H. Barker, and P.M. LaForce. 1998. Performance-based physician reimbursement and influenza immunization rates in the elderly. American Journal of Preventive Medicine 14: 89-95. Kralewski, John E., et al. 2000. The effects of medical group practice and physician payment methods on costs of care. Health Services Research 35(3): 591613. Krasnik, Allan, P. Groenwegen, and Poul A. Pedersen. 1990. Changing remuneration system: effects on activity in general practice. BMJ 300:1698-1701. Kristiansen, Iwar S., and K. Holthedahl. 1993. The effect of the remuneration system on the general practitionner's choice between surgery consultations and home visits. Journal of Epidemiology and Community Health 47:481-4. Kristiansen, Iwar S., and Gail Mooney. 1993. The general practitioner's use of time: Is it influenced by the remuneration system? Social Science & Medicine 37:393-9. Kristiansen, Iwar S., and Per Hjortdahl. 1992. The general practitioner and laboratory utilization: Why does it vary? Family Practice 9: 22-7. Managed Care Outlook. 2001. 'Interest in capitation back off.' Available at http://www.managedcaremag.com/archives Me Guire, Thomas G. 2000. Physician agency, in Anthony J. Culyer and Joseph P. Newhouse, ed., Handbook of Health Economics, 1:463-536. Amsterdam: Elsevier Science. Martin, Shelley. 2000. CMA survey shows fee-for-service not dead yet. Canadian Medical Association Journal 163(5): 601. Miller, Robert H., and Harold S. Luft. 1997. Does managed care lead to better or worse quality of care? Health Affairs 16(5): 7-25. Mulligan, Pamela. 2002. La capitation: une direction mal avisee dans la reforme des soins de premiere ligne. Le Medecin defamille canadien 48: 244-6. Namovicz-Peat, S., J. Swann, and K. Tureski. 2001 HMO & PBM strategies for pharmacy benefits: Recent results, current practices, future plans. Atlantic Information Services. Washington: Atlantic Information Services. Pauly, Mark, and Mark A. Sattertwaite. 1981. The pricing of primary care physicians' services: A test of the role of consumer information. Bell Journal of Economics 12: 488-506.

Physician-Payment Methods and Efficiency

239

Philips, Karen, and William R. Swan. 1996. Health care system in transition: Canada. Preliminary version. Copenhagen: World Health Organization, Regional Office for Europe. Red Book 2002. Available at http://www.nhs.Uk/redbook/2.htm Rice, Thomas. 1993a. A model is only as good as its assumptions: A reply to Peele. Journal of Health Economics 12: 209-11. - 1993b. Demand curves, economists, and desert islands: A response to Feldman and Dowd. Journal of Health Economics 12: 201-4. Robinson, James C. 2001. Physician organization in California: crisis and opportunity. Health Affairs 20(4): 81-96. - 2001. Theory and practice in the design of physician payment incentives. Milbank Quarterly 79(2): 148-77. Rochaix, Lise. 1993. Financial incentives for physicians: The Quebec experience. Health Economics 2:163-76. Rosser, Walter, and Jan Kasperski. 2002. En faveur d'une remuneration mixte. Le Medecin defamille canadien 48: 247-9. Sandier, Simone. 1990. Health Services Utilization and Physician Income Trends, in Health care systems in transition: The search for efficiency. Paris: OECD. Scott, Anthony. 2000. Economics of general practice, in Anthony J. Culyer and Joseph P. Newhouse, ed., Handbook of Health Economics 1:1175-1200. Scott, Anthony, and Jane Hall. 1995. Evaluating the effects of GP remuneration: Problems and prospects. Health Policy 31:183-95. Showstack, Jonathan A., Deborah W. Garnick, Kenneth E. Rosenfeld, et al. 1987. Episode-of-care physician payment: A study of coronary artery bypass graft surgery. Inquiry 24: 376-83. Stearns, Sally C., Barbara L. Wolfe, and David A. Kindig. 1992. Physician responses to fee-for-service and capitation payment, Inquiry 29: 416-29. Tai-Seale, Ming, Thomas H. Rice, and Sally C. Stearns. 1998. Volume responses to Medicare payment reductions with multiple payers: a test of the McGuirePauly model. Health Economics 7:199-219. Whynes, David, and Darrin L. Baines. 1998. Income-based incentives in UK general practice. Health Policy 43:15-31. Wilkin, David, Steve Gillam, and Anna Coleman, eds. 2001. The national tracker survey of primary care groups and trusts 2000-1: Modernising the NHS. London: King's Fund; and Manchester: National Primary Care Research and Development Center. Available at: http://www.npcndc.man.uk/publications/ tracker%20report%200001 .pdf

This page intentionally left blank

PART THREE CHANGE AND THE PUBLIC

This page intentionally left blank

8 Public Involvement in the Development of a Health Care Vision HARLEY D. DICKINSON

We live in turbulent times. New scientific knowledge, technological innovations, and organizational reforms render traditional truths, occupational hierarchies, and established social roles obsolete at increasingly rapid rates. The perpetual transformation of the health care system, for example, is accompanied by calls for greater public involvement in a variety of decision-making and service-delivery activities. In responding to these calls, the public - in the roles of patient, consumer, volunteer, and citizen - is imbued with new rights and responsibilities relative to governments, service providers, and others. Changes of this sort are contentious because they affect established needs and interests (Lalonde 1974; Epp 1986; Charles and DeMaio 1993; Abelson, Lomas, Eyles, et al. 1995; Dickinson 1994,1996, 2002; Lomas 1997; Dickinson and Torgerson 1998-9; Bolaria and Dickinson 2001; Church 2001; OECD 2001). Among these contentious changes are calls for informed and shared clinical decision-making as a way to increase the involvement of the public as patients (Makoul, Arntson, and Schofield 1995; Short 1996; Caress 1997; Charles, Garni, and Whelan 1997; Partridge 1997; Ubel and Lowenstein 1997; Gafni, Charles, and Whelan 1998; Guadagnoli and Ward 1998; Gwyn and Elwyn 1999; O'Connor et al. 1999; Robinson and Thompson 2000; Stevenson, Barry, Britten et al. 2000; Barry et al. 2001; Weston, 2001). Similarly, medical savings accounts (MSAs), a recently proposed way to expand consumer involvement through the application of market principles to the provision and consumption of healthcare services, are seen by proponents as a way to save medicare and by opponents as undermining its foundational principles (Litow and

244 Harley D. Dickinson

Muller 1998; Ramsay 1998; Deber 1999; Gratzer 1999; Stein 2001; McMahon and Zelder 2002). Still another proposed innovation are patient charters, which, proponents argue, enhance system accountability by defining the rights and responsibilities of health care consumers/patients relative to those of service providers and governments. Opponents claim that they simply entrench existing power relations and interests (Flood and Epps 2001; IRPP 2001; Gratzer 2002). Efforts to forge new relationships between government and the voluntary sector are multidimensional. On the one hand, some focus on the relationship between enhanced public participation, increased levels of social capital, and improved population health status. Others maintain that increased reliance on volunteers marks an abandonment of the sick and the transfer of uncompensated responsibility for the provision of care to the community, particularly to women (Armstrong and Armstrong 1996; McDowell 1986; Bolaria 1988; Putnam 2001; Veenstra 2001; Woolcock 2001). On the other hand, some see the negotiation of a new governmentvoluntary sector accord as part of a neo-corporatist redefinition of citizens' rights and responsibilities relative to democratic policy development and implementation. Others see it as contrary to the principles of liberal democracy and a process that will further marginalize the politically unorganized (Crichton et al. 1997; Hunold 2001; Wilensky 1981; Wooley 2001; Reed and Selbee 2001; Good 2001; Brock 2001). Clearly, the public occupies a variety of roles and has numerous competing interests relative to the health care system. It is impossible to compartmentalize completely these social roles or to differentiate clearly the public's various, and sometimes competing, interests (Charles and DeMaio 1993; Lomas 1997). Having said that, however, in this paper I concentrate primarily on efforts to involve the public in its role as citizen. In doing so, I address the following general questions: • How can and should the public be involved in setting overall directions for the health care system? • What mechanisms are most effective in ensuring that the system reflects public values and principles? In my answers to these questions, I argue that deliberative publicconsultation procedures have several features that recommend their expanded use for the formation of both informed public opinion and the political will to act on it. The creation of informed public opinion

Public Involvement in a Health Care Vision 245

and its conversion into actionable political will are the means by which the public, acting as citizens, can be most meaningfully involved in developing and maintaining an overall vision for the health system consistent with its values and principles. Regional health authorities (RHAs)1 are a promising institutional site for linking deliberative public-consultation procedures to policy and operational levels of decision making. I begin by discussing two dimensions of the democratic deficit and the crisis of legitimacy faced by contemporary societies. The first is a tendency towards elitism in political decision making inherent in the institutions and procedures of representative democracy. The second relates to medical dominance as a defining feature of Medicare and as a barrier to public involvement in health care decision making in particular. Following this, I identify regionalization interaction as a strategy to involve the public in deliberative consultation related to health care governance and service delivery, and thereby to facilitate health system reform. I conclude that the results to date are equivocal but promising. The Uninvolved Public, Medical Dominance, and the Democratic Deficit Contemporary democracies are faced with declining political party membership, decreasing voter turnouts, widespread cynicism, and lack of trust in the institutions of representative government (Dalton, Burklin, and Drummond 2001). In combination, these features constitute a democratic deficit. The democratic deficit is related to a political crisis of legitimacy. Both are rooted in a belief that those who make and implement policy do not adequately represent the interests of the general public (Habermas 1976). The democratic deficit and the associated legitimacy crisis exist at two levels. The first is in the context of the institutions and processes of representative democracy in general. The second is in the context of specific policy domains such as health care. In this section, I discuss both. I also discuss public participation as a means to reduce the deficit and resolve the crisis. Paradoxically, representative democracies are characterized by a structural tendency towards elitism and the disempowerment of the general public. This occurs through, and to the extent that, the political role of citizens is limited to periodic voting, a process described as 'a

246 Harley D. Dickinson

regulated acclamation procedure for alternately appointed elites to exercise power' (Habermas 1971, 68). To the extent that this is an accurate description, it is not surprising that the public is foregoing the opportunity to play its appointed role. The generally agreed upon corrective to the democratic deficit is more democracy. Several influential analysts, for example, advocate the introduction of direct democracy as a means to increase citizen control of policy making and implementation (Arnstein 1969; Feingold 1977; Charles and DeMaio 1993). Others do not make the assumption that the goal of public participation is direct democracy and citizen control (Connor 1986; OECD 2001). Connor (1986), for example, argues that the goal is conflict avoidance and, failing that, conflict resolution while leaving intact representative forms of decision making. Similarly, the OECD (2001) specifies three strategies for increasing public involvement: informing, consulting, and actively engaging the public in proposing policy options and participating in policy dialogue. In this model, authority to make final decisions remains with elected governments (OECD 2001, 2). From the perspective of proponents of direct democracy, the forms of public participation advocated by these models are mired in tokenism (Arnstein 1969) and may contribute to a deepening of the democratic deficit, not its reduction. Popular support for direct democracy versus reformed representative democracy is split. The strongest support for direct democracy is among citizens at the periphery of politics - the less interested, the less informed, and adherents of extreme parties with strong populist and nativist tendencies. This feeds fears about direct democracy and the tyranny of the majority (Dalton, Burklin, and Drummond 2001,151). It is not just the politically marginalized, however, who support more participatory forms of democracy. According to opinion polls, most people in Western democracies favour moving towards a more participatory style of democratic government (Dalton, Burklin and Drummond 2001,145). This is not to say that all are in favour of direct democracy and citizen control. Abelson, Lomas, Eyles, et al. (1995) report that participants in their deliberative polling study favoured different decision makers for different types of decisions and different decision contexts. Relative to devolved authority, for example, participants favoured an advisory role for interested citizens, with decisionmaking authority assigned to traditional decision-making bodies, including elected officials and experts. This last point draws attention to the practice of delegating decision-

Public Involvement in a Health Care Vision 247

making authority to third parties. This has been referred to as government-by-proxy (Salamon 1981; Kettl 1988; Brudney 1990; Reed and Selbee 2001). In the case of health care, control over decision making and service delivery has historically been delegated to the medical profession. It has been argued, in fact, that relative to health care decision making, the medical profession constitutes a 'state within a state' and, in particular circumstances, a threat to democracy (Taylor 1960; Badgeley and Wolfe 1967). To the extent that this is true, health system governance is more technocratic than democratic and provides little opportunity for public participation (Freidson 1973; DeSario and Langton 1987). This societal delegation of power to the medical profession occurred at a time when infectious diseases were the main source of morbidity and mortality. Advances in medical science during the late nineteenth and early twentieth centuries enabled the medical profession to prevent and cure many of these illnesses. It was widely believed that the continued development and application of medical science to the diagnosis and treatment of illnesses would result in continued improvement in population health status. In this context, the central task of policy makers was ensuring that individuals had access to physician and hospital services regardless of their ability to pay. Medicare, built on the principles of universality, comprehensiveness, portability, public financing, and administration on a non-profit basis, eventually was instituted as the means to achieve that goal. The medical profession vigorously resisted the introduction of medicare because it was seen as a threat to professional autonomy (Coburn, Torrance, and Kaufert 1983; Coburn 1998; Blishen 1991; Touhy 1999; Taylor; 1960, 1978; Badgeley and Wolfe 1967). As a result of this resistance, the profession secured several concessions from the government that entrenched medical autonomy and dominance by creating a health care system characterized by private practice and public payment (Naylor 1986). The private nature of clinical practice decisions put physicians beyond external scrutiny and management, and, combined with a fee-for-service system of payment that tied physician incomes to fee levels and volume of services provided, created a system with built-in tendencies towards high levels of provider-driven demand and rising costs. By the mid-1970s, and through the 1980s, these inflationary tendencies gave rise to a variety of cost-control initiatives (Armstrong 1997; Fuller 1998; Dickinson and Hay 1988; Weller and Manga 1983; Vayda, Evans, and Mindell 1979), but none of these

248 Harley D. Dickinson

initiatives fundamentally affected the inflationary dynamic of medicare or the extent of medical dominance (Touhy 1999). In recent years, however, several factors have converged to generate a renewed interest in reducing medical autonomy and dominance. These include substantial small-area variations in diagnostic and therapeutic interventions that appear, in many cases, to be related more to physician idiosyncrasies than to medical need (Roos and Roos 1994; Crombie, Cross, and Fleming 1992; Anderson and Mooney 1990), and relatively high levels of medical error (Health Edition 2002c, 2). In combination, these factors have raised doubts about whether medical practice is sufficiently grounded in medical science to warrant the profession's control over the form and content of its work (Lomas and Contandriopoulos 1994). Calls for evidence-based medicine can be seen, in part, as a response to this (Sackett et al. 1998; Gray 1997; Dickinson 1998). Concerns that evidence-based medicine conceals a threat to professional autonomy are evident in the critical response by some clinicians (Grahame-Smith 1995). Other factors, not directly related to the nature and organization of medical practice, are also at play to reduce medical autonomy and dominance. Demographic and epidemiologic shifts, for example, have resulted in a relative decline in infectious diseases and an increase in chronic conditions and accidents as the major sources of morbidity and mortality (Omran 1971; Lalonde 1974). For the most part, the medical profession is unable to cure these conditions and the hospital-centred health care system is not well designed, in terms of effectiveness or efficiency, for the long-term care and management of people with chronic conditions. The demographic and epidemiologic shifts also contribute to a reduction in the knowledge gap between physicians and the general public. Patients and their families often have more knowledge than physicians about how best to manage chronic conditions. Increased levels of education among the population and easy and widespread access to health-related information made available by new information and communications technologies are also seen to contribute to a reduction in the knowledge gap (Blishen 1991). This shift in the knowledge/power base is associated with emergence of patient/consumer rights movements and demands for increased opportunities to participate in health care decision making and service delivery. Organizational and governance reforms such as regionalization have increased opportunities in this regard, as has elucidation of the health-promotion policy paradigm (Epp 1986; Blishen 1991).

Public Involvement in a Health Care Vision 249 Health Promotion: From Access to Health Care to Health for All Increased public involvement in health care decision making and service delivery is at the heart of the health-promotion policy paradigm. The health-promotion framework emerged in the early 1970s as a result of research showing that a relatively small proportion of the population health status was attributable to clinical medicine. Factors outside the health care system, including human biology, the physical and social environments, and individual lifestyle choices were identified as important determinants of health. Relative to these factors, health care services accounted for a relatively small proportion of population health status. It follows that ever increasing expenditures on health care services result in diminishing returns in terms of health status (Evans and Stoddart 1994). Thus, from this perspective, the greatest gains in population health status were to be found in encouraging behaviours that preserved and promoted health and discouraging those that did not. It was expected that identifying and supporting health-promoting behaviours would increase population health status and reduce both the need for, and unnecessary use of, expensive medical care and hospital services (Lalonde 1974; Epp 1986). Critics, however, pointed out that these ideas could be interpreted as a form of victim blaming and abandonment of the sick (Bolaria 1988; McDowell 1986). These concerns notwithstanding, the concept became increasingly entrenched in the policy discourse. By the late 1980s and early 1990s, a series of provincial commission reports recommended a number of common reforms, including increased emphasis on health promotion and decreased emphasis on institutional care; better coordination and integration of services; and increased public participation in policy planning and service delivery. In all provinces, except Ontario, regionalization was adopted as the means to achieve these goals (Angus 1991; Angus et al. 1995; Crichton et al. 1997; Dorland and Davis 1995; Mahtre and Deber 1992). Regionalization: Promoting Health and Cutting Costs Many of the reforms called for in the various reports were not new. Neither was the notion that regionalization is an effective way to achieve them. Saskatchewan, for example, experimented with regionalization as early as 1946 (Feather 1991a,b). Despite this, regionalization was never the dominant mode of system governance and service

250 Harley D. Dickinson

delivery. Following the introduction of Medicare, it was pronounced dead (Taylor 1978). Its resurrection began in the mid-1970s when governments in the Northwest Territories, Quebec, Manitoba, and Ontario (Vayda, Evans, and Mindell 1979; Taylor 1978; Badgeley 1982) regionalized aspects of health care as a way to deal with rising costs. These fiscal concerns were exacerbated by the fact that medicare was not producing the expected health outcomes. Thus, the search was on to find a way to implement health system reforms in the face of powerful provider interests committed to the status quo. Efforts to establish a new balance of power took the form of a neo-corporatist strategy involving extensive public consultation exercises and the formation of new partnerships and collaborations with a wide range of stakeholders (Boudreau 1991a,b). In Canada, interest in the neo-corporatist model was heightened as a result of a 1980 OECD conference on the growing fiscal crisis of the welfare state (Crichton et al. 1997). Wilensky (1981,194), in a paper presented at the conference, argued that countries adopting neo-corporatist models of policy formation were more likely than the least corporatist countries, such as Britain, the United States, and Canada, 'to contain the bargaining power of physicians and other provider groups, and thereby move the [health care] system towards real health outputs/ Regionalization, thus, re-emerged as a neo-corporatist strategy to involve the 'public' in a quest to actualize the health-promotion policy paradigm and to break the 'Medicare Pact' (Lomas 1997d). Cast in these terms, it is clear that the goals of the health promotion policy paradigm are not to be understood only as add-ons to the existing physician-dominated and hospital-centred health care system. Regionalization must also be understood as a means to 'demedicalize' and deinstitutionalize health care. Increased public involvement is necessary (although not sufficient) to form both the public opinion and the political will necessary to bring about the new balance of political power required to implement these reforms successfully. This relates to the previously discussed concerns about the democratic deficit. Policy makers and reformers are concerned that if the public is not involved in, and supportive of, proposed reforms to the health care system, it may not be possible to make the desired changes, or at least it may not be possible to make them without unacceptable political costs. This is a concern to the extent that regionalization and health promotion are seen as threats to medicare.

Public Involvement in a Health Care Vision 251

By creating new planning and management capacities between provincial and local levels of health system governance and service delivery, regionalization is intended to be a means to help establish and mobilize new stakeholders in reform efforts (Dorland and Davis 1995; Lewis 1997; Lomas, Woods, and Veenstra 1997a,b,c; Lomas 1997d; Church and Barker 1998; Lewis et al. 2001). It is a contradictory process, however, that involves both centralization and decentralization of authority and responsibility and, associated with this, a tension between democratic (community control) and technocratic (professional control) of decision making (Rasmussen 2001, 256). Relative to the centralization-decentralization of decision-making capacity, regionalization invariably resulted in the replacement of a relatively large number of local hospital and other facility boards with a smaller number of RHAs. In the early 1990s, for example, the government of Saskatchewan replaced more than four hundred local boards with thirty-two district health boards. This number subsequently was reduced to twelve. The scale of centralization was similar or greater in all jurisdictions that regionalized their health care systems. Critics of regionalization argue that the elimination of local hospital boards has de-democratized health care decision making by reducing the number of opportunities for public involvement. Proponents, on the other hand, argue that it has resulted in enhanced democratic participation, especially in jurisdictions, like Saskatchewan, that created partially elected boards. Most other jurisdictions, however, have appointed boards. Several provinces have indicated their intention to introduce elected or partially elected RHA boards, but so far, Alberta is the only province to have done so. Saskatchewan, on the other hand, replaced its partially elected boards with entirely appointed boards in the most recent round of reforms. This clearly de-democratizes decision making. Although there are contradictory tendencies, regionalization has resulted in increased efforts to increase public involvement in, and to assert lay (public) control over, health care providers and system planning and management. These efforts have taken a number of different forms. I discuss some below. Informing the Public Meaningful public involvement is the best prophylactic and therapeutic response to the democratic deficit. Of course, it presupposes an

252 Harley D. Dickinson

informed public. The most direct way for governments to inform members of the public is to produce and disseminate information for their use. This takes the form either of providing 'passive' access to information requested by members of the public or of taking 'active' measures to disseminate information regardless of public requests (OECD 2001, 2). Standard annual public meetings and reports, along with brochures, newsletters, pamphlets, news releases, and various other print materials, are common ways to disseminate information. Various electronic media are increasingly used for this purpose, too, including cable television, radio, and Internet sites. Providing information to the public is an important way for all levels of government to meet accountability requirements. Certainly, RHAs attempt to engage the public through the provision of information. In Saskatchewan, for example, RHAs are required by law to provide public access to meetings and minutes and to present annual reports and health service information. The provision of information by governments is generally thought to be a foundation stone of representative democracy. There are concerns, however, that governments may provide information in a selfserving fashion, rather than as a means to enhance informed public involvement in policy discussions. When this occurs, it heightens public suspicion and cynicism and contributes to a deepening of the democratic deficit. Thus, it is necessary to have external, arms-length scrutiny of government activities, even ones that seem to be as unproblematic as providing information to the public. Experience also shows that it is insufficient to leave government provision of information on a voluntary basis. Governments often are reluctant to make information too freely available. The two main reasons are protection of individual's privacy and state security. In many cases, these are legitimate grounds to restrict freedom of information. Sometimes, however, reluctance to make information public may be an attempt to avoid accountability. Despite recent setbacks, the legal frameworks and mechanisms for ensuring public access to government information have expanded and improved over the last two decades. In 1980, for example, about 20 per cent of OECD countries had freedom-of-information laws. This proportion had increased to 40 per cent in 1990 and 80 per cent in 2000 (OECD 2001, 3). Yet, as important as it is, the provision of information is limited as a means to involve the population in public affairs. The approaches used are not interactive. Information generally flows only

Public Involvement in a Health Care Vision 253

one way and it may be non-responsive to the particularities of peoples' interests and concerns. They fail to provide opportunities to engage in discussion about the validity and meaning of the information provided. To address these shortfalls, increased efforts have been devoted to developing more interactive ways to consult the public. Consulting the Public Public consultation is another common means to involve the public in decision making. Public consultation processes consist of two-way exchanges of information between governments and the public. Abelson, Forest, Eyles, et al. (2001) classify consultation methods as either deliberative or non-deliberative. Non-deliberative methods involve exchanges of information but not deliberative dialogue aimed at achieving mutual understanding; deliberative-consultation procedures do include a dialogical component. Public opinion polls are at the nondeliberative end of the consultation continuum. Public hearings and focus groups are more interactive, and various types of citizen advisory and decision-making bodies are at the most deliberative end of the consultation continuum. Experience with non-deliberative consultation procedures shows that they too are useful but limited (Abelson, Forest, Eyles, et al. 2001): they tend to be sporadic and reactive complaint sessions involving more or less isolated individuals (Eksterowicz and Cline 1988). Deliberation is seen as the means to overcome these limitations (Eksterowicz and Cline 1988). Procedures such as citizen juries, citizen panels, planning cells, consensus conferences, and deliberative polling are all deliberative public-consultation methods that have potential application in the health policy domain at all levels of decision making - federal, provincial, and regional (Abelson, Forest, Eyles, et al. 2001). An Excursus on Deliberation, Discourse Ethics, and Forms of Representation What is deliberation? Deliberation is dialogical and public. This distinguishes it from personal contemplation (also often termed deliberation) and non-deliberative consultation and information giving. Proponents argue that it is the essence of public rationality and democracy and the only sure cure for the democratic deficit (Habermas 1984, 1987,1996; Majone 1989; Dryzak 1990; Fischer and Forrester 1993). As

254 Harley D. Dickinson

Mathews (1995, 42) states, 'deliberation is the DNA of democracy, it both forms and informs a public/ The principal objective of deliberative procedures is to achieve mutual understanding and, ideally, agreement on the normative legitimacy (justice or fairness) of a policy among all those whose needs and interests are affected by it. Discourse ethics specifies the means to do this (Dryzak 1990; Benhabib and Dallmayr 1990; Habermas 1990,1994). Discourse ethics takes the form of normative guidelines that specify the 'who, what, and how' of deliberation directed towards answering the following general question: 'In a world characterized by an irreducible plurality of values, and in the absence of a universal morality, how can we resolve, or at least accommodate, value differences in a way that is acceptable to all without resort to manipulation, threat or violence?' Specific versions of this question must be answered relative to all policy and practice decisions if those policies and practices are going to be rationally justifiable and acceptable to all those who are affected by them. Who may participate in deliberations? Anyone who is competent to speak and act, in principle, has a right to take part in discourse. In modern, mass societies, of course, it is impossible to achieve this degree of participation. It was for this reason, in part, that representative forms of democratic governance were developed and why they remain essential. The current democratic deficit, however, signals a crisis of legitimacy for existing institutions of representative democracy. Development and deployment of deliberative-consultation procedures is part of the attempt to resolve this crisis without abandoning representative forms of governance to direct democracy or to non-democratic forms of authoritarian and elite (technocratic) decision making. Deliberative consultation procedures generally are linked to neocorporatist forms of interest representation, although Hunold (2001) maintains they are also compatible with other forms of representation. DeSario (1987, 135) identifies four main forms: formal, ascriptive, descriptive, and substantive. Formal political representation entails entrusting elected officials with decision-making power that they exercise in the way they see fit. As we have seen, dissatisfaction with this form of representative democracy is thought by many to be the main source of the democratic deficit. Ascriptive representation refers to a situation where some group is delegated authority to make decisions in particular policy domains. It

Public Involvement in a Health Care Vision 255

is premised on the assumptions that those to whom this authority is delegated possess superior specialized knowledge and that they will use it to pursue general interests, not narrow self-interests. I argued previously that doubts about the validity of these assumptions as they apply to the medical profession are contributing to a legitimacy crisis in health care and calls for increased public involvement. Descriptive representation refers to the idea that participants in deliberative and decision-making processes should reflect the sociodemographic characteristics of the community as a whole. The underlying premise is that people with common characteristics share, and will promote, common interests. Finally, substantive representation is based on the premise that public consultation and decision making should involve members of groups espousing and working to realize particular interests. It assumes that active members of interest-based groups are more motivated to become informed and to participate in policy development and implementation than those with only diffuse interests. The legitimacy of neocorporatist forms of policy making and implementation rests upon these assumptions. Descriptive and substantive forms of representation have assumed the greatest significance in the context of calls for both deliberative consultation and direct democracy (DeSario 1987). Faced with a democratic deficit, governments are actively developing relationships with interest-based groups in order to extend and strengthen neo-corporatist decision-making capacity (Good 2001; Reed and Selbee 2001). Proponents of neo-corporatism, or substantive representation, recognize that not everyone is equally motivated or mobilized to participate in various aspects of civic life. Reed and Selbee (2001) refer to the minority of Canadians who are disproportionately engaged as Canada's civic core. Based on their analysis of the 1997 National Survey of Giving, Volunteering and Participating, they found that 28 per cent of Canadian adults account for 83 per cent of total hours volunteered, 77 per cent of total charitable dollars donated, and 69 per cent of civic participation. In terms of socio-demographic characteristics, the civic core is older, religious (but not Catholic), and relatively well-educated compared to the general population. Given this, initiatives like the federal government's effort to establish a government-voluntary sector accord (Good 2001) causes some to be concerned that it is an exercise in mobilizing bias (DeSario 1987). Reed and Selbee (2001) agree but maintain that the effect is directed

256 Harley D. Dickinson

towards the expansion of social capital and, therefore, is desirable: "The civic core, although small, is clearly a pillar of enormous significance in maintaining a just and mutually caring society; we would speculate that it may also have a central role in supporting democratic governance as well.' Discourse ethics specifies that everyone whose needs and interests are affected by a decision have the right to participate directly in policy deliberations. In practice, however, participation is generally mediated through various forms of representation - descriptive, substantive, ascriptive, and formal political. Representativeness, considered both as goal and as process, is not achieved through any singular means. Rather, as we have seen, there are numerous forms of representation, 'each by virtue of its own peculiar nature seeing, reflecting, [and] attempting to effectuate a slightly different facet of that great conglomeration of desires and interests that make up the electorate' (Pennock 1968, 27, quoted in DeSario 1987, 135). Thus, there is not one best way to represent interests in the exercise of power. Discourse ethics addresses the issue of who may be involved in deliberations. It also addresses the question: 'What may be discussed?' In principle, every participant in a discourse has a right to express his or her needs and interests, desires and preferences, principles and values. Similarly, any other participant may challenge the validity of such assertions. The person so challenged is obliged to provide reasons and arguments in support of his or her assertions. Argumentation, motivated to achieve mutual understanding, is the essence of rational deliberation (Toulmin 1958; Habermas 1984; Dickinson 1998). Discourse ethics also provides normative guidelines concerning how participants in deliberations are to interact. As we have seen, participants in deliberative discourse have the right to assert anything and to challenge any assertion made by others. They must not only exercise their own rights in these regards, however; they also have obligations to respect the rights of others to do so. Some argue that participants in deliberative discourse have obligations not only to respect others' rights to participate in deliberations but also to ensure that all other participants actually exercise their rights (Chambers 1995). These normative expectations presuppose that participants will be self-interested. Discourse ethics does not begin with the unrealistic expectation that participants will adopt Rawls's 'veil of ignorance,' or some other artificial device, as a way to move beyond self-interest to

Public Involvement in a Health Care Vision 257

altruism (Lomas 1997). Rather, the rights that participants in deliberations have publicly to assert anything and to challenge any assertions made by others discourages narrowly self-serving strategic communication, such as misrepresentation, deceit or manipulation, threats, coercion, and violence. By exposing strategic communicative action, deliberation discourages it and, in the process, contributes to the achievement of rational understanding and agreement based only on the force of the better arguments. Only agreements achieved through this means are recognized as genuine and legitimate (Habermas 1990, 89). Following from the discourse ethics is the principle that valid policies are those 'that meet (or could meet) with the approval of all affected in their capacity as participants in a practical discourse' (Habermas 1990,93). It is obvious that implementation of the norms of rational deliberation as specified in discourse ethics is impossible. It is possible, however, to approximate more or less closely these deliberative procedures and, thereby, to approximate realization of the above-mentioned policy principle that follows from them. Various deliberative consultation processes are more or less well designed to approximate these normative procedures (Abelson, Forest, Eyles, et al. 2001; Pickard 1998). Proponents of deliberative procedures argue that even a rough approximation of the ideal is worth pursuing. Several possible benefits of doing so are commonly identified. These include increased legitimacy of policy decisions that results from stakeholder and public involvement. The increased legitimacy is seen to come from the fact that deliberative consultations allow consideration of a broader range of options than may have occurred otherwise. Associated with this is the capacity that deliberative procedures create for participants to clarify their value preferences and to express the intensity and relative weights of their preferred options. More important, perhaps, deliberative procedures also are the basis of social learning, which, among other things, is understood as the achievement of new understandings of oneself, one's needs and interests, and one's rights and responsibilities relative to others. Policy development, implementation, and compliance are most effective when they are the result of mutual understanding and agreement based on processes of social learning. Effective institutional reforms, that is, organizational and procedural innovation, are predicated on social learning. Finally, social learning that takes place through participation in deliberative procedures has a correlate in the enhancement of delibera-

258 Harley D. Dickinson

tive skills and abilities of individuals who participate. That is, practice makes perfect - or at least better, and more civically literate - individuals. Such individuals are precisely those who increase the stock of social capital and thereby help sustain a just and mutually caring society with increased democratic capacity (Veenstra and Lomas 1999; Reed and Selbee 2001). Deliberative consultation is the most actively participatory form of public participation. The bulk of experience with using deliberative procedures for setting health services and policy priorities has been in the United Kingdom (Abelson, Forest, Eyles, et al. 2001). Yet both nondeliberative and deliberative consultation procedures have been, and are, used by various levels of government in Canada as well, from the national to the provincial, municipal, and regional levels. Deliberative Consultation: Active Public Participation Active public participation in policy formation and implementation is conceived of as a partnership between government and citizens (OECD 2001). Within this partnership, citizens have a right to be actively engaged in the policy-making process and governments have an obligation to enable meaningful public involvement and to take the results into account in making decisions. There is no consensus around the types of decisions various publics want to be, or should be, involved in (Abelson, Lomas, Eyles, et al. 1995; Lomas 1997). Nor is there a single, best way to involve the public in deliberative consultations. Various tools, procedures, and institutional means and forms of representation to enable effective deliberative consultation are being experimented with (Pickard 1998; OECD 2001). In Canada, at the federal level of government, efforts in this regard have increased since the early 1990s through the public consultation processes of the National Forum on Health (Ham 2000) and, more recently, the Commission on the Future of Health Care in Canada (the Romanow Commission) and the Kirby Committee. At the provincial level of government, there are numerous examples of task forces and commissions using public-consultation procedures. At the regional level, various initiatives of this sort also are evident. I concentrate on these. RHAs and Deliberative Procedures

An assessment of the decision-making needs of Saskatchewan RHAs done in the mid-1990s identified increased public involvement in

Public Involvement in a Health Care Vision 259

health system planning and governance as a top priority (Kouri 1996). A recent national survey also indicated that developing effective public consultation procedures was the second-ranked priority for RHA board members after population health research (Kouri 2002). Clearly, but not surprisingly, public consultation is a priority for RHAs. What do we know about RHA practices in this regard? Abelson, Forest, Smith, et al. (2001) found that regional health advisers in Ontario and decision makers in Quebec used public consultation procedures such as issue forums, workshops, focus groups, private meetings, public hearings, and surveys. They identify a trend away from large, formal, and often confrontational, public hearings, as the method of choice for public consultation, towards smaller, less formal, and more consultative processes such as open-house meetings. Such meetings seem to be an increasingly popular way to consult with the public among health care advisers in Ontario. They are defined as an opportunity for interested members of the public 'to drop in' to local or neighbourhood establishments to meet the decision makers, obtain information, discuss the issue under deliberation and offer feedback' (Abelson, Forest, Smith, et al. 2001,16). The August 2002 Newsletter of the Canadian Centre for Analysis of Regionalization and Health (CCARH) reported some of the findings of a recently completed national survey of public participation practices of RHAs (CCARH 2002,3). The Effective Consultation Project was funded by the Canadian Health Services Research Foundation (CHSRF) and conducted by Julia Abelson, Pierre-Gerlier Forest, John Eyles, and their colleagues.2 The RHAs that responded reported using a range of methods to involve the public. In descending order, they were focus groups (41 per cent), public meetings (39 per cent), small group meetings (34 per cent), surveys (29 per cent), open houses (10 per cent), and public hearings (9 per cent) (CCARH 2002, 3). Of these methods, all but surveys could have deliberative dimensions, with focus groups probably being the most deliberative. Between 70 and 80 per cent of the responding RHAs involved the public to set goals and priorities and to help design services. Around one-quarter involved the public in resource-allocation decisions (CCARH 2002, 3). Public involvement in these areas is consistent with the goal of ensuring that the health care system reflects the public's principles and values (Lomas 1997; Mitton and Donaldson 2002). The survey also showed that over 90 per cent of the responding RHAs attempted to involve citizens and the general public. About 77

260 Harley D. Dickinson

per cent reported involving both patients/health services users and other stakeholder groups, while about 64 per cent attempted to involve broad-based community organizations. Over 40 per cent of responding RHAs reported involving all the above-mentioned groups (Abelson, Forest, Eyles, et al. 2002). These data support the contention that RHAs pursue multiple public consultation strategies that employ different modes of interest representation. Consultations with patients, other health services users and stakeholder groups, and broad-based community organizations seem to fit the model of substantive representation discussed above. The widespread involvement of interest-based groups in the consultation process is also indicative of a neo-corporatist policy-making and implementation strategy. At the same time, efforts to involve citizens and the general public in consultation processes are consistent with descriptive forms of interest representation. The data give some support to the view that RHA boards recognize different forms of interest representation as legitimate and employ various consultation procedures. Relative to strategies for securing public participation, the survey showed that the most common means was through existing RHA committees (80 per cent), followed by the use of existing community networks, mailed invitations, the media, word of mouth, and electronic means (CCARH 2002,3). These data are interesting insofar as they suggest that the primary means to secure public participation and consultation is through the regular structures and functioning of the RHAs themselves. This supports the notion that RHAs constitute sites for enhanced public participation in health care policy and service delivery. I return to this point below. Which jurisdictions in Canada have been most successful in involving the public? This is a difficult question to answer because of the diversity of organizational and decision-making structures that exist in Canada and because of the variety of consultative processes being used. Nonetheless, if one uses the number of RHAs per province that reported using four or more public involvement procedures as an indicator of where the greatest efforts to involve the public have taken place, western Canada clearly sets the pace. In Manitoba and British Columbia, five RHAs reported using four or more public consultation procedures. This is followed by Alberta and Saskatchewan with three each. Nova Scotia and New Brunswick each have one RHA reporting the use of four or more public participation procedures. No RHAs reported using four or more public participation procedures in any of

Public Involvement in a Health Care Vision 261

the remaining regionalized provinces or territories (Abelson, Forest, Eyles, et al. 2002). Analysis of the instrumental use of deliberative public consultation procedures by RHAs needs to be supplemented by an analysis of RHAs as deliberation. That is to say, regionalization itself, or more precisely, establishment of RHA boards, can be understood as the creation of new public spaces where deliberation, or at least an approximation of deliberation, can, and does, occur (Finlay and Dickinson 2001). RHAs As Deliberative Consultation In this section, I suggest that the various regular and special meetings of RHA boards can approximate deliberative consultation methods. Such meetings, and the talk that takes place at them, are not widely recognized as deliberation. One reason for this is the tendency to see RHAs only as decision-making entities and not as sites where public consultation, deliberation, and social learning take place. Much of the talk that occurs in the context of formal RHA board meetings is not immediately related to decision making. For that matter, much of it is not deliberative either. Based upon ongoing HEALNet-funded research on the nature and use of evidence by RHA board members, the greatest proportion of talk appears to be for the purpose of information exchange. In particular, it consists of management personnel providing reports on various operational aspects of the health care system for which the RHA board is responsible. This is consistent with the findings of Mitton and Donaldson (2002), Lomas, Veenstra, and Woods (1997b), and Kouri, Dutchak, and Lewis (1997). Another proportion of RHA board talk, however, is directly linked to the making of decisions, while a third proportion can be understood as approximating deliberation. In this context, the discussion focuses on issues of board identity, its collective needs and interests, its rights and responsibilities relative to various stakeholders, and issues of how to develop and sustain those relationships in ways that enable the board to realize its objectives and meet its responsibilities. Talk can include reports of meetings with external interest groups that have occurred through the regular functioning of board committees or the activities of management personnel. It also involves face-to-face meetings with stakeholders. Deliberation is not decision making. It is, rather, a process of rational public opinion formation. Deliberation qua deliberation is a means to achieve mutual understanding and non-coerced agreement. Relative to

262 Harley D. Dickinson

democratic decision making, its function is to create informed public opinion and a political will to act, that is, to create the collective capacity and will to make justifiable choices. Deliberation, related to forging informed public opinion and a political will to act, should be uncoupled from the immediate decision-making needs (Habermas 1984). My argument is that RHA boards, both structurally and procedurally, constitute relatively new sites for deliberative public consultation. The structure and functioning of RHA boards is substantially equivalent to features identified as characteristic of effective public consultation procedures. On the basis of a review of relevant literature, Lomas (1997,110) suggests that the most effective means to involve the public is through representative panels of ten to twenty members who are provided small honoraria, who are replaced on a regular basis, and who meet regularly to set priorities via the deliberative achievement of a consensus view. These characteristics support effective public consultation in at least four ways (Lomas 1997). First, panels that meet on an ongoing basis make it worthwhile to ensure that the membership is representative. Lomas discusses two forms of representativeness. The first is what was previously identified as descriptive representativeness. That is, members can be chosen to reflect the socio-demographic characteristics of the general population in an area. Panels that are representative in this way, according to Lomas, are appropriate for consultation regarding broad service-category priorities. The second form of representativeness discussed is a form of substantive representation. Here Lomas suggests a spectrum of patients to be consulted on more specific questions of service. The second advantage of ongoing panels identified by Lomas (1997, 110) is related to the fact that regular interaction among members enhances the willingness of panel members 'both to take a collective rather than an individual view and to arrive at a consensus view/ To the extent that this is the case, it is consistent with the expectations associated with the application of discourse ethics. Third, constituting panels on an ongoing basis makes considerable investments in information acquisition worthwhile and allows for a critical interpretation of the relevance of the data for decision making. It also enables panel members to discuss information and its relevance with various experts. Finally, establishing panels characterized by the features and procedures described above makes the process of public consultation and input both visible and accountable. Such publicity in

Public Involvement in a Health Care Vision 263

the process helps minimize strategic actions and self-serving behaviours on the part of stakeholders and those with vested interests. To summarize, I suggest that the structural and procedural features of the panels described by Lomas, as well as the advantages that derive from them, can all be seen as descriptive of RHAs. In addition RHAs, in some jurisdictions, have the advantage of being not only deliberative bodies but also decision-making entities. Suggesting that the RHA boards, either elected or appointed, are sites where deliberation can occur is not to say that this potential is always realized. It is not. However, it is true that structural imperatives and organizational incentives can create an interest in achieving rational understanding and agreement among the occupants of decision-making positions. Those who are in positions of public responsibility for decision-making and who are publicly accountable to all those affected by them, as are RHA board members, will, more or less, be so motivated. This capacity can be further enhanced by conscious efforts to ensure that RHA board deliberations approximate as closely as possible the norms identified by the discourse ethics. It can also be enhanced by the use of a range of the deliberative and non-deliberative consultation procedures currently available (Abelson, Forest, Eyles, et al. 2001). Unfortunately, though RHA board members are in positions characterized by structurally generated interests to achieve mutual understanding through deliberative means, and though they are motivated to engage in that type of public involvement, counteracting characteristics of the decision-making environment undermine realization of the goal of informed deliberation and decision making. Impediments to Informed Deliberation

Informed deliberation and decision-making require relevant information available in a timely fashion and in comprehensible and useful formats. This is a perennial problem because RHAs often do not have the capacity to collect the needed information on their own (Badgeley 1982). The problems of availability and accessibility of relevant information are particularly germane where RHAs are legally responsible for the health status of the populations within their regions. For example, the results of a needs assessment done with RHA board members and senior management in Saskatchewan show that they were concerned about their ability to make informed decisions relative to health pro-

264 Harley D. Dickinson

motion because of inadequate information (Kouri 1996). And this is not just a Saskatchewan problem. A national survey of RHAs showed that 79 per cent of all RHAs in the country identified population health research as the most important type of research needed. Community participation research was ranked as the second priority by 65 per cent of all RHAs (Kouri 2002,17). Not only do RHAs lack the necessary information to make rational policy decisions relative to health promotion, they also lack the resources and the jurisdiction to address the most important determinants of health - poverty, environment, education, justice, employment, and working conditions - to name a few. As a result, healthpromotion efforts are restricted to the things that RHAs feel they can affect - primarily individual lifestyle choices and certain types of selfhelp and mutual-aid initiatives directed at the family and community levels (Burgess 1996). Given this limited scope for action, RHAs tend to understand health promotion in terms of reduced use of medical, hospital, and other forms of institutional care (Burgess 1996). This has at least two potentially contradictory effects. First, it often results in unpaid home-based care, usually provided by women, being substituted for professionally provided hospital and institutional care. The burdens, costs, and negative consequences associated with this trend are inadequately recognized or compensated. Second, de-emphasizing medical and hospital care may result in inadequate provision of, and access to, necessary health care services, or at least it may result in the perception of such inadequacies. Those who focus attention on waiting lists and waiting times generally interpret them as indicators of a crisis of availability caused either by the putative inadequacies of public administration or by inadequate levels of government funding. This is especially a concern where health care services and healthpromotion budgets are combined so that increased expenditures on health promotion come from resources gained through decreased expenditures on health care services. This is certainly the case in Saskatchewan where not only are the two budgets combined but resources can flow only through a one-way valve - that is - resources can be moved from institutional care to community-based and home care services, but not the other way (Lewis, Kouri, Estabrooks, et al. 2001). This undermines the legitimacy of health promotion among the general public, who tend to conflate health with health care (Lomas and Contandriopoulos 1994).

Public Involvement in a Health Care Vision 265

Paradoxically, regionalization, by providing numerous sites for active public involvement in health care planning and system governance, appears to have helped mobilize public resistance to health reform, at least those dimensions of health reform thought to jeopardize access to physician and hospital services. This resistance takes the form of critiques of health promotion and regionalization. From Health for All to Access to Health Care: Around the Policy Circle Evidence of an inversion of policy priorities, from health promotion to access to health care, is found, among other places, in the themes being addressed by the Commission on the Future of Health Care in Canada. For example, the public consultation exercises of this commission focused on three options for securing new resources for Medicare: fundamentally reforming the nature and delivery of health care services so as to increase efficiencies and thereby secure additional resources without need for new money; increasing the investment of public money in the system; or finding new resources by allowing increased private sector participation. These general options were advanced through concrete reform proposals. A similar emphasis is present in the three most recent provincial commission reports - Clair in Quebec, Fyke in Saskatchewan, and Mazankowski in Alberta. These reports give little indication that the health promotion policy paradigm is a priority. The Saskatchewan government's most recent plan for health reform also highlights this point. Developed in response to the recommendations of the April 2001 Fyke Commission report, and based on its own public consultations, it states prominently, and repeatedly, that it 'does not call for any hospital closures or conversions', and that communities 'will be involved in decision-making if a hospital cannot be maintained due to shortages of key health professionals or other circumstances' (Saskatchewan Health 2001,5). It may appear from this that efforts to demedicalize and de-institutionalize health care are in abeyance and that the health promotion policy framework has been abandoned. This, I believe, would be too hasty a judgment. The above quoted statement notwithstanding, the Saskatchewan government's action plan is replete with an energetic commitment to enhanced managerialism in the form of plans to introduce performance indicators and increased system accountability. Key

266 Harley D. Dickinson

elements of the Saskatchewan government's action plan include, for example, a drastic reduction in the number of RHAs from thirty-two to twelve, government-appointed CEOs for the reformed regions, the elimination of elected boards, reform initiatives in primary health care, and the creation of a 'Quality Council.' All are claimed to be responses to the fact that people have told the government 'they want more efficient management, and more accountability' (Saskatchewan Health 2001, 7). Regionalization under Attack The Saskatchewan government's action plan marks a turn towards a more technocratic management regime and away from what arguably was the most democratic of the regionalized health care systems in Canada. In a sense, this brings it more in line with regionalization in other provinces. The managerialism of regionalized health care systems across the country is seen as a threat to the autonomy of physicians and hospitals (MacKinnon 2002; Borsellino 2002; Borsellino 1997a,b; Gray 1995). It is perhaps not surprising that strong criticism of regionalization has recently come from the CEO of the Ontario Hospitals Association (OHA) and the editor of the Medical Post. As we have seen, Ontario is the only jurisdiction where hospital management has not been placed under the control of RHAs. Given the success of regionalization in reducing the number and use of hospital beds in those jurisdictions where it has been introduced, it is not surprising to find resistance to it coming from this quarter. Despite this there surely are advocates for its introduction into Ontario. An 8 April 2002 OHA conference entitled Regionalization: Is It the Answer for Ontario? suggests this may be the case. Yet, in a presentation at that conference, the president and CEO of the OHA declared regionalization in the rest of the country a failure. According to him, it has failed to contain costs, it has failed to achieve a better integration and coordination of services, and it has failed to increase community involvement (MacKinnon 2002). The main reason given for its failure is that regionalization provided no role for physicians in the reform process. The editorial in the 30 April 2002 issue of the Medical Post vigorously endorsed the argument that regionalization has failed and echoed the view that lack of physician involvement is the main cause of this (Borsellino 2002).

Public Involvement in a Health Care Vision 267

But declarations of the failure of regionalization may be as premature as was its previously reported death (Taylor 1978). None of the factors previously identified as motivating its introduction have abated. That said, however, regionalization has not been successful in demedicalizing health care, or in altering the state/profession accommodation at the heart of medicare (Touhy 1995). Nor has it been successful in dislodging the widely held belief that the primary determinant of health is medicine (Lomas and Contandriopoulos 1994). Regionalization has been much more successful in de-institutionalizing health care delivery. In particular, it has been successful in significantly reducing the number of hospitals, hospital beds, rates of utilization, duration of stay, and costs. In some senses, it has been less effective in expanding community-based and home care services and in integrating and coordinating services. Rising prescription drug expenditures, however, suggest a predominantly pharmaceutical strategy for providing community-based care. Relative to the goal of increased public involvement - the main focus of this paper - it is hard to say whether regionalization has been successful or not. If the measure of success is the successful institutionalization of the health promotion policy paradigm, then it has failed. RHAs, just like provincial governments and the federal government, have been unsuccessful in convincing people that health promotion works. The main reason for this failure, however, is probably public resistance to the demedicalization and de-institutionalization of the health care system - a process that is widely understood as a threat to Medicare and to health. If I am correct in this assessment, then public involvement has been successful, but successful in resisting a major health policy initiative, not in implementing it. The most difficult, and important, task facing health reformers is still changing the popular belief that providing medical care is the same as promoting health (Lomas and Contandriopoulos 1994, 282). Conclusion In these turbulent times, nothing is more volatile than the health care system. Even experts in health policy and health care reform are hard pressed to keep abreast of changes, let alone to know what they all mean. Is it any wonder that the public is concerned, if not confused, about what is going on and what the future holds?

268 Harley D. Dickinson

There is no guaranteed solution to this state of anxious uncertainty, but getting involved is strongly recommended. There are, as we have seen, contradictory imperatives and interests both pushing for, and pulling against, expanded opportunities for meaningful public involvement in health care decision making. Furthermore, the balance of these forces varies from time to time and from place to place. Similarly, different procedures for public involvement result in different outcomes. Nevertheless, I have argued throughout this paper that deliberative public consultation procedures have several features that strongly recommend their expanded use in policy development and implementation. This is particularly the case with regard to the formation of public opinion and political will, especially as a means to address questions of fairness. Relative to this, meaningful public involvement in policy relevant decision making requires the creation of organizational capacities and the institutionalization of deliberative procedures. As we have seen, substantial progress has been made in these regards. There is also much optimism that new information and communication technologies will greatly expand the capacity and opportunities for public involvement. Much of this potential, however, has yet to be realized. I do not advocate direct democratic decision making but rather public involvement in achieving agreement on goals, principles, and values for the health care system. Deliberative participation procedures provide a means for ensuring that needs and interests are interpreted and considered in deciding on the goals of health care policy and the means of achieving them. Democratically constituted or representative appointed RHA boards provide an important institutional means for linking deliberative public consultation directed to public opinion and political will formation to the decision-making process. Linking deliberative consultation with deliberative decision making seems ideally suited to realization of the various benefits claimed for applied discourse ethics. Similarly, initiatives such as the National Forum on Health and the Commission on the Future of Health Care in Canada, as well as provincial commissions, provide important, albeit, ad hoc, opportunities to engage the public in the process of generating informed public opinion and the political will to act in important policy domains. By so doing, they provide an opportunity for us all to deliberate on the kind of society we want to live in and the kind of people we want to be.

Public Involvement in a Health Care Vision 269 NOTES

I would like to thank Susan Robertson, Tara Finlay, and Julie Kosteniuk for assistance on various aspects of the preparation of this paper. I am grateful to Julia Abelson for permission to use data from the Effective Consultation Project. I also want to thank the two anonymous reviewers for their critical insights and helpful suggestions, some of which I have adopted. This undoubtedly is a better paper because of the help and input of all these people. I am, however, entirely responsible for any weaknesses and errors it may still contain. 1 Throughout this paper, I use the phrase regional health authority (RHA) to refer to the sub-provincial structure of health care policy making and service delivery. The names of these entities vary by province. 2 Julia Abelson generously gave me permission to report some of the unpublished findings from this study. My interpretation of the data may not reflect the views of the researchers or the funding agent.

REFERENCES

Abelson, J., et al. 1995. Does the community want devolved authority? Results of deliberative polling in Ontario. CAM/153(45): 403-12. Abelson, J., et al. 2002. The 3-minute public participation survey: Summary of findings. Draft report from the Effective Public Consultation Project Towards more meaningful, informed and effective public consultation: A comparative study of Regional Health Authorities in Canada. A research project funded by the Canadian Health Services Research Foundation. Abelson, J., P-G. Forest, J. Eyles, et al. 2001. Deliberations about deliberative methods: Issues in the design and evaluation of public consultation processes. McMaster University Centre for Health Economics and Policy Analysis, research working paper 01-04, June 2001. Abelson, J., P-G. Forest, P. Smith, et al. 2001. Obtaining public input for health systems decisionmaking: Past experiences and future prospects. McMaster University Centre for Health Economics and Policy Analysis research working paper 01-03, April 2001. Anderson, T.F. and G. Mooney, ed. 1990. The challenge of medical practice variations. London: Macmillan. Angus, D.E. 1991. Review of significant health care commissions and task forces in

270 Harley D. Dickinson Canada since 1983-84. Ottawa: Canadian Hospital Association/Canadian Medical Association/Canadian Nurses Association. Angus, D.E., et al. 1995. Sustainable health care for Canada. Ottawa: University of Ottawa Press. Armstrong P. 1997. Privatizing care, in P. Armstrong et al., ed., Medical alert: New work organizations in health care. Toronto: Garamond Press, 11-30. Armstrong, P., and H. Armstrong. 1996. Wasting away: The undermining of Canadian health Care. Toronto: Oxford University Press. Arnstein, S.R. 1969. A ladder of citizen participation. Journal of the American Institute of Planners, July, 216-24. Badgeley, R., and S. Wolfe. 1967. Doctors' strike: Medical care and conflict in Saskatchewan. Toronto: Macmillan of Canada. Badgeley, R. 1982. Regionalization of health services in Canada. Israel Journal of Medical Sciences 18: 375-83. Barry, C.A., et al. 2001. Giving voice to the lifeworld: More humane, more effective medical care? A qualitative study of doctor-patient communication in general practice. Social Science & Medicine 53:487-505. Beierle, T.C. 1999. Using social goals to evaluate public participation in environmental decisions. Policy Studies Review 16(3-4): 75-103. Benhabib, S., and F. Dallmayr, ed. 1990. The communicative ethics controversy. Cambridge, Mass., and London: MIT Press. Blishen, B. 1991. Doctors in Canada: The changing world of medical practice. Toronto: University of Toronto Press in association with Statistics Canada. Bolaria, B.S. 1988. The politics and ideology of self-care and lifestyles, in B.S. Bolaria and H.D. Dickinson, ed., Sociology of health care in Canada. Toronto: Harcourt Brace Jovanovich, 537^19. Bolaria, B.S., and H.D. Dickinson. 2001. The evolution of health care in Canada: Toward community or corporate control?, in W.C. Cockerham, ed., The Blackwell Companion to Medical Sociology. Maiden, Mass., and Oxford, U.K.: Blackwell Publishers, 199-213. Borsellino, M. 1997a. NS doctors prepare to fight regionalization. Medical Post 33:12. - 1997b. Doctors wary of regionalization plans in BC. Medical Post 33: 80. - 2002 (30 April). Editorial: Regional mumbo-jumbo displayed for what it is. Medical Post 38(17). Available at http://www.medicalpost.com/mdlink/ english/members/medpost/data/3817/08A.HTM> (1 May 2002). Boudreau, F. 1991a. Stakeholders as partners? The challenges of partnership in Quebec mental health policy. Canadian Journal of Community Mental Health 10: 7-28.

Public Involvement in a Health Care Vision 271 - 1991b. Partnerships as a new strategy in mental health policy: The case of Quebec. Journal of Health Politics, Policy and Law 16: 307-30. Brock, K.L. 2001. Promoting voluntary action and civil society through the state. Isuma, 2(2): 53-61. Brudney, J.L. 1990. Expanding the government-by proxy construct: Volunteers in the delivery of public services. Nonprofit and Voluntary Sector Quarterly 19(4): 315-28. Burgess, M. 1996. Health care reform: Whitewashing a conflict between health promotion and treating illness?, in M. Stingl and D. Wilson, ed., Efficiency vs. equality: Health reform in Canada. Halifax: Fernwood Publishing, 153-62. Caress, A. 1997. Patient roles in decision-making. Nursing Times 93(31): 45-8. Carroll, W.K., ed. 1992. Organizing dissent: Contemporary social movements in theory and practice. Toronto: Garamond Press. CCARH (Canadian Centre for Analysis of Regionalization and Health). 2002. RHAs experiment with public participation methods, Newsletter, August, 3. Available at http://www.regionaliozation.org (16 September 2002). Chambers, S. 1995. Feminist discourse/practical discourse, in J. Meehan, ed., Feminists read Habermas: Gendering the subject of discourse. New York and London: Routledge 163-80. Charles, C, A. Gafni, and T. Whelan. 1997. Shared decision-making in the medical encounter: What does it mean? (or it takes at least two to tango). Social Science & Medicine 44(5): 681-92. Charles, C., and S. DeMaio. 1993. Lay participation in health care decisionmaking: A conceptual framework. Journal of Health Politics, Policy and Law 18(4): 881-904. Church, J. 2001. Citizen participation in health (care) decision-making: What we know and where we might go. The shift: A newsletter on shifting paradigms in health and health promotion 3(3): 1, 3. Church, J., and P. Barker. 1998. Regionalization of health services in Canada: A critical perspective. International Journal of Health Services 28(3): 467-86. CM A, 1995. Regionalization of health services in Canada: A Canadian Medical Association policy position paper. May 1995. Coburn, D. 1998. State authority, medical dominance, and trends in the regulation of the health professions: The Ontario case, in D. Coburn, C. D'Arcy, and G. Torrance, ed., Health and Canadian society: Critical perspectives, 3rd edition. Toronto: University of Toronto Press, 332^46. Coburn, D., G.M. Torrance and J.M. Kaufert. 1983. Medical dominance in Canadian historical perspective: The rise and fall of medicine? International journal of Health Services 13(3): 407-32.

272 Harley D. Dickinson Cohen,}., and A. Arato. 1994. Civil society and political theory. Cambridge, Mass., and London: MIT Press. Connor, D.M. 1986. A new ladder of citizen participation. Constructive Citizen Participation 14(2): 3-5. Crichton, A., et al. 1997. Health care a community concern? Developments in the organization of Canadian health services. Calgary: University of Calgary Press. Crombie, D.L., K.W. Cross, and D.M. Fleming. 1992. The problem of diagnostic variability in general practice. Journal of Epidemiology and Public Health 46(4): 447-54. Dalton, R.J., W.F. Burklin, and A. Drummond. 2001. Public opinion and direct democracy. Journal of Democracy 12(4): 141-53. Deber, R. 1999. Medical savings accounts: A fine idea unless you're sick. Health Policy Forum, spring, 4-5. DeSario, J. 1987. Consumers and health planning: Mobilization of bias?, in J. DeSario and S. Langton, ed., Citizen participation in public decision making. Westport, Conn.: Greenwood Press, 133-51. DeSario, J., and S. Langton. 1987. Citizen participation and technocracy, in J. DeSario and S. Langton, ed., Citizen participation in public decision making. Westport, Conn.: Greenwood Press, 3-17. Dickinson, H.D. 1994. The changing health care system: Controlling costs and promoting health, in B.S. Bolaria and H.D. Dickinson, ed., Health, illness and health care in Canada, 2nd edition. Toronto: Harcourt Brace, 106-29. - 1998. Evidence-based decision-making: An argumentative approach. International Journal of Medical Informatics 51(2.3): 71-82. - 1996. Health reforms, empowerment and the democratization of society, in M. Stingl and D. Wilson, ed., Efficiency vs. equality: health reform in Canada. Halifax: Fernwood Publishing, 179-89. Dickinson, H.D., and D. Hay, 1988. The structure and cost of health care in Canada, B.S. Bolaria and H.D. Dickinson, ed., in Sociology of health care in Canada. Toronto: Harcourt Brace Jovanovich, 51-73. - 2002. Health care, health promotion and health reforms, in B.S. Bolaria and H.D. Dickison, ed., Health, illness and health care in Canada, 3rd edition. Toronto: Nelson Thompson Learning, 351-71. Dickinson, H.D., and R. Torgerson. 1998-9. Sustainability and health: The prairie ecosystem study. Health and Canadian Society 5(2): 177-88. Dorland, J.L., and S.M. Davis, ed. 1995. How many roads ...? Regionalization and Decentralization in health care. Kingston and Ottawa: Queen's University and CMA. Drass, K. A. 1982. Negotiation and the structure of discourse in medical consultation. Sociology of Health and Illness 4(3): 320-41.

Public Involvement in a Health Care Vision 273 Dryzak, J.S. 1990. Discursive democracy: Politics, policy, and political science. Cambridge, U.K., and New York: Cambridge University Press. Eksterowicz, A., and P. C. Cline. 1988. Citizen participation: Opportunities to influence public policy. Newsletter, Center for Public Service, University of Virginia, 65(1): 1-6. Epp, J. 1986. Achieving health for all: A framework for health promotion. Ottawa: Health and Welfare Canada. Evans, R.G., And G.L. Stoddart. 1994. Producing health, consuming health care, in R.G. Evans, M.L. Barer, and T.R. Marmor, ed., Why are some people healthy and others not? The determinants of health of populations. New York: Aldine de Gruyter, 27-64. Feather, J. 1991a. From concept to reality: Formation of the Swift Current health region. Prairie Forum, 16(2): 59-80. - 1991b. Impact of the Swift Current health model: Experiment or model? Prairie Forum 16(4): 225-18. Feingold, E. 1977. Citizen participation: A review of the issues, in H.M. Rosen, J.M. Metsch, and S. Levy, ed., The consumer and the health care system: Social and managerial perspectives. New York: Spectrum. Finlay, T., and H.D. Dickinson. 2001. Policy level decision-making: Approximating the rules of rational discourse. Turning your thinking upside down. Fourth International Conference on the Scientific Basis of Health Services, Sydney Conference Centre, Sydney Australia, 22-25 September. Fischer, F., and J. Forrester, ed. 1993. The argumentative turn in policy analysis and planning. Durham, N.C., and London: Duke University Press. Flood, C. and T. Epps. 2001. Can a patients' bill of rights address concerns about waiting lists? Draft working paper, Health Law Group, Faculty of Law, University of Toronto. Available at http://www.irpp.org> (5 April 2002). Fraser, N. 1989. Unruly practices: Power, discourse, and gender in contemporary social theory. Minneapolis: University of Minnesota Press. Freidson, E. 1973. Profession of medicine: A study in the sociology of applied knowledge. New York: Dodd Mead. Fuller, C. 1998. Caring for profit: How corporations are taking over Canada's health care system. Vancouver and Ottawa: New Star Books and the Canadian Centre for Policy Alternatives. Gafni, A., C. Charles and T. Whelan. 1998. The physician-patient encounter: The physician as a perfect agent for the patient versus the informed treatment decision-making model. Social Science & Medicine 47(3): 347-554. Good, D.A. 2001. A Government-voluntary sector accord. Isuma 2(2): 46-52. Grahame-Smith, D. 1995. Evidence-based medicine: Socratic dissent. BMJ 310: 1126-7.

274 Harley D. Dickinson Gratzer, D. 1999. Code blue: Reviving Canada's health care system. Toronto: ECW Press. - 2002. Opinions/medicare: Chartering to nowhere. Medical Post 38(17). http://www.medicalpost.com/ (1 May 2002). Gray, C. 1995. CMA-cosponsored conference raises many questions about the future of regionalized health care. CMAJ153: 642-5. Gray, J. A.M. 1997. Evidence-based healthcare: How to make health policy and management decisions. Edinburgh: Churchill Livingstone. Guadagnoli, E., and P. Ward. 1998. Patient participation in decision making. Social Science & Medicine 47(3): 329-39. Gwyn, R., and G. Elwyn. 1999. When is a shared decision not (quite) a shared decision? Negotiating preferences in a general practice encounter. Social Science & Medicine 49: 4347-447. Habermas, J. 1971. Toward a rational society: Student protest, science and politics. London: Heinemann Educational Books. - 1976. Legitimation crisis. London: Heinemann. - 1984. The theory of communicative action, Volume 1: Reason and the rationalization of society. Boston: Beacon Press. - 1987. The theory of communicative action, Volume 2: Lifeworld and system: A critique of functionalist reason. Boston: Beacon Press. - 1990. Moral consciousness and communicative action. Cambridge, Mass.: MIT Press. - 1994. Justification and application: Remarks on discourse ethics. Cambridge, Mass., and London: MIT Press. - 1996. Between facts and norms: Contributions to a discourse theory of law and democracy. Cambridge, Mass.: MIT Press. Ham, L. 2000. Strengthening government-citizen connections: Health policy in Canada. Paris: OECD. Hamowy, R. 1984. Canadian medicine: A study in restricted entry. Vancouver: Eraser Institute. Health Canada. 2002. Some key health promotion stepping stones. Health Policy Research 1(3):2. Health Edition (a). 2002. Message to Romanow: More federal funding. 6(11), 15 March, 1-2. Health Edition (b). 2002. 6(18), 2 May, 3. Health Edition (c). 2002. Medical error: A global health care issue. 6(35), 6 September, 2. Hunold, C. 2001. Corporatism, pluralism, and democracy: Toward a deliberative theory of bureaucratic accountability. Governance: An International Journal of Policy and Administration 14(2): 151-67.

Public Involvement in a Health Care Vision 275 Illich, 1.1976. Limits to medicine. Medical nemesis: The expropriation of health. Toronto and London: McClelland and Stewart in association with Marion Boyars. IRPP. 2001. A patients' charter in IRPP task force on health policy, 25-8. Available at http://www.irpp.org (5 April 2002). Kettl, D.F. 1988. Government by proxy: (Mismanaging federal programs. Washington: D.C.: CQ Press. Kouri, D. 1996. Assessing board decision-making needs: Saskatchewan district health boards. Occasional paper no.l. Saskatoon: HEALNet Regional Health Planning. Kouri, D. 2002. HEALNet Regionalization Research Centre. Conference presentation Regionalization in health care: What's really working? Vancouver, 2-23 February. Kouri, D., J. Dutchak, and S. Lewis. 1997. Regionalization at age five: Views of Saskatchewan health care decision-makers. Saskatoon: HEALNet Regional Health Planning. Lalonde, M. 1974. A new perspective on the health of Canadians: A working document. Ottawa: Minister of Supply and Services. Langton, Stuart. 1978. Citizen participation in America. Lexington, Mass.: Lexington Books. Lewis, S. 1997. Regionalization and devolution: Transforming health, reshaping politics? HEALNet: Regional Health Planning, Occasional paper no.2, October 1997. Lewis, S., et al. 2001. Devolution to democratic health authorities in Saskatchewan?: An interim report. CMAJ 164(3): 343-7. Littow, M., and S. Muller. 1998. Feasibility of health care allowances in Canada, prepared by Milliman and Robertson. Toronto: Consumer Policy Institute. Lomas, J. 1997. Reluctant rationers: Public input into health care priorities. Journal of Health Services Research and Policy 2(2): 103-11. Lomas, J., and A-P. Contandriopoulos. 1994. Regulating limits to medicine: Towards harmony in public-and self-regulation, in R.G. Evans, M.L. Barer, and T.R. Marmor., ed., Why are some people healthy and others not? The determinants of health of populations. New York: Aldine de Gruyter 253-83. Lomas, J., J. Woods, and G. Veenstra. 1997a. Devolving authority for health care in Canada's provinces: 1. An introduction to the issues. CMAJ 156(3): 371-7. - 1997b. Devolving authority for health care in Canada's provinces: 2. backgrounds, resources and activities of board members. CMAJ 156(4): 513-20. - 1997c. Devolving authority for health care in Canada's provinces: 3. Motivations, attitudes and approaches of board members. CMAJ 156(5): 669-76.

276 Harley D. Dickinson - 1997d. Devolving authority for health care in Canada's province: 4. Emerging issues and prospects. CMAJ 156(6), 817-23. McDowell, 1.1986. National strategies for health promotion. Letters to Canadian Journal of Public Health 77(6): 448. MacKinnon, D. 2002. Regional authorities in Canada: Lessons for Ontario. Notes for remarks made at the OHA conference on regionalization: Is it the answer for Ontario? 8 April, Ottawa. Available at http://caohat03-oha. com/oha/webnews.nst/(ww)/PSPR-58ZP6V?OpenDocument (10 April, 2002). McMahon, F., and M. Zelder. 2002. Making health spending work. A Fraser Institute occasional paper. Public Policy Source 54. Mahtre, S.L., and R.B. Deber. 1992. From equal access to health care to equal access to health: A review of Canadian health commissions and reports. International Journal of Health Services 22(4): 645-68. Majone, G. 1989. Evidence, argument and persuasion in the policy process. New Haven, Conn.: Yale University Press. Makoul, G., P. Arntson, and T. Schofield. 1995. Health promotion in primary care: Physician-patient communication and decision-making about prescription medications. Social Science &1 Medicine 41(9): 1241-54. Mathews, D. 1995. Public deliberation: What talk radio can't do. The Responsive Community, fall, 37-42. Mitton, C, and C. Donaldson. 2002. Setting priorities in Canadian regional health authorities: A survey of key decision makers. Health Policy 60: 39-58. Northcott, H.A. 1994. Threats to medicare: The financing, allocation and utilization of health care in Canada, in B.S. Bolaria and H.D. Dickinson, ed., Health, illness and health care in Canada, 2nd edition Toronto Harcourt Brace, 65-82. O'Connor, A.M., et al. 1999. The Ottawa patient decision aids. Effective Clinical Practice 2(4): 163-70. O'Connor, J. 1976. The fiscal crisis of the state. New York: St Martin's Press. OECD. 2001. Engaging citizens in policy-making: Information, consultation and public participation. PUMA Policy Brief No. 10, July 1-6. Omran, A.R. 1971. The epidemiologic transition: A theory of the epidemiology of population change. Millbank Memorial Fund Quarterly 49: 509. Partridge, C. 1997. Editorial - The patient as decision maker. Physiotherapy Research International, 2(4): iv-vi. Pennock, J.R. 1968. Political representation: An overview, in J.R. Pennock and J.W. Chapman, ed., Representation. New York: Atherton Press. Pickard, S. 1998. Citizenship and consumerism in health care: A critique of citizens' juries. Social Policy and Administration 32(3): 226^4.

Public Involvement in a Health Care Vision 277 Pleiger, D. 1990. Policy networks and the decentralization of policy making. Bulletin European Social Security 55/6: 37-9. Putnam, R. 2001. Social capital: Measurement and consequences. Isuma 2(1): 41-51. Ramsay, C. 1998. Medical savings accounts: Universal, accessible, portable, comprehensive health care for Canadians. Vancouver: Fraser Institute. Rasmussen, K. 2001. Regionalization and collaborative government: A new direction for health system governance, in Duane Adams, ed., Federalism, democracy and health policy in Canada. Montreal: McGill-Queen's University Press, 239-70. Reed, P.B., and L.K. Selbee. 2001. Canada's civic core: On the disproportionality of charitable giving, volunteering and civic participation. Isuma 2(2): 28-33. Robinson, A., and R.G. Thompson. 2000. The potential use of decision analysis to support shared decision making in the face of uncertainty: The example of atrial fibrillation and warfarin anticoagulation. Quality in Health Care 9: 23844. Roos, N.P., and L.L. Roos. 1994. Small area variations, practice style, and quality of care, in R.G. Evans, M.L. Barer, and T.R. Marmor, ed., Why are some people healthy and others not? The determinants of health of populations. New York: Aldine de Gruyter, 231-52. Sackett, D.L. et al. 1998. Evidence-based medicine: What it is and what it isn't. Centre for Evidence-based Medicine. Available at http://cebm.jr2.ox.ac.uk/ ebmisisnt.html (13 July 1999). Salamon, L.M. 1981. Rethinking public management: Third-party government and the changing forms of government action. Public Policy 29(3): 255-75. Saskatchewan Heath. 2001. Saskatchewan, healthy people. Healthy province. The action plan for Saskatchewan health care: A summary. Regina: Saskatchewan Health. Available at http://www.health.gov.sk.ca (28 November, 2001). Short, S.D. 1996. Patient compliance, client participation and lay reskilling: A review of some sociological work on lay participation in health care decision making. Health Care Analysis 4:168-77. Stanbury, W.T., and MJ. Fulton. 1988. Consultation and public participation processes in government policy making: A conceptual framework. University of Ottawa working paper, 88-12. Stein, J. G. 2001. The cult of efficiency. Toronto: House of Anansi Press. Stevenson, F.A., et al. 2000. Doctor-patient communication about drugs: The evidence for shared Decision Making. Social Science and Medicine 50: 829-^40. Taylor, 1978. Health insurance and Canadian public policy: The seven decisions that created the Canadian health insurance system. Montreal: McGill-Queen's University Press.

278 Harley D. Dickinson - 1960. The role of the medical profession in the formation and execution of public policy. Canadian Journal of Economics and Political Studies 25/91): 10827. Toulmin, S. 1958. The uses of argument. Cambridge, U.K.: Cambridge University Press. Tuohy, C.H. 1995. What drives change in health care policy? A comparative perspective. Timlin Lecture, University of Saskatchewan, 30 January. - 1999. Accidental logics: The dynamics of change in the health care arena in the United States, Britain, and Canada. New York: Oxford University Press. Ubel, P.A., and G. Lowenstein. 1997. The role of decision analysis in informed consent: Choosing between intuition and systematicity. Social Science & Medicine 44(5): 647-56. Veenstra, G. 2001. Social capital and health. Isuma 2(1): 72-81. Veenstra G., and J. Lomas. 1999. Home is where the governing is: Social capital and regional health governance. Health & Place 5:1-12. Vayda, E., R.G. Evans, and W.R. Mindell. 1979. Universal health insurance in Canada: History, problems, trends. Journal of Community Health 4: 217-31. Weller, G.R., and P. Manga. 1983. The push for reprivatization of health care services in Canada, Britain and the United States. Journal of Health Politics Policy and Law 8(3): 495-18. Weston, W.W. 2001. Commentary: Informed and shared decision-making: The crux of patient-centred care. CMAJ 165(4): 438-9. Wilensky, H.L. 1981. Democratic corporatism, consensus, and social policy: Reflections of changing values and the 'crisis' of the welfare state, in OECD, ed., The welfare state in crisis: An account of the conference on social policies in the 1980s. Paris: OECD, 185-95. Woolcock, M. 2001. The place of social capital in understanding social and economic outcomes. Isuma, 2(1): 11-17. Wooley, F. 2001. The strengths and limits of the voluntary sector. Isuma 2(2): 21-7.

9 Public Participation and Citizen Governance in the Canadian Health System JULIA ABELSON AND JOHN EYLES

Public participation in the health sector in Canada has taken many forms since the establishment of a universal, publicly administered health system. Citizen users routinely act as 'consultants' in service and program planning, offering their advice about service needs and problems. Interested citizens act as resource allocators through membership on priority-setting committees and, ultimately, as health system governors in their role as hospital, health centre, or health authority board members. The public also participates as 'protector of the public interest' through membership on regulatory bodies. Increasingly, the public is being asked to offer its views about the future directions of the health system, through either opinion polls or more in-depth public dialogues. Formal, explicit public-participation mechanisms have been called for recently within a federal/provincial/territorial agreement to ensure the health system's accountability to the public. With so many distinct and overlapping roles for the public, and goals for its participation in the governance of the health system, public participation can contribute in many ways to the performance of the health system and to the expression of democratic values held in Canadian society. This paper discusses these issues and is guided by the following questions: 1. What can public participation, and democratic processes more broadly, contribute to the health system, its governance, and the values held by Canadians towards and in the health system? 2. What has public participation in the Canadian health system contributed towards:

280 Julia Abelson and John Eyles

a) the efficient functioning of organizations in the health sector? b) strengthening citizen commitment towards health programs and policies? c) encouraging the expression of democratic values in Canadian society? 3. What are reasonable objectives for public participation in the future governance of the Canadian health system and how can more effective, legitimate processes be created to achieve these objectives? Before addressing each of these questions, we briefly discuss a number of concepts that will be referred to throughout the paper to ensure a common understanding at the outset. First, public participation is a broad set of practices that includes passive forms of citizen involvement, where the public's views are sought as an input to a planning or decision-making process, and more active involvement through direct participation in decision-making processes and structures. Public participation also encompasses citizen engagement and deliberative democracy. In some sections, we refer to specific forms of public participation that are undertaken within different issue and decision-making contexts. In other sections, we simply refer to the broad range of participation activities. With respect to the 'public' in public participation, there are multiple meanings of public and so we need to cast our eye widely to consider all those who have a stake in the health system, including users and potential users of care, public and patient advocates, and providers and experts, who are also citizens as well as taxpayers. Given the paper's orientation to democratic processes and governance, we are primarily interested in the public as actual and potential 'citizen governors,' with a necessary bias towards those demonstrating an interest in this role. Discussions of governance also feature prominently. Like participation, meanings of governance vary among fields of study and intended audiences. In this paper, governance is viewed as the interaction between citizens, experts, and elected representatives in the development and implementation of policies designed to solve social problems. Health system governance may be considered at one or more levels: 1) the macro level (i.e., central or national authority), which has responsibility for the public sector as a whole (e.g., federal or provincial governments or pan-Canadian authorities); the meso level, which includes regional or functional authorities determined either geographically or by population group; and the micro level of specific

Public Participation and Citizen Governance 281

organizations or communities (Forest et al. 1999). Given the paper's focus on issues of public participation, we are specifically interested in the contributions of citizens, through their democratic participation and interactions with experts and elected officials, to the governance of the health system. How Public Participation Can Contribute to the Health System and the Expression of Democratic Values Public Participation and Citizen-Governance Ideals The contribution of a participatory citizenry to successful democracy has been the subject of long-standing (and seemingly never-ending) debate among democracy theorists and practitioners. These debates frequently focus on the merits of representative versus participatory democracy1 (i.e., how directly should citizens be involved in government decision making), the desired goals for participation, and the methods for achieving them. Tensions exist between views of participation as an essential element of successful democracy (and inherently desirable in its own right) and participation as a means for achieving something else, be it a specific decision outcome, a desire for more informed, accountable or legitimate decision making, or perhaps to delay or share the blame for a difficult decision. Lying somewhere between these two basic tenets is a desire for public participation to contribute to a more educated and engaged citizenry.2 Choosing between public-participation goals and the means for achieving them is shaped by broader social and economic contexts and changes and by the values held by the governing elite and publics. Over the last few decades, public confidence and trust in representative democracy and traditional political institutions has declined as interest in politics has increased. This relationship has produced a frustrated public, dissatisfied with the status quo yet unable to find meaningful ways to participate in society (Nevitte 1996; Ekos 1996). The precise source of this dissatisfaction is thought to result from the interaction of economic, cultural, and political changes that have altered citizens' views about, and desired relationships with, their political institutions (Inglehart 1995; Inglehart, Nevitte, and Basanez 1996). It is also believed to explain the shift away from more traditional notions of representative democracy towards those of direct, engaged citizen participation (Graham and Phillips 1998).

282 Julia Abelson and John Eyles

Changing views about public participation are also shaped by shifts in political discourse (e.g., the relative roles of the market versus the state in service provision), structural changes linked to changes in public-values systems, and evolving concepts of citizenship. For example, the post-Second World War period leading up to the establishment of many of Canada's significant social welfare programs was a time of collective problem solving and 'pan-Canadian citizenship' (Maxwell 2001,5). The poor health of army recruits, the ravages of the Depression, and the inadequacy of employment wages to provide basic health propelled public opinion and politicians towards the creation of new economic and social policies that would ensure access to basic government services and, in turn, provide the seeds for a national health insurance program (Taylor 1987). These collective policy goals were pursued alongside unprecedented levels of affluence and expanded educational opportunities over the next twenty-five years (into the 1970s) and dramatically changed social structures characterized by a strong welfare state and the collective values of universality, equity, and equal opportunity. Over time, however, these collective goals have been replaced by more individualistic goals, shaped by economic conditions, periods of retrenchment, and the dominance of neo-conservative values, followed by the introduction of market forces into public-sector management (Nevitte 1996). These individualistic goals have, in turn, inspired new public-participation ideals of increased public sector accountability and a customer-service orientation espoused by new public-management principles (Osborne and Gaebler 1992; Borins 1995). The 'new public participation' also hinges on a perceived need for social capital to enable citizens in communities and organizations to solve problems collaboratively (Putnam 1993). Associational relationships and civic participation are thus seen as being built on mutuality, that is, trust (formed and displayed through personal and social relationships and in institutions) and commitment to the common good that may be manifested in commitment to community or local-level affiliations or governing institutions (Giddens 1991; Veenstra and Lomas 1999). These calls for increased civic participation, capacity building, and the creation of social capital are a partial response to the rise of individualism in the 1980s and view a re-created community as the cornerstone to improvements in social and economic conditions (Putnam 1993; Sandel 1996; Bellah 1985).

Public Participation and Citizen Governance 283 Participation and Governance Ideals in the Health System To what extent do new public-participation and governance ideals resonate with those in the health system? As its broad goals, public participation in the health system is primarily concerned with: • improving the quality of information regarding the population's needs and preferences; • encouraging public debate over the fundamental direction of the health system; • ensuring public accountability for the processes within and outcomes of the system; and • protecting the public interest. Taken together, the pursuit and achievement of these goals are indicators of the health system's performance with respect to desired goals. In addition, a strong democratic presence in and commitment to the governance of public institutions are also believed to contribute to the development and maintenance of a civic culture that promotes the values of solidarity and tolerance, also a potentially desirable goal for the health system. As we discuss in the following sections of the paper, public participation in its various forms may both contribute to and thwart the achievement of these desired goals. Public Participation and Improved Information • A basic objective of public participation is to obtain information from the public about its needs and preferences with respect to a responsive health system. This objective was the basis for the creation of some of the earliest citizen-representative health authorities (e.g., health systems agencies in the United States, community health councils in the United Kingdom, and district health councils in Ontario) (Checkoway 1981; Checkoway and Doyle 1980; Klein 1976; Tuohy and Evans 1984). Citizen users are often the most appropriate individuals to provide this information since they can articulate service needs, gaps in service availability, or problems with service delivery. The public, generally speaking, feels comfortable providing this type of information and believes it is well suited to the role (Litva et al. 2002; Abelson, et al. 1995). This type of involvement can contribute to more effective governance when users' views have been honestly incorporated into the

284 Julia Abelson and John Eyles

decision-making process and when a range and diversity of views and types of users are involved. It also requires sensitivity to the vulnerability of service users who may be reluctant to 'speak out' for fear of losing service. Embracing a customer-centred, patient-rights movement in a health system historically dominated by a hierarchy of professions, therefore, requires a significant shift of mindset. Public Participation, Public Values and the Values of the Health System The public may be a crucial 'idea source' when it comes to expressing the values that should root, and ultimately, guide the health system. Yet, creating a transparent process for involving the public in this way can be difficult, challenged by a complicated relationship between public values, social institutions, and changing political discourse. For example, it is not clear whether the post-Second World War consensus on the solidarity principles that shaped the Canadian health insurance system (e.g., equity, universality, and accessibility) is still held by the vast majority of Canadians or if the consistently strong support for a universal health system has been shaped by the system itself and its symbol of Canadian identity (Giacomini et al. 2003; Maxwell 2001; Health Action Lobby 1992; National Forum on Health 1997). Values may also be shaped by interests such as those held by professional organizations looking to protect their organizational culture, industry stakeholders, and consumer- and patient-advocacy groups seeking to mobilize support for or opposition to various policy proposals. The policy elite also wields influence over beliefs and, in turn, support for their policy agendas. All groups use the media as the primary vehicle for interest-group 'claims-making' (Best 1989) and policy-elite 'crafted talk' (Jacobs and Shapiro 2000). The establishment of commissions, such as the Commission on the Future of Health Care in Canada, provide additional opportunities for shifting beliefs through the commissioning of ideas (Jenson 1994) and identifying the tensions underlying these values and interests, as well as the potential methods for resolving them. In these instances, the lines between public 'pawn' and 'participant' become blurred. To illustrate this complex relationship, consider recent surveys, opinion polls, and public dialogues that have focused on Canadians' eroding confidence in the health care system (Donelan et al. 1999), the 'future sustainability' of the health care system, and attitudes towards the expansion of insured health services or towards increased privatization (PriceWaterhouseCoopers 1997; Ekos 1999). These evaluations

Public Participation and Citizen Governance 285 of system performance and deterioration are efforts to gauge public opinion, but they also have the potential to shape Canadians' values toward these issues. Furthermore, predominantly attitudinal surveys do not reflect on how the public has reached these views. More detailed analyses of public values towards the health system indicate that the public's values are not reflected as closely in the current system as they once were. More prevalent today are the values of selfreliance, autonomy, and choice as well as those of compassion and future investment (Maxwell 2001; Mendelsohn 2002). The current values discourse is also more pluralistic than that which prevailed when Canada's social programs were being established. Now, there is a pluralism of values across population groups such as the young and old, the affluent and the poor, and across regions and governing parties (i.e., those that have embraced market-style reforms versus those that support a large role for the state). This pluralism complicates efforts to pursue unified social policy goals, to maintain a collective approach, and to encourage, through participation, the expression of democratic values (Maxwell 2001; Mendelsohn 2002). Within this values discourse, questions arise about the compatibility of choice and autonomy with socialinsurance models (see Klein 2001). Given that collectivist values appear to be less dominant than they once were, the health care system and its principles may need reinventing to emphasize choice and autonomy within a model of compassion and investments for all Canadians. It is clear that the public can play an important role in shaping the values expressed by the health system, but there is no single way to elicit or articulate a nation's collective values. We are often interested in ascertaining Canadians' values towards the health system with little recognition that Canadians' values exist in individuals and communities, and that there may be conflicts between individual and community values with no straightforward way to aggregate from one level to the other (Giacomini et al. 2003). Furthermore, it is unlikely that publicparticipation processes will be sufficient for encouraging the expression of democratic values given the current trend towards more pluralistic and individualistic values. Public Participation and Commitment to Public Institutions An increasingly fractured set of public values requires a highly effective set of institutions to manage conflict and find common ground to inform public-policy decisions. Our earlier discussion described a citizenry disenchanted with its public institutions and the means available

286 Julia Abelson and John Eyles

for public participation. Political theorists such as Rousseau and Barber argue that participation contributes to an educated and engaged citizenry potentially supporting collective problem solving and public institutions (Barber 1984). Other arguments suggest that, over time, the public moves from shallow 'public opinion' to deeper, more reasoned 'public judgment' and becomes more willing to take responsibility for its views and more committed to the policies it supports (Yankelovich 1991). At present, the interested public appears eager to participate in meaningful ways but is neither active nor engaged, owing in part to weak institutions. To what extent does this phenomenon exist in the health system and what does 'public commitment to public institutions' mean in the context of the health system? At a macro level, commitment to public institutions might be considered in terms of support for a publicly funded and administered health system. Through the participatory process, citizens become more knowledgeable about the health care system and more interested in pursuing collective problem solving through public means and shared resources, and, hence, develop an increased commitment to a publicly funded system. While this argument is reasonably compelling, it is not clear whether or how much public participation is necessary to achieve this kind of commitment to public institutions, although it may be a useful mechanism for mitigating the erosion of this commitment. At the meso level, public participation in a regional health authority might also promote greater commitment to public institutions. For example, the experimentation with elected health board members in Saskatchewan, Quebec, and Alberta represents one method for fostering public commitment to public institutions, in this case, the health system. We discuss the performance of such mechanisms in the next section. The opportunity to foster increased commitment to public institutions is perhaps most likely to be realized at the micro or local-governance level through civic participation such as community associations, community health centre boards, and community health councils. In the United Kingdom, citizen-driven community health councils have had a long history of providing input into local health governance structures (Klein and Lewis 1976). In Canada, community health boards were established in Nova Scotia as arms of the regional health boards in the 1990s and the newly established district health authorities, vested with the responsibility for developing community- service plans. Despite these examples, regionalization of the health sector in

Public Participation and Citizen Governance 287 Canada has led to the dismantling of hundreds of local health governance structures and traditional forms of civic participation (Veenstra and Lomas 1999; Lomas 1997a). A further challenge is the complexity of some public-policy sectors such as health care, which pose a barrier to increasing the civic competence necessary for more than token citizen participation (Dahl 1992). Public Participation for Efficiency or Efficient Public Participation? If public participation and democratic processes are valued as critical inputs to policy decision making and, consequently, health system performance, then it is also plausible to consider how public participation contributes to the 'efficient functioning' of the health system. Once again, our examination of this relationship is challenged by the different underlying goals of participation. For example, public participation is often viewed as an end in itself, a desirable goal for democracies and public policy making and, as such, does not demonstrate obvious benefits beyond the achievement of democratic goals. In the early 1990s, public participation was seen as an instrument for achieving broader health system goals such as improved service coordination and integration (i.e., efficiency) through regionalization and devolution. Considered in this way, what aspects of efficiency does public participation have the potential to influence? In economic terms, there are three different types of efficiency to consider with respect to the provision of public goods such as health and health care: • Technical efficiency - for a given output, the inputs required to produce it are minimized or the maximum output is produced from a combination of inputs. • Cost-effectiveness - similar to technical efficiency, but the cost of the inputs is also accounted for (i.e., for a given output there will be a specific combination of inputs that will be cost-effective) - 'doing things right/ • Allocative efficiency - links supply and demand by considering the values and preferences of those consuming the outputs (i.e., produce types and amounts of output that people value most) - 'doing the right things/ Applying the first two concepts of efficiency to the case of public participation might push us to look for technically efficient ways to involve the public to produce a given output. But the public-participa-

288 Julia Abelson and John Eyles

tion literature rarely describes examples of technically efficient or costeffective participation although there is discussion of effective participation processes (Beierle and Cayford 2002; Rowe and Frewer 2000; Petts 2001; and Webler, Tuler, and Krueger 2001). A commonly held view is that democratic processes, at best, slow decision making but are essential to ensure that legitimate decisions are made, and at worst, can significantly alter or derail a decision-making process if not carefully 'managed' (Abelson and Lomas 1996). In short, public participation is rarely viewed as compatible with the technically efficient or cost-effective functioning of the health system. Of greater relevance is the notion of allocative efficiency, which considers both the supply/production side of the equation and the demands, preferences, or values that shape what is produced. The essence of this argument is that information obtained from the public about values, needs, and preferences can contribute to the efficient allocation of societal resources (see Hurley, Birch, and Eyles 1995). Thus, public participation may potentially contribute to the effective performance of the health system by helping create a fully informed citizenry, transparency, and, ultimately, accountability, and in this way contributes to the achievement of efficiency goals. The Darker Side of Public Participation While it is often assumed that public participation is inherently beneficial to policy decision making, some instrumental goals for public participation have the potential to thwart its well-intentioned promotional efforts. A community's desire to pursue a narrowly defined set of collective interests, for example, often mobilizes citizens to participate actively to achieve their goals. In the education sector, this has manifested itself in pressure being exerted among some groups to opt out of the public education system in pursuit of greater choice (see Stein 2001). The social cohesion fostered within these communities threatens broader collective values of solidarity and tolerance. There are perhaps fewer examples of communities 'opting out' within the health sector, but there is little doubt that the intense mobilization witnessed in response to the threatened hospital closures throughout the 1990s entrenched values of intolerance and parochialism within some communities, particularly those where social cleavages previously existed (Abelson 2001). Public participation at regional and local levels, furthermore, may weaken the attachment to larger-scale public systems and exacerbate

Public Participation and Citizen Governance 289

current jurisdictional conflicts in Canada between federal-provincial and provincial-local roles in the health system. The emotional attachment that communities develop towards local institutions such as hospitals has clearly contributed to the inefficient allocation of resources. Local hospital restructuring experiences across the country are excellent examples of this local-resource attachment phenomenon where communities have fought to keep multiple hospital sites open at the expense of other system goals such as expanded community care (Abelson 2001; Lepnurm and Lepnurm 2001). An increased emphasis on direct democracy at the expense of representative democracy is another double-edged sword. Mounting concerns about governments being too far removed from citizens and about citizen representatives (i.e., elected public officials) inadequately representing citizen interests or having too many competing interests to represent have led to calls for greater and more direct citizen involvement in democratic processes through various forms of public participation and citizen engagement. A shift from the historical tradition of rational expert-driven to greater citizen-controlled decision making does not guarantee improvements and may lead to a requirement for greater resource expenditures in the process of training and preparing citizens for their decision-making roles. In this first section of the paper, we have considered the potential contributions of public participation to the achievement of different health system goals (e.g., efficiency, commitment to public institutions, expression of democratic values). In doing so, we underscore the conflicts that arise in pursuing different health system goals, the multiple goals of public participation itself, and the positive and detrimental influences that public participation may have on the achievement of desired health system goals. The Canadian Experience with Public Participation in the Governance of the Health System Different Means for Different Ends The first calls for public participation in Canada's health system date back to the 1970s when citizen representation on regional health services delivery or administrative boards was recommended by early reports on health system reform (Community Health Centre Project 1972; Foulkes 1973) and implemented, in the form of district health

290 Julia Abelson and John Eyles

councils in Ontario (Tuohy and Evans 1984). Citizen participation was viewed as a means for rationalizing, integrating, and coordinating health services through local structures that would, through community representation, better respond to local health needs. Outside Ontario, which has now had close to thirty years of experience with district health councils and consumer representation, experiments with regional-governance structures did not occur until the 1990s. Like its 1970s predecessor, the devolution movement of the 1990s (and accompanying calls for increased public participation) was unanimously accepted by provincial policy makers in the absence of supporting evidence or well-articulated plans for its widespread use. Indeed, an international review of devolution initiatives found few studies of its impact on government performance with equivocal results about its merits or weaknesses (Canadian Medical Association 1993). Other evidence questioned the ability of these newly formed structures to achieve community control in the health care sector (Marmor and Morone 1980; Bjorkman 1985). Calls for greater community control in the health sector signalled a major departure from previous decades of expert and bureaucracy-controlled decision making and health system governance. The prevailing view had been that only professionally controlled governance was compatible with the pursuit of health system efficiency and equity goals. In the early 1990s, provincial governments called for a move away from this privileged relationship between health care bureaucracies and dominant structural interests, namely, physicians and hospitals, to increased community control and the creation of new governance structures (Rasmussen 2001). While community control through citizen governance has been more rhetoric than reality in much of Canada's public health system, we note a fundamentally different history of public participation in Quebec, where recommendations of the Castonguay-Nepveu's Commission of Enquiry on Health and Social Welfare (1967-71) identified the importance of public and user involvement in health care decision making and positioned such involvement centrally within its reforms by legislating public participation through regional health board elections (Charles and DeMaio 1993; Forest et al. 2000). Citizen governance has also been a feature of the primary care delivery system in Canada, although it has been largely relegated to the role of an 'alternative' model to the traditional physician-governed organizations. Centres locaux de services communautaires (CLSCs) were

Public Participation and Citizen Governance 291

introduced in Quebec in 1972 to provide primary medical and social services to geographically defined populations. Community health centres were introduced in Ontario in the late 1970s as communitygoverned, globally funded primary health care organizations designed to improve access to health and health care for the socially disadvantaged and difficult-to-serve populations (Hutchison, Abelson, and Lavis 2001; Church 1994). While citizen governance may be a sentinel measure of democratic control of the health system, the public participates in a variety of other ways that have the potential to influence decision making. Public-consultation exercises have been the traditional mechanism through which the public provides input to the health system. These are typically government-initiated processes conducted at various stages of a decision-making process to inform planning, priority setting, or resource-allocation decisions. There is a rich tradition of public consultation in Canada at all levels of government. In the health system, both federal and provincial governments and regional and local health authorities (through community, district, and regional health councils and boards) routinely seek the public's input, as do higherprofile (and ever-present) provincial health commissions, to inform their recommendations.3 Public representation on regulatory bodies has long been used as a method of ensuring professional accountability to the public and protecting the public's interest. Legislation governing regulatory bodies stipulates public representation on statutory committees and the governing council, but the degree and type of public involvement on these bodies varies provincially and by health-professional sector. The experiences with each of these public-participation approaches are assessed in the following sections. Public Participation in Regionalized, Devolved, and Community Health Governance Structures: Promise and Reality Canada's ten-year experiment with regionalization and devolution and attempts to increase democratic control over health system governance provides an excellent laboratory for judging the public-participation experience. Unfortunately, there is little solid evidence to make any claims about whether and how increased democratic control has contributed to improved system performance. This situation is not unique to Canada, or the health system in particular (Zakus and Lysack 1998).

292 Julia Abelson and John Eyles

What can be assessed with reasonable rigour, however, is the extent to which increased democratic control has been achieved; how public consultations have contributed to the quality of information used in decision making; and how they have contributed to the public's sense of ownership and commitment to the health system. Elected regional health boards were established at different times in Saskatchewan and Quebec as tools for achieving political accountability and, hence, enhanced responsiveness to local needs. By selecting board members through an electoral process, health boards would be directly accountable to their communities. The experience with these elections is that only those who had the most to lose or gain (i.e., concentrated interests) stood for election. In the health sector, these are typically those who derive their incomes from the system. The communities' diffuse and fragmented interests pose further barriers to participation (Alford 1975; Marmor and Morone 1980; and Tuohy and Evans 1984). Low voter turnout has been the norm (e.g., 10 per cent in Saskatchewan's 1999 board elections), with a tendency for higher turnout in rural communities and in highly politicized communities where major cuts were being contemplated (Lomas 2001; Rasmussen 2001). A 1997 survey of Saskatchewan board members found that 63 per cent of members believed that health care reform had increased local control over health care services and 62 per cent believed that the quality of health care decisions had improved under the devolved authorities (Lewis et al. 2001). When compared with the experiences found elsewhere in Canada, however, 'there was little that could be attributed exclusively to the electoral process' (346). Furthermore, the experience has demonstrated that health board elections are costly and cumbersome, they produce low voter turnout, and they have failed to foster a more active, engaged citizenry committed to collectivist goals. In light of these experiences, their continued use should be questioned if efficient, effective participation and public commitment are desired goals. The Quebec reforms of 1991 'Une reforme axee sur le citoyen' opened the door for broader citizen involvement. Created for the purposes of bringing the health system closer to the population, the establishment of Quebec's regional health and social services boards was also seen to play a role in breaking the monopoly that interest groups appeared to have over decision-making processes (Forest et al. 2000).

Public Participation and Citizen Governance 293

The goal of increasing community participation through board membership has been achieved, but establishing community control and empowerment has clearly not occurred (O'Neill 1992; Godbout 1981, 1987; Eakin 1984a, 1984b; Forest et al. 2000; and Abelson et al. 2002). The Nova Scotia experience with regional health boards was similar despite the same representation from consumers and providers at the community health board level and a two-thirds to one-third majority of consumer representatives at the regional health board level. Increased public representation is no guarantee for meaningful participation in the context of an 'implicit power hierarchy in health care ... made up of prearranged tiers of professionals, administrators and consumers'; the lack of consumer ties to constituency-support networks; and disparities in technical and administrative skills between consumers and professionals (Sullivan and Scattolon 1995). Furthermore, early accounts of Saskatchewan health board activities indicate that these were few open meetings, no published agendas, and no opportunities for the public to comment on plans. The only opportunities for citizens to participate came at election time (Rasmussen 2001). In British Columbia, early enthusiasm for a new 'closer to home' form of citizen participation and an open process for designing the shape of the regional health authorities quickly gave way to frustrated participants who no longer felt that they were able to influence the boards the way they had envisioned at the outset (Prankish et al. 2002). Fractured decision making and an intensely competitive relationship between politicians and professionals have provided little scope for collaborative problem solving and caused the public to withdraw from their participatory activities, suggesting that 'if there is any crisis in health care, it is in the governance of the health system' (Maxwell 2001, 22). Thus, citizen governance in the health system (and in many other public policy sectors) has remained an elusive goal for decades. Multiple conflicting goals of participation, the reluctance for those with concentrated interests to allow their futures to be determined by others, and the reluctance of the public to take responsibility for decision making have presented significant barriers. Many organizations in the health system have recognized these challenges and are seeking ways to involve the public more meaningfully in their decision-making processes while moving away from the pursuit of goals of increased community control (Regionalization Research Centre 2000; Abelson et al. 2002).

294 Julia Abelson and John Eyles

Public Participation on Regulatory Bodies: Governing the Health Professions Public representation on regulatory bodies has long been used as a method of ensuring accountability of health care professionals to the public but there have been few attempts to document or assess the public's experiences with this form of participation.4 It is difficult to determine, therefore, whether or how the broader experiences of citizen governance of the health system is generalizable to that of public participation on regulatory bodies. A review of Governing Councils of Colleges of Physicians and Surgeons across Canada5 reveals considerable variation in the governing council composition of these bodies, with a range of 14 to 25 per cent public membership on seven councils, 30 per cent public representation on one council, and a 40 to 50 per cent range of public representation on two councils. While the number of public representatives alone is no guarantee of their empowerment, a commitment to greater than 40 per cent public membership by the Ontario and British Columbia colleges is an important symbol. Consulting the Public on Decision Making in Health Care Local efforts to consult with the public during health planning and more recent priority-setting and resource-allocation decision-making processes have provided valuable case studies of public participation. The health services restructuring exercises that have taken hold of communities across the country over the past five to seven years are examples of intense, issue-based participation rarely seen since the health care system's establishment. Mobilization occurred in response to threats of hospital closures and provided several lessons about issueand interest-based public participation. First, efforts to involve the public locally have largely failed with respect to the outcome-oriented goals of participation (i.e., contributing to a decision-making process geared towards shifting resources from hospitals to communities) but have been hugely successful in mobilizing community-wide participation. Judged in these terms, public participation in local health services restructuring processes contributes little, if anything, to the achievement of efficient and equitable health system goals or outcomes and may, as discussed earlier, contribute to greater inefficiencies in resource allocation. A potential benefit of these restructuring processes may, however, be the fulfillment of

Public Participation and Citizen Governance 295

participation's educative goals through the community's enhanced understanding of the complexities of decision making in health care and the relative contributions of hospitals and community-based service delivery in meeting population health needs (Abelson et al. 2002). The failure of these public-participation processes (with respect to efficient resource-allocation decisions) must be considered in designing future public participation processes. Characteristics of these failures include: the success of the media and stakeholder groups in framing of the issue as a decision about health care cuts and government retrenchment despite local health council attempts to reframe the debate as one of resource reallocation; the absence of legitimate deliberation (i.e, shared learning, considered public judgment) because of the abstractness of issues such as primary care reform and communitybased delivery compared to highly recognizable hospital-based care; the lack of legitimate opportunities for public participation with provincial governments making the final decisions and controlling the public-consultation agenda; the lack of clarity about the public's role as consultant versus decision maker; and the deference displayed to, and the power wielded by, the local health care elite, resulting in restrained attempts to offer and promote alternative solutions (Abelson 2001). Furthermore, communities (defined geographically or by shared needs or characteristics) are not all alike and engage in different forms of participation shaped by different community characteristics such as language and religion, education and income levels, and the presence of shared community values. The presence of sharp social cleavages, for example, can mobilize intense, issue-driven participation that may conflict with solidaristic principles (Abelson 2001). It is questionable, therefore, whether the public and geographic communities should be involved in local cost-cutting exercises such as decisions about which local hospitals should be closed. These are 'tragic choices' that communities are not likely to make comfortably and should not, therefore, be expected to make (Calabresi and Bobbitt 1978). Public-consultation processes in the health sector have been fraught with problems that have produced angry, frustrated citizens and hindered policy development. Consultations that are presented as openended processes but framed very narrowly raise the ire of participants who judge the process to be a 'done deal' (Deber and Williams 1995; Aronson 1993). These experiences have resulted in calls for public-consultation processes to have greater clarity of purpose and clearer lines

296 Julia Abelson and John Eyles

of accountability between the consultation and the final decision taken (Abelson et al. 2002). The Public's Perspective on Public Participation With the exception of satisfaction surveys and some limited polling data, we know little about how members of the public views themselves in their actual or potential roles as citizen governors, resource allocators, or health system consultants. The public's apparent dissatisfaction with current political institutions and participatory opportunities hardly provides any concrete evidence on which to base future design processes. The limited evidence that does exist points to the public as 'reluctant rationer' of services (i.e., unwilling to take responsibility for setting priorities either among broad health service categories or specific services) (Lomas 1997b; Litva et al. 2002). Evidence also suggests that those citizens willing to consider this responsibility have an insatiable appetite for information about the costs and benefits of various programs and services, information that may be unavailable, costly to obtain, or difficult to interpret (Abelson et al. 1995; Lomas 1997b; and Litva et al. 2002). Participants cite generally positive experiences with their involvement in some consulting roles although they are becoming increasingly impatient when they perceive themselves to be a rubber stamp for decisions that are already taken (Coote and Lenaghan 1997; Litva et al. 2002). Experiences with public involvement in deliberative exercises such as citizens' juries and panels have generated positive feedback from participants who welcome the opportunity to become more informed about their local health system; however, these same people also express concerns about the outcome of the process given the substantial time investment. Participants in deliberative processes also tend to emerge from these experiences with a fuller understanding of the complexities of decision making in the health sector and, hence, renewed respect for existing decision makers (Abelson et al. 1995; 2003). A recent study of the participation in actual health care priority-setting decisions found that public representatives view themselves as contributing to both the process and the outcomes of the decision-making process (Martin, Abelson, and Singer 2002). Another study of public preferences for participation in the United Kingdom aptly summarizes the public mood suggesting that citizens are seeking 'accountable consultation' - consultation that provides them with

Public Participation and Citizen Governance 297

opportunities to express their views (but not the responsibility for decision making) and a guarantee that their contributions will be heard and an explanation provided for the decision ultimately made (Litva et al. 2002). The Future of Public Participation and Citizen Governance in the Health System The signing of the Social Union Framework Agreement in February 1999 has committed all provincial and territorial governments (with the exception of Quebec) to a variety of principles for treating and meeting the needs of Canadians. The framework also attempts to reflect and give expression to Canadians' fundamental values. Relevant to the health sector is the commitment to ensuring 'adequate, affordable, stable and sustainable funding for social programs' (Social Union Framework 1999,1). Relevant to public participation in the health system and its contribution to the efficient functioning of the health system is the agreement for governments to improve public accountability and transparency by ensuring 'effective mechanisms for Canadians to participate in developing social priorities and reviewing [health system] outcomes' (Social Union Framework 1999, 3). As with previous commitments to public involvement in the health sector, the commitment described in the Social Union Framework is well intentioned but may be doomed to fail without carefully articulated goals and means. A major challenge remaining is the establishment of transparency about the purpose and mandate of the framework agreement itself. Intergovernmental agreements are extremely difficult for the public to digest. They are opaque products of complicated institutional arrangements far removed from the public's day-to-day interaction with the health system, and, therefore, at risk of generating expectations that will be left unrealized or simply dropped from the public-participation agenda. This most recent effort to institutionalize public participation in health system decision making warrants attention to the weaknesses that have plagued previous public-participation efforts: • the domination of citizens by powerful groups interested in involving the public when it suits their purpose; • the tendency of policy makers to tout citizen governance as a critical element to achieve more responsive decision making while using

298 Julia Abelson and John Eyles

these structures as instruments of cost cutting and restructuring; • the ability for only the most educated and sophisticated and arguably the most unrepresentative and biased 'publics' to participate as citizen governors; and • the increasing cynicism of a public weary of pre-determined illegitimate public-consultation processes, reluctant to take responsibility for decision making and seeking more accountable consultation. A more positive set of local experiences surely exists across the country but these are more difficult to identify and assess, in part because they are often regarded as 'trivial' or 'marginal' examples of local citizen-initiated processes, despite their success in effectively mobilizing communities and contributing to collaborative problem solving and to a greater sense of commitment and ownership of the health system. They may not even be immediately recognizable as examples of public participation in the health system (e.g., volunteer activities, local fundraising, resource pooling among community health agencies); instead, they may be seen as a form of civic-minded participation inextricably linked to trust, commitment, and collaborative problem solving. We discuss these examples in more detail in subsequent sections of the paper. In summary, public participation in the governance of the health system, despite its potential, has contributed little to: • efficient functioning of the health system; • strengthening the commitment of citizens towards health programs; or • encouraging the expression of democratic values of solidarity and tolerance in Canadian society. In the remaining sections of the paper, we focus on the development of strategies to realize public participation's potential for achieving these goals, which we believe are at least partially attainable. Which Way Forward? Recommendations for Enhancing Democratic Processes and Citizen Governance in the Health System We offer a set of guiding principles as well as selected strategic and technical recommendations designed to strengthen democratic processes towards the achievement of improved citizen governance in the health system.

Public Participation and Citizen Governance 299 Guiding Principles Recognize the Multiple Goals of Participation and Their Inherent Tensions: As long as there are competing objectives for public participation, there will be difficulty reaching agreement on how best to proceed. Designing equitable public-participation processes, for example, may not be compatible with public-participation processes that can contribute to the efficient functioning of the health system, or with the goal of promoting public participation to foster greater commitment to public institutions and the expression of democratic values. The specific objectives for public participation must be clearly articulated for different governance levels, different participants, and different public-participation processes. The trade-offs of pursuing one goal for the achievement of another must be recognized. Acknowledge a Major Role for Interested Parties and Capitalize on the Public's Tendency towards Issue-Based Participation: We address several issues pertinent to an enhanced role for individual public participants in the health system. First, we believe that it is unreasonable to expect 'disinterested' citizens to function as citizen governors (i.e., public members on health councils and boards) and advocate instead more flexible definitions that assume that citizen participants are interested members of their community, not simply users or consumers as they are often labelled. The broader public-participation literature and, more specifically, the health care participation literature, have clearly emphasized the link between interests, issues, and participation among the public (Abelson 2001; Abelson et al. 1995). It may also be the case that only 'interested' citizens and community members will be willing to dedicate their time to the more intensive deliberative approaches advocated by citizen-engagement proponents. The task that remains, then, is determine what constitutes 'interested,' a difficult concept to sort out in the health arena. The conventional view of interested parties in the health sector has been based on the notion of individuals who derive their income from the health care system or who have a particular stake in the health care system, and for whom active participation enables them to influence decisions based on a narrow set of interests (e.g., hospital administrators seeking to block the closure of their hospital). Ideally, interests should be conceived of much more broadly to capture interest in collectivist goals such as the efficient and equitable functioning of the health system. To operation-

300 Julia Abelson and John Eyles

alize this concept at the community level, for example, a group of interested citizens might be brought together from a broad base of community organizations that are not directly involved in health services delivery but for whom the health system is seen as an important community resource. The selection of participants who pursue collective over individual interests is both difficult to achieve and may contradict some democratic principles (i.e., providing equal opportunities for public participation) but may, nevertheless, be an important vehicle for promoting this particular public-participation goal. Moving from the interested to the 'disinterested citizen/ we note that this person is not likely to participate unless affected by an issue. When she does get involved, she wants to know why she is being asked to participate and that her participation will make a difference (i.e., that it will be considered in the decision-making process). She may prefer to provide the input without taking overall responsibility for the final decision or may be willing to share a more expanded role with experts and elected officials. In essence, this participant is seeking 'accountable consultation/ which she perceives to be missing from the health system currently (Litva et al. 2002; Abelson et al. 1995). The movement to replace conventional standing committees with issuebased task forces and working groups in a variety of health planning organizations is one way to begin to address this demand for more 'purposeful consultation' although its effects on decision making are unknown (Abelson et al. 2002). Strategic Recommendations Establish and Enhance a Role for 'Health System Advocates' at Local/ Regional, Provincial, and National Governance Levels: At the citizen-user level, patient charters have become a popular tool for institutionalizing patient rights and responsibilities although the relative weight given to rights versus responsibilities in these charters can differ enormously. The key is to provide citizen users with adequate means for communicating their concerns about their care through accessible, easily identifiable channels (i.e., they need to know where to go). In some instances, these processes may also need to ensure an adequate separation between the 'the complainer' and the service provider or funder, particularly when the users are vulnerable clients such as the elderly and the mentally ill. Such an approach has been adopted in the recently released Report from the Premier's Health Quality Council in New

Public Participation and Citizen Governance 301 Brunswick, which has proposed a charter of rights and responsibilities (Government of New Brunswick 2002). 'Health system advocates' have been proposed as a supporting mechanism of the charter to 'facilitate citizen interface with the system ... when issues need to be addressed, when information that is not easily accessible is needed, or when a citizen feels he/she is not receiving appropriate information or treatment' (Government of New Brunswick 2002, 85). Advocates would be appointed at both the regional health authority and the provincial level, with appropriately integrated functions to ensure clarity of roles and responsibilities. A variant of this model could also be implemented at the national level to create, in effect, a 'guardian' of the public health system. The National Round Table on the Environment and the Economy (NRTEE) is one such model. It is an independent advisory body that provides decision makers, opinion leaders, and the Canadian public with advice and recommendations for promoting sustainable development. A parallel health roundtable could be created that would promote the health system goals of high quality, accessibility, and efficiency. Opportunities for citizens to contribute directly through complaints and submissions should still be available through fully transparent mechanisms such as the dispute-settlement procedures established under the Social Union Framework Agreement or parallel provincial mechanisms. Develop and Cultivate Civic Infrastructures As 'Institutions' for Democratic Participation: If we accept the notion that the existing governance structures in the health system have failed to foster participatory decision making and collective problem solving (and perhaps should never have been relied upon to achieve these goals), then how are the public's calls for meaningful and accountable participation to be met? In an increasing number of public-sector areas, existing social and civic networks are being relied upon to build the needed capacity for participation and shared decision making. Civic organizations are considered important and attractive channels of public participation, examples of 'better government by offering deliberative consultation involving extensive interests' (Lindsay 2000). Civic participation and strong associational networks can also increase the effectiveness of democratic citizenship and governance (Putnam 1993). How might this work in the health system? 'Building social capital and civic infrastructure is largely a matter of removing the constraints

302 Julia Abelson and John Eyles

that often truncate that self-organizing process, and of improving the space it needs to flourish' (Rosell 1999, 50). One approach to removing these constraints might be to use regional health authorities to foster and build this civic infrastructure through user and citizen networks (Veenstra and Lomas 1999). Community health boards in Nova Scotia and citizen advisory committees in Saskatchewan are examples of these institutions, since they appear to be the right size to foster interaction around local health issues and have not, as some might have expected, functioned as veto points for decisions that are unpalatable to communities. These structures have flourished with a committed group of local leaders who have had public-participation experiences in multiple sectors and who, on an ad hoc basis, come together to address local problems.6 In the case of the Nova Scotia community health boards, these groups have grown from existing community networks and are now in a better position to understand and weigh the needs, preferences, and values of their constituents than could a nonlocal public body (see Mays 2000). This approach is not without challenges, including its reliance on communities with shared collective interests and associational networks that cross social classes. Communities with social capital (i.e., able to engage in collaborative decision making) are typically found in more homogeneous communities where there are fewer competing political goals that divide a community. Some communities, however, may overcome their divisions more easily than others, pointing, once again, to the presence of community values as distinct from individual values. This approach's strength may also be a weakness in its reliance on the same group of civic leaders to move from issue to issue and the potential this creates for volunteer fatigue, stifling innovation or consolidating power within a community elite. Efforts would need to be made to mitigate this risk. Technical Recommendations Improving the Legitimacy of Public-Participation Processes: Improved legitimacy of public-participation processes requires a clearer articulation of goals, intended products (i.e., why the public is being asked to participate), and selection of participants. This is necessary for transparency. Accountability for the public's involvement can be achieved by maintaining an ongoing dialogue with individual and community participants at various points in the decision-making process. But, as yet,

Public Participation and Citizen Governance 303

there is no consensus about which approaches are most effective for these different community and decision-making contexts. Citizen engagement and deliberative processes are widely viewed as promising solutions to a perceived 'democratic deficit' and to produce more legitimate participatory processes. Calls for two-way, interactive engagement are ubiquitous throughout the public-sector governance literature as a mechanism for fostering strengthened commitment to public institutions, shared values, and collective problem solving (Gastil 2000). Attempts to engage citizens through deliberative processes have been experimented with only recently in the health sector, and primarily in the United Kingdom, not in Canada (Mclver 1998). These processes remain vulnerable to powerful stakeholder interests who seek to initiate or sponsor them and thereby control the information that is selected and presented and that ultimately influences the outcome of deliberations (Dunkerley and Glasner 1998). While the principles of deliberation such as informed dialogue leading to reasoned public judgment have the potential to contribute to both instrumental (i.e, improved decision making) and developmental (i.e., commitment to health programs and policies; expression of democratic values) goals, mechanisms are needed to reduce their vulnerability to the same problems that have plagued traditional participation practices. Greater emphasis also needs to be given to the outcomes of these processes (i.e., producing concrete input, recommendations, or decisions) to ensure that they become more than a democratic luxury (Gastil 2000). Pursuing Efficient Participation: Public participation at all costs is not sustainable in a health system preoccupied with efficiency and Value for money/ Those designing public-participation processes must consider both the costs and the effects of these processes. Neither have been adequately addressed by the literature or decision-making organizations although participation researchers have begun to turn their attention to effective process (Rowe and Frewer 2000; Webler, Tuler, and Krueger 2001; and Beierle and Cayford 2002). Deliberative methods warrant particular scrutiny given the substantial costs associated with their implementation. With respect to allocative efficiency and public participation, consideration needs to be given to the role of an informed citizenry and how much information is required to achieve this goal. The evidence suggests that the public's demands for information to inform decision making are substantial. However, over time, as public opinion turns to more

304 Julia Abelson and John Eyles

informed 'public judgment/ the public assumes a greater responsibility for its views, is able to consider the trade-offs associated with different positions, and is willing to accept the costs and the benefits of a given public policy (Yankelovich 1991). A final consideration is that some of the most democratically driven participation processes (particularly those that produce issue-based community mobilization designed to protect the status quo) will lead to the inefficient allocation of resources. Conclusions The analysis and recommendations presented in this paper emerge from the mixing of conceptual and theoretical ideas, empirical research on the experiences with public participation in the health system, and some practical advice for enhancing democratic processes in the health system to contribute more effectively to the performance of the Canadian health system. Our review of the public-participation experience suggests that, at present, an increasingly pluralistic and fractured public does not appear, by its participation, to increase the efficiency or the broad public commitment to the health care system, or to deepen the values of solidarity and tolerance. However, there may be ways, via regional and local processes, in which citizens, motivated by interest, can nevertheless foster commitments to the system as a whole. More fundamentally, however, the desire for public participation to contribute to the efficient functioning of the health system must be balanced against the value of public participation in its own right and as a basic feature of democratic society. Although the contributions of public participation have not likely altered the health system's performance per se, it is conceivable that our health system might be worse off without the modest attempts to involve the citizenry that have taken place over the past forty years. If public involvement in the health system (however defined) is considered a value in its own right, and this view appears to be largely supported, the basis upon which we judge its success or failure, and the quality of the evidence used to make this judgment, needs more careful consideration. Participation may never produce greater efficiencies, nor is public participation likely a necessary condition for the efficient functioning of the health system. It has the potential, however, through the adoption of some of the recommendations made here, to strengthen citizen commitment to health programs and to encourage the expression of democratic values. As governments deal with increasingly challenging questions about

Public Participation and Citizen Governance 305 how to manage and sustain a high-quality, accessible, and efficient health system, the public will undoubtedly be asked and encouraged to participate in these decisions. An increasingly savvy public is likely to recognize if it is being asked to share the blame for the consequences of difficult and painful decisions or to be legitimately involved in charting the future direction of the health system. Decision makers would be wise to tread carefully as they determine how best to involve the public in this process.

NOTES We are grateful to two anonymous reviewers for their comments on an earlier version of this paper. We thank Patricia Collins, Colin McMullan, and Lydia Garland for their research assistance. We are also grateful to a number of researchers and decision makers across Canada with whom we have interacted over the past several years as part of a collaborative-research program on public participation and effective public consultation in health systems decision making (funded by the Health Evidence Application and Linkage Exchange Network, the Canadian Health Services Research Foundation, and le Conseil de la sante et du bien-etre du Quebec). We have found these interactions, over and above the published work that has arisen from this research program, extremely helpful in shaping some of the ideas expressed in this paper. 1 See Pateman (1970) for a detailed discussion of this issue. 2 Aristotle, Rousseau, and J.S. Mill discuss the educational and developmental benefits to the citizenry of citizen participation and its positive influence over citizens' commitment to public interests. 3 Recent examples include the Report from the Premier's Health Quality Council in New Brunswick, 2002; the Premier's Advisory Council on Health for Alberta, 2002; la Commission d'etude sur les services de sante et les services sociaux du Quebec, 2001; and the Fyke Commission, 2001. Throughout the 1990s, almost every provincial government established a commission or panel to propose health system reforms and some governments such as Nova Scotia and Ontario established provincial health councils as advisers to the provincial government. 4 Although we were unable to find any empirical studies of public- participation experiences on regulatory bodies, there are numerous accounts of public attempts to seek justice for medical errors that have fuelled patients'

306 Julia Abelson and John Eyles rights movements internationally and the establishment of patients' charters to address concerns about the health system's lack of public accountability. One of the most comprehensive reviews of a regulatory body from the public's perspective is an analysis of the United Kingdom's General Medical Council (see Stacey 1992). 5 These data were collected from websites and through personal contacts with the colleges or medical boards in each province. 6 These views reflect preliminary findings from a cross-jurisdictional public participation study led by J. Abelson and P.-G. Forest and funded by the Canadian Health Services Research Foundation.

REFERENCES

Abelson, J. 2001. Understanding the role of context in local health care decision making: Case study results from Ontario, Canada. Social Science & Medicine 53(6): 777-93. Abelson J., and J. Lomas. 1996. In search of informed input: A systematic approach to involving the public in community decision-making. Health Care Management FORUM 9:48-52. Abelson, J., et al. 1995. Does the community want devolved authority? CMAJ 153: 403-12. - 2002. Obtaining public input for health systems decision making: Past experiences and future prospects. Canadian Public Administration 45(1): 70-97. - 2003. Deliberations about deliberative methods: Issues in the design and evaluation of public participation processes. Social Science & Medicine 57(2): 239-51. Alford, R.R. 1975. Health care reform and structural interests. Excerpt entitled: Alternative views of the study of health care, in Robert Alford, ed., Health Care Politics: Ideological and interest group barriers to reform. Chicago: University of Chicago Press, 9-21. Aronson, J. 1993. Giving consumers a say in policy development: Influencing policy or just being heard? Canadian Public Policy 19,367-78. Barber, B. 1984. Strong democracy. Berkeley: University of California Press. Beierle, T.C., and J. Cayford. 2002. Democracy in practice: Public participation in environmental decisions. Washington, D.C.: Resources for the Future. Beierle, T., and D. Konisky. 2000. Values, conflict, and trust in participatory environmental planning. Journal of Policy Analysis and Management 19(4): 587-602. Bellah, R. 1985. Habits of the heart: Individualism and commitment in American life. Berkeley: University of California Press.

Public Participation and Citizen Governance 307 Best, J. 1989. Claims, in Joel Best, ed., Images of issues: Typifying Contemporary social problems. New York: Aldine de Gruyter, 1-3. Bjorkman, J. 1985. Who governs the health sector? Comparative European and American experiences with representation, participation and decentralization. Comparative Politics 17: 399-420. Borins, S. 1995. The new public management is here to stay. Canadian Public Administration 38(1): 122-32. Calabresi, G., and P. Bobbitt. 1978. Tragic choices. New York: W.W. Norton and Company. Canadian Medical Association Working Group on Regionalization and Decentralization. 1993. The language of health system reform: Report of the working group on regionalization and decentralization. Ottawa: Canadian Medical Association. Charles, C, and S. DeMaio. 1993. Lay participation in health care decisionmaking: A conceptual framework. Journal of Health Politics, Policy and Law 18: 881-904. Checkoway, B. 1981. Citizens and health care. New York: Pergamon Press. Checkoway, B., and M. Doyle. 1980. Community organizing lessons for health care consumers. Journal of Health Politics, Policy and Law 5(2): 13-26. Church, W.J.B. 1994. Structural interests and health politics in Ontario: The case of community health centres, Ph.D. Thesis, University of Western Ontario, London, Ont. Community Health Centre Project. 1972. The community health centre in Canada. Ottawa: Information Canada. Coote, A., and J. Lenaghan. 1997. Citizens' juries: Theory into practice. London: Institute for Public Policy Research. Dahl, R. 1992. The problem of civic competence. Journal of Democracy 3(4): 4559. Deber, R., and A. Williams. 1995. Policy, payment and participation: Longterm care reform in Ontario. Canadian Journal on Aging 14(2): 295-318. Donelan, K., et al. 1999. The cost of health system change: Public discontent in five nations. Health Affairs 18(3): 206-16. Dunkerley, D., and P. Glasner. 1998. Empowering the public? Citizens' juries and the new genetic technologies. Critical Public Health, 8(3): 181-92. Eakin, J. 1984(fl). Survival of the fittest? The democratization of hospital administration in Quebec. International Journal of Health Services 14(3): 397-412. - 1984(b). Hospital power structure and the democratization of hospital administration in Quebec. Social Science & Medicine 18(3): 221-8. Ekos Research Associates Inc. 1996. Rethinking government. Ottawa: Ekos Research Associates. - 1999. Benchmark survey. Ottawa: Ekos Research Associates.

308 Julia Abelson and John Eyles Forest, P.-G., et al. 1999. Issues in the governance of integrated health systems. Report submitted to the Canadian Health Services Research Foundation. - 2000. Participation de la population et decision dans le systeme de sante et de services sociaux du Quebec. Study prepared at the request of the Conseil de la sante et du bien-etre du Quebec. Foulkes, R. 1973. Health security for British Columbians: Volumes 1 and 2. Victoria: Queen's Printer. Prankish, J., et al. 2002. Challenges of community participation in health-system decision making. Social Science & Medicine 54:1471-80. Gastil, J. 2000. Is face-to-face citizen deliberation a luxury or a necessity? Political Communication 17: 357-61. Giacomini, M., et al. 2003. The policy analysis of 'values talk': Lessons from the Canadian health reform. Health Policy, forthcoming. Giddens, A. 1991. Modernity and self-identity: Self and society in the later modern age. Stanford: Stanford University Press. Godbout, J. 1987. La democratie des usagers. Montreal: Boreal Express. - 1981. Is consumer control possible in health care services? The Quebec case. International Journal of Health Services 11(1): 151-67. Government of New Brunswick. 2002. Health renewal. Report from the Premier's Health Quality Council. Available at http://www.gnb.ca/op-cpm/ephqc.pdf. Graham, K.A. and S.D. Phillips. 1998. Making public participation more effective: Issues for local government, in Graham and Phillips, ed., Citizen engagement: Lessons in participation from local government. Toronto: Institute of Public Administration of Canada, 1-24. Health Action Lobby (HEAL). 1992. Medicare: A value worth keeping. Available at http://www.cna-nurses.ca/heal/healframe.htm. Hurley, J., S. Birch, and J. Eyles. 1995. Geographically-decentralized planning and management in health care: Some informational issues and their implications for efficiency. Social Science and Medicine 41(1): 3-11. Hutchison, B., J. Abelson, and J. Lavis. 2001. Primary care in Canada: So much innovation, so little change. Health Affairs 20(3): 116-31. Inglehart, R. 1995. Changing values, economic development and political change. International Social Science Journal 47(3): 379^103. Inglehart, R., N. Nevitte, and M. Basanez. 1996. The North American trajectory: Social institutions and social change. New York/Berlin: Aldine de Gruyter. Jacobs, L., and R. Shapiro. 2000. Politicians don't pander: Political manipulation and the loss of democratic responsiveness. Chicago: University of Chicago Press. Jenson, J. 1994. Commissioning ideas: Representation and royal commissions,

Public Participation and Citizen Governance 309 in S. Phillips, ed., How Ottawa spends 1994-95: Making change. Carleton University Press, 39-70. Klein, R. 2001. Estimating the financial requirements of health care. BMJ 323: 1318-9. Klein, R., and J. Lewis. 1976. The politics of consumer representation: A study of community health councils. London: Centre for Studies in Social Policy. Lepnurm, R., and M. Lepnurm. 2001. The closure of rural hospitals in Saskatchewan: Method or madness? Social Science & Medicine 52(11): 16891707. Lewis, S., et al. 2001. Devolution to democratic health authorities in Saskatchewan: an interim report. CMAJ 164(3): 343-7. Lindsay, I. 2000. The new civic forums. Political Quarterly 71(4): 404-11. Litva, A., et al. 2002. 'The public is too subjective:' Public involvement at different levels of health-care decision making. Social Science 61 Medicine 54(12): 1825-37. Lomas, J. 1997a. Devolving authority for health care in Canada's provinces: 4. Emerging issues. CMAJ 156(6): 817-23. - 1997b. Reluctant rationers: Public input to health care priorities. Journal of Health Services Research and Policy 2(2): 103-11. - 2001. Past concerns and future roles on regional health boards. CMAJ 164(3): 356-7. Mclver, S. 1998. Healthy debate? An independent evaluation of citizens' juries in health settings. London: King's Fund Publishing. Marmor, T., and J. Morone. 1980. Representing consumer interests: unbalanced markets, health planning and the HSAs. Milbank Memorial Fund Quarterly/ Health and Society 58(1): 125-65. Martin, D., J. Abelson, and P.A. Singer. 2002. Participation in health care priority setting through the eyes of the participants. Journal of Health Services Research & Policy 7(4): 222-9. Maxwell, J. 2001. Toward a common citizenship: Canada's social and economic choices. REFLEXION no.4. Ottawa: Canadian Policy Research Networks. Mays, N. 2000. Legitimate decision making: The Achilles' heel of solidaristic health care systems. Journal of Health Services Research and Policy 5(2): 122-6. Mendelsohn, M. 2002. Canadians' thoughts on their health care system: Preserving the Canadian model through innovation. Saskatoon: Commission on the Future of Health Care in Canada. National Forum on health. 1997. Canada health action: Building on the legacy. Values Working Group synthesis report. Ottawa: Health Canada Communications. Nevitte, N. 1996. The decline of deference. Peterborough, Ont.: Broadview Press.

310 Julia Abelson and John Eyles O'Neill, M. 1992. Community participation in Quebec's health system: A strategy to curtail community empowerment? International Journal of Health Services, 22(2): 287-301. Osborne, D., and T. Gaebler. 1992. Reinventing government: how the entrepreneurial spirit is transforming the public sector. Reading, Mass.: Addison-Wesley. Pateman, C. 1970. Participation and democratic theory. Cambridge, U.K.: Cambridge University Press. Petts, J. 2001. Evaluating the effectiveness of deliberative processes: waste management case studies. Journal of Environmental Planning and Management 44(2): 207-26. PriceWaterhouseCoopers. 1997. Canada Health Monitor, Survey #17, Highlights Report. Toronto. Putnam, R. 1993. Making democracy work: Civic traditions in modern Italy .Princeton, N.J.: Princeton University Press. Rasmussen, K. 2001. Regionalization and collaborative government: A new direction for health system governance, in Duane Adams, ed., Federalism, democracy and health policy in Canada. Montreal: McGill-Queen's University Press, 61-106. Regionalization Research Centre. 2000. A survey of regional health authorities' research priorities. Available at http://www.regionalization.org Resell, S. 1999. Culture, values and social institutions, in Steven Resell, ed., Renewing governance: Governing by learning in the information age. Don Mills, Ont.: Oxford University Press Canada, 33-59. Rowe, G., and L. Frewer. 2000. Public participation methods: A framework for evaluation. Science, Technology and Human Values 25(1): 3-29. Sandel, M. 1996. Democracy's discontent: America in search of a public philosophy. Cambridge, Mass.: Harvard University Press. Social Union Framework. 1999. A framework to improve the social union for Canadians. Available at http://www.uni.ca/socialunion.html (7 December 2001). Stacey, M. 1992. Regulating british medicine. Chichester, U.K.: John Wiley and Sons. Stein, J. G. 2001. The cult of efficiency. Toronto: House of Anansi Press. Sullivan, M., and Y. Scattolon. 1995. Health policy planning: A look at consumer involvement in Nova Scotia. Canadian Journal of Public Health 86(5): 317-20. Taylor, M. 1987. Health insurance and Canadian public policy: The seven decisions that created the Canadian health insurance system and their outcomes. Montreal: McGill-Queen's University Press. Tuohy, C., and R. Evans. 1984. Pushing on a string: The decentralization of

Public Participation and Citizen Governance 311 health planning in Ontario, in The costs of federalism. R. Golembiewski and A. Widawsky, edv New Brunswick, N.J.: Transaction Books, 89-116. Veenstra, G., and J. Lomas. 1999. Home is where the governing is: Social capital and regional health governance. Health and Place 5:1-12. Webler, T., S. Tuler, and R. Krueger. 2001. What is a good public participation process? Five perspectives from the public. Environmental Management 27(3): 435-50. Yankelovich, D. 1991. Coming to public judgment: Making democracy work in a complex world. Syracuse, N.Y.: Syracuse University Press. Zakus, J.D. and C. Lysack. 1998. Revisiting community participation. Health Policy and Planning 13(1): 1-12.

10 Practical Strategies for Facilitating Meaningful Citizen Involvement in Health Planning JAYNE RENEE PIVIK

The concept of including citizens in the process of defining their health needs and developing solutions to meet those needs has been supported by the World Health Organization (WHO 1978), the federal government (Epp 1986; Ministry of Health 1993), provincial governments (Government of Quebec 1988; Ontario Ministry of Health 1989; Premier's Commission on Future Health Care for Albertans 1989; Nova Scotia Royal Commission on Health Care 1989; Saskatchewan Commission on Directions in Health Care 1990), regional health organizations (e.g., Association of District Health Councils of Ontario 1991; Regional Municipality of Ottawa-Carleton 1995; Regional Municipality of the Halton Health Department 1993), consumer organizations (Consumers' Association of Canada 1993, Consumers' Health Forum of Australia 1990, 1996), and health researchers (Gott and Warren 1991; Lilley 1993; Meleis 1992; Mhatre and Derber 1992; Siler-Wells 1988), just to name a few. Support for increasing citizen participation (CP) in health planning was in large part influenced by the WHO and the United Nations International Children's Emergency Fund's 1978 Alma Ata Declaration (World Health Organization 1978). Central to this declaration is the conviction that individuals have the right and responsibility to participate in the planning and implementation of their health care. The result of the declaration, aimed at bringing health within the reach of all by the year 2000, was an increase not only in CP but also in decentralization of health services, an emphasis on health promotion, and a focus on de-institutionalization. In response to this initiative, the Canadian government released a document entitled Achieving Health for All: A Framework for Health Pro-

Facilitating Citizen Involvement in Planning 313

motion (Epp 1986), which encouraged CP in policy development for health care reform. Specifically, it stated: 'Encouraging public participation means helping people to assert control over the factors which affect their health. We must equip and enable people to act in ways that preserve or improve their health. By creating a climate in favour of public participation, we can channel the energy, skills and creativity of community members into the national effort to achieve health' (9). Consequently, many provinces developed a decentralized health care system that included regional boards of health, which are responsible for budgeting, service management, monitoring, and evaluation of health and social services. A review of six Canadian provincial health commissions and reports conducted by Mhatre and Derber (1992) identified increased CP as an important principle for improved health care, along with decentralization and, in some cases, devolution of authority from the provincial to the regional level (also see Godbout 1981; Zakus and Lysack 1998, for reviews). Recently, federal/provincial/territorial deputy ministers of health and four large organizations (the Canadian Coordinating Office of Health Technology Assessment, the Canadian Health Services Research Foundation, the Canadian Institute for Health Information, and the Institute of Health Services and Policy Research) conducted a national consulation process to identify priority areas related to health services and policy issues. Human resources were identified as the dominant issue for health research for the next two to five years, with particular emphasis placed on meeting the health needs of marginalized or underserved groups in a sustainable and cost-effective manner (Canadian Health Services Research Foundation 2001). Consumer and advocacy organizations have also rallied together to insist on an equitable and influential role in health planning. One example is the recommendation from the Research to Action Forum, that people with disabilities play a central role in the development and implementation of policy related to disability and have better opportunities for involvement in service planning (Research to Action Forum: Recommendations and Summary of Findings 1999). CP and Health Planning Health Planning Health planning essentially involves identifying problems or gaps within a specific area and developing an action plan for implementing

314 Jane Renee Pivik

solutions. A framework described by Rice (1993) for assessing health services typifies the problem-identification stage. Initially, the area is defined in terms of social systems, environmental factors, and demographics. Community members are then recruited from community organizations such as school boards, public health organizations, elected officials, etc. Following recruitment, a community health profile with key indicators and measures that assist with setting priorities is developed (e.g., census data, hospital records, police records). Finally, a list of problems to be solved as well as available resources is compiled. Once the goals and objectives have been identified, the community under investigation examined, an inventory of existing services taken, and demographic and epidemiological data gathered, the second stage of the health planning process is to develop a plan for implementing the solutions. Based on the results of the needs assessment, planners then determine priorities, develop recommendations and implementation strategies, define evaluation criteria, prepare a final report, and establish mechanisms for updating the plan (Association of District Health Councils of Ontario 1991). Citizen Participation in Health Planning CP in health planning refers to involvement in the identification, development, and evaluation of services, programs, policies, and laws aimed at keeping citizens healthy, that is, maintaining their emotional, physical, social, and spiritual well-being (Lilley 1993). Strategies for including citizen input in health planning can be broadly categorized as passive or active, depending on the participation level and the amount of decision-making power involved. Types of CP in Health Planning Consultation Strategies Traditionally, CP in health planning has involved a passive or consultative role, where community members are asked to provide input related to their needs and concerns, program planning, and resource allocation (Regional Municipality of the Halton Health Department 1993). Citizen consultation is aptly described by Saltman (1994) as a Voice' and conducted using such methods as surveys, community forums, focus groups, public meetings, and key informant interviews.

Facilitating Citizen Involvement in Planning 315 Involvement Strategies

Active participation or citizen involvement reflects a Voice and choice' for community members (Saltman 1994). In citizen involvement mechanisms, community members are involved in the development, implementation, and evaluation of health planning (Meleis 1992). Examples of citizen involvement strategies include membership on boards or committees in provincial health councils, regional health agencies, local boards of health, hospital advisory committees, hospital boards of directors, or specific task forces (Lilley 1993). Another form of citizen involvement is the rapid participatory-appraisal process developed by the WHO (WHO 1988), where key community informants work collaboratively with health professions to collect and analyse community information and determine priorities based on that effort (Rifkin 1996). It is clear that the main criteria differentiating citizen involvement from citizen consultation is the citizens' level of participation in decision making. Advantages of CP Many advantages to CP in health planning have been reported (Bracht and Tsouros 1990; Feather, McGowan, and Moore 1994; Lilley 1993; Rifkin 1996; White 1982). These include: • health care that reflects the specific needs, values, culture, and attitudes of the community; • decision making that is more accountable because of the incorporation of specific concerns identified in the community; • more efficient use of scarce resources; • enhancement of community awareness of health issues; • increased support for the resulting programs and services; • opportunity for increased networking between health care service providers and community members; • access to local resources and skills of community members; • mechanism for public feedback; • opportunity for training and educating community members for future community-development activities; and • enhanced sense of control and empowerment within the community. Including community members in needs assessment and problem

316 Jane Renee Pivik

solving provides the opportunity for different perspectives, the pooling of resources, and creative approaches. For example, one of the projects of the 'Healthy Localities' strategy utilized a brainstorming session with community members on the topic of nutrition. Creative community-based solutions involved convincing local grocery stores to set up a fruit and vegetable stand for people in a high-rise development, the introduction of healthier food in school canteens, and a change in the menus based on feedback from individuals receiving Meals on Wheels (Turner 1990). Challenges of CP With such reported advantages, one would expect that CP in health planning would be the norm and practised consistently, which unfortunately is not the case. Reported barriers to CP include resource limitations (Checkoway, O'Rourke and Bull 1984; Weaver and Pivik 1997a), perceived status differentials, processes that are not fully accessible, poor communication, differing definitions of participation, conflicting vested interests, an incongruence between stated purpose and practice, lack of representativeness, tokenism, role strain, and the stress associated with change in role definition (Lilley 1993; Lord 1989; Valentine and Capponi 1989). As well, Weaver and Pivik (1997a) found that health planners identified time constraints, difficulty reaching marginalized populations, and lack of knowledge and training as significant challenges to CP. Canadian Experiences Involving Citizens in Health Planning Case Illustrations In order to identify what methods have been successful in the past and where changes can be made for the future, a few illustrations of CP in health planning in Canada are presented. Aronson (1993) describes a public-consultation process conducted by the Ontario Ministry of Community and Social Services where elderly people were asked for their feedback on a reform document for long-term care policies. This exercise was a clear case of citizen consultation, not involvement. Further, citizens' contributions were solicited in the later stages of the policy-making process, calling into question its purpose. Finally, the consultation process was flawed. When citizens gave their input, min-

Facilitating Citizen Involvement in Planning 317

istry facilitators attentively listened to the concerns, anecdotes, and questions and then asked them to turn their comments into suggestions for reform (which was difficult for many of the participants). Lessons that can be gleaned from this experience include ensuring that CP occurs at the formative stages of planning and that processes are accessible, and recognizing that citizen input reflects experiential knowledge and may be delivered as stories or personal experiences. For this last point, it is the job of the planner to take those stories and interpret them, not the role of the citizen to make their experiences fit the health plan or objectives. Fortunately, Labonte and Feather (1996) have developed a unique and user-friendly way in which citizens' storytelling can be analysed and used to identify health concerns and priorities. Factors that facilitated and hindered CP in health planning have been examined for regional health planning organizations in eastern Ontario (Weaver and Pivik 1997a). Six district health councils and six public health units were surveyed to study the following: the organizational culture of the agency in relation to CP; when CP is used; which mechanisms are used; important skills of citizens and staff for facilitating CP; resources; barriers to CP; and suggestions for eliminating those barriers. Facilitators identified for public involvement and consultation included clear organizational policies and guidelines, adequate staff and time to pursue community involvement, resources for training, information, education, and materials, and financial resources for involving community members. Regarding community-based health planning, an effort in British Columbia can serve as an example. Community-based health planning groups were organized with a mandate to develop a health plan that reflected the needs of its constituents based on the policy 'New Directions for a Healthy B.C.' (Higgins 1999). The effort involved citizen input in the early phases of planning and included a plan that involved public education. However, this experience speaks to the challenge of ensuring representation of citizen input or involvement and, particularly, developing processes accessible to marginalized populations. Of the four health planning groups that were developed, youth, single parents, persons with physical disabilities or mental illness, and members of First Nations were not adequately represented. In another community-based experience, Nelson, Lord, and Ochocka (2001) closely examined the factors and processes of citizen involvement in mental health planning and policy development. Three different organizations were studied over two and a half years: a Canadian Mental Health Association, an organization geared toward housing for

318 Jane Renee Pivik

survivors/consumers with mental illnesses, and a local self-help organization. This study clearly indicates that citizens are capable of successful involvement in health planning and service delivery given the opportunity and proper support. Factors such as value-based mandates, participatory processes and structures, willingness for power sharing, and flexible and individualized support services were a few of the key factors influencing successful involvement. Constraints inhibiting citizen involvement included a lack of governmental funding, inconsistent support from some stakeholder groups, and the need for skills training and education. Understanding successful strategies and existing challenges provides useful information for facilitating CP. Based on previous experiences, recommendations for facilitating meaningful CP have been developed for federal, provincial, regional, and community-based initiatives. Federal Government Recommendations Federal Government Sponsored Research Initiative Typically, the decision regarding when, where, and which type of citizen input is needed for a particular project is under the auspices of health planners and bureaucrats. In fact, only a few studies have attempted to determine the public's view of which type of participation would interest them. Notably, Abelson et al. (1995) questioned forty-six randomly selected citizens and forty-six attendees at a town hall meeting about their willingness and interest in participating in health and social services needs identification and decision making. In this sample, there was more interest in assuming a consulting role (73.9 and 50.0 per cent, respectively) than in being held responsible for decisions (8.7 and 15.9 per cent). However, since most of these respondents reported middle-class incomes and high education levels, with a third of the randomly selected citizens and over half of the town hall attendees employed in either health or social services, the sample was clearly not representative. Further, the authors did not ask respondents to indicate what informational needs, resources, or training they might need to participate. It is recommended that the federal government initiate a research program focused on access to equitable health. This can be achieved by funding either a research-chair position or a special competition through the federal granting agencies associated with health. It is recommended that this mandate include conducting a national survey of citizens-at-large, con-

Facilitating Citizen Involvement in Planning 319 319

sumer groups, and advocacy organizations, to assess interest, accessibility, and challenges to citizen participation in health planning. Questions should be asked about the level and type of participation that they might be interested in, the resources needed for participation, accessibility issues, current barriers and possible solutions to participation, and their current skills, knowledge, and expertise. This information will provide not only a guide for better methods for reaching citizens but also, perhaps, a database of citizens interested in participation. As well, information about the perspectives and needs of citizensat-large compared to vested interest groups can be acquired. Instead of assuming that citizens are typically apathetic towards participating in health planning (Abelson et al. 1995; Church and Barker 1998), this effort will determine where and how to facilitate their involvement. Concurrently, surveys sent to health planners would provide valuable information about current practices, resources, and support needed for enhancing CP and the challenges that make CP difficult to achieve. Policies based on the results of this research can then be developed in order to address accessibility, representation, and process issues associated with facilitating citizen participation in health planning. Provincial Government Recommendations Tangible political support is considered essential for facilitating CP in health care planning and decision making (Church and Reville 1989; Kushner 1996; Mattessich and Monsey 1992; Meleis 1992; Siler-Wells 1988; Sullivan and Scattolon 1995). The following recommendations for provincial governments centre on the development and implementation of policies associated with addressing the social influences on health, ensuring representation, and improving accessibility to participation in health planning by its citizens. Social Influences on Health It is recommended that provincial governments develop and implement policies that directly link health with social services. One of the greatest determinants of health inequalities identified by Raphael (2000) is poverty. He argues that communication among health, social development, policy organizations, and municipal sectors needs to occur more frequently in order to address the problem collaboratively. Babwin (1998) also stresses the need to be cognizant of the health impacts associated with housing, crime reduction, and safety issues.

320 Jane Renee Pivik

Labonte (1992) has shown that public health workers are aware of the health influences of unemployment and homelessness but regard poverty more in relation to lifestyle factors than in terms of health. Interestingly, Labonte also finds that health professionals view health inequalities more narrowly than either other individuals or community groups. This finding was confirmed in a study by Rowan (1998), where both consumers and health providers identified depression, isolation, and stress as negative influences on health. However, the consumers focused on the broad-based social impact of these factors on family and community whereas the providers were primarily concerned with how these issues affected the aged or persons with disabilities. These results indicate the influence one's role has on the type of information elicited, further stressing the need for representative feedback. When the relationship between social issues such as unemployment and homelessness (Church 1996) is linked with health consequences, it is likely that more individuals (particularly marginalized populations) will understand the importance of participating in health planning. Ensuring Equitable Representation It is recommended that provincial governments develop and implement policies that promote equitable representation in health planning. What is a truly representative group? According to Checkoway (1979), CP should consist of consumers who are broadly representative of social, economic, linguistic, and racial populations as well as residing in the proximity of the health service area. He further suggests that citizens who are representative of constituency organizations have more support and are more likely to represent a stronger and more diverse voice for consumers, an opinion supported by consumer organizations, planners, and researchers alike (Bastian 1996; Brotchie and Warm 1993; Consumer's Health Forum of Australia 1990, 1996; Gott and Warren 1991; Lilley 1993; Siler-Wells 1988; Sullivan and Scattolon 1995). However, whether citizens or consumers/users are sought for participation is dependent on the goal of the initiative. Citizens can provide a community-based perspective reflecting the specific needs and priorities identified by their communities as important. Leaders in the community have often shown themselves to be effective in this role. Users or consumers, on the other hand, can provide specific experiential knowledge of a health issue such as barriers to equitable health care services or the social impact of a disease. Both types of participa-

Facilitating Citizen Involvement in Planning 321

tion are important and should be sought as needed. In relation to health policy development, Kuschner (1996) recommends participation by 'consumerists' or individuals with a focus on consumer rights (e.g., access, choice, information, advocacy, and equity) and an understanding of the policy process. She suggests approaching national umbrella organizations that deal with consumer issues, such as the Consumer Association of Canada. One of the major concerns associated with representativeness is the lack of marginalized individuals providing their perspective. We need to identify the under-represented groups and develop mechanisms for gaining their input. Marginalized populations have been defined by Laverack and Labonte (2000) as 'those most in need, not already able to meet their own needs, with limited access to resources or who exist outside power structures' (258). For example, marginalized populations in relation to health planning have included individuals who are stereotyped owing to unemployment, disabilities, ethnicity, age, or gender. Disadvantaged groups tend to have higher health needs but are also the least likely to have a say or participate in identifying needs and priorities for health services (Robertson and Minkler 1994; Kieffer 1984; Whitehead 1992), and without active support and involvement they are unlikely to participate (Laverack and Labonte 2000). Higgins (1999) has examined the strategies for increasing participation in marginalized populations (First Nations bands, street youth, youth, single parents, persons with physical disabilities, and persons with mental illness) using focus-group discussions. She finds that traditional strategies such as committee meetings, public forums, and surveys are not useful mechanisms for these groups. The people in her study were interested in participating but wanted that participation to happen in their own settings, such as day-care facilities, community organizations, Aboriginal friendship centres, and schools, or via the Internet. They suggested that health planners spend time 'walking in their shoes' in order to gain a better understanding of their realities the first step towards experiential participation. Higgins recommends that, to reach under-represented groups, one must go where the people are, be it coffeehouses, churches, community centres, etc., actively listen and observe, and then engage individuals in discussing their needs and concerns. Finally, we must acknowledge and respect the diverse contributions and strengths of citizens as participants. She suggests that 'meaningful participation may constitute sharing experiences as recipients of services, providing input and feedback about the quality

322 Jane Renee Pivik

of health care, as well as identifying needs in the planning of programs. This may be in the form of citizen panels, discussion groups or innovative methods such as at religious gatherings and as display structures in malls' (p. 33). The importance of gaining this particular input is especially important since 'average citizens' rate the needs of disadvantaged populations very low (Lomas 1997). Improving Accessibility for Involvement It is recommended that provincial governments develop and implement policies that promote improved accessibility in health planning. Improving accessibility in health planning involves determining how and where participation is requested. Church (1996) describes the barriers that survivors of mental illnesses faced in trying to have their voices heard during legislative subcommittee hearings on mental health. The meetings were held in large cities and the survivors were expected to pay for their travel and accommodation (with reimbursement given later), respond to a committee agenda that focused on legislated issues such as centralization/decentralization, and deal with the difficulties associated with public speaking. A more accessible process would have included up-front travel and accommodation expenses, an agenda that allowed these survivors to address issues on which they have clear expertise (e.g., what it is like to have a specific illness and society's response to that illness), and optional formats for expressing opinions, such as having letters read to the committee from survivors. Regarding where to gain citizen input, the need to go to the people was also indicated in a study by Rowan (1998), where less than half of the consumers from a variety of organizations (service clubs, social service groups, religious groups, schools, community groups, self-help groups, others) attended-focus group meetings at a local hospital to discuss health services needs, whereas over 90 per cent of the participants from the hospitals attended. Babwin (1998) also found that going into the community and meeting with community leaders 'on their turf is important for participation. This is a strategy that has been proven successful over and over again (Deville-Almond 1998; Pivik and Weaver 1997; Rodney et al. 1998; Weaver and Pivik 1997b). Other policies that would facilitate meaningful participation reflect the processes for gaining input. It is recommended that the following four principles, based on the legal requirements of consultation identified by Edwards (1998) be observed. These are: consultation should take place in the formative stages of the initiative; the consultee must have sufficient

Facilitating Citizen Involvement in Planning 323

and adequate information to make informed decisions; adequate time must be given to make decisions; and community decisions must be given conscientious consideration. Citizens need to know that their participation is valued. Policies that require that participation be sought in the formative stages of the initiative will strengthen the belief that the participants' opinion was given conscientious consideration, as noted by Aronson (1993). It is also important to give citizens an opportunity to reflect on their opinions, as shown by Shiell, Hawe, and Seymore (1997) in their testretest study of health outcomes, where a substantial number of individuals changed their opinion following consideration of the issues. Jordan et al. (1998) recommend that, if health authorities are interested in substantive recommendations from the public, opportunities for informed and deliberative participation such as citizens' juries or userconsultation panels be used. Policies that promote information provision to citizens are another important antecedent for meaningful participation. It is recommended that provincial governments provide resources to support the development of coalitions, to provide opportunities for public debate on health issues (Sabatier 1987), and to establish citizen-research centres to assist community involvement in rule making and policy decision (MacDermid et al. 1993). Partnerships with academics could assist in this initiative for educating the public, as evidenced by the public workshops on health organized by St Francis Xavier's University extension department (1999). Entitled 'People's School on Health,' these workshops covered such topics as globalization, inequalities in health, health public policy (what is it and how can we influence it?), health impact assessment, and solutions. In summary, provincial-government recommendations that would support meaningful participation include the development and implementation of policies that promote social and economic alliances with health, ensure representation of citizen input, facilitate accessibility of CP, and provide informational and education opportunities which enable informed decisions. Regional-Level Recommendations The Multi-Modal Consumer Involvement Model The regional-level recommendations focus on the development of a comprehensive working model that incorporates the provincial-policy recommendations. Entitled the Multi-Modal Consumer Involvement

324 Jane Renee Pivik Figure 10.1 The multi-modal consumer involvement model

Model (see Figure 10.1), it addresses the challenges associated with ensuring accurate feedback from the community, the difficulty in reaching marginalized populations, the social influences on health, and the need for representativeness of community feedback. The model is based on the concept of data triangulation (Jick 1979), where multiple methods of data acquisition are used simultaneously to promote accuracy. Collaborative Health Networks It is recommended that regional health authorities formally develop collaborative health networks. Creating formalized collaborative health networks addresses the social impacts on health and provides a mechanism for pooling skills, talents, and resources as well as a vehicle for information collection and dissemination (Proenca 1998). Collaborative health networks would involve organizations whose mandates intersect with

Facilitating Citizen Involvement in Planning 325

health issues (e.g., poverty, housing, crime, social support, education, etc.) such as public health units, social services, law enforcement, schools, community organizations such as the Independent Living Resource Centres, shelters, religious organizations, businesses, and so on. Each network would have a specific focus, such as the priorities identified by the regional health authority. According to Proenca (1998), effective collaborative health networks are characterized by high levels of connectivity and integration, have organizational cultures that emphasize teamwork, mutual respect, reciprocity, and community accountability, and are coordinated by a community health manager or leader. Gamm (1998), in his study of effective health networks, stresses the importance of carefully designing a program and the need for information sharing across the organizations involved. These networks then become both a means of identifying community needs and priorities and a vehicle for addressing the problems identified. The development of the network must be a collaborative effort with all those involved, where participants mutually identify prevailing values, goals, and principles. A collaborative-partnership approach will identify roles, responsibilities, expectations, and participants' strengths and constraints and define decision-making and conflict-resolution strategies (Pivik 1997). Community Consultants It is recommended that regional health authorities dedicate resources for hiring community consultants. The most salient feature of the model related to meaningful CP is the introduction of community consultants whose sole aim is to assist in identifying the health needs and priorities of the community. Once provincial priorities have been identified, consultants with experiential knowledge of the targeted priority can be identified. For example, the identified priorities of the Ontario Community Health Centres health promotion programs related to child and family health (Ontario Ministry of Health and Long-Term Care 2002) include domestic violence prevention/treatment; parenting education, both individual and group, to improve healthy child development; anti-racist initiatives and other programs to promote tolerance, cultural diversity, and acceptance of minority groups; and education and counselling related to weight/body image issues, peer relationships, healthy sexuality programs for street youth including drop-ins, and programs for teen mothers. Community consultants for this program may include survivors of domestic violence situated at a halfway house, individuals

326 Jane Renee Pivik

of different races or religions within their communities, people who have dealt with eating disorders, or teen mothers who are situated at youth drop-in centres. With their experiential knowledge, these consultants understand first-hand the realities of social and health services associated with these issues. Further, people in similar situations are more likely to open up to them and they probably already have contacts in the community. Finally, they represent marginalized populations and are likely to provide unique perspectives and experiences. In order to facilitate access to community members, community consultants need to be situated within the community. According to Rodney et al. (1998), the benefits of using community members for outreach work includes 'improved accessibility and sustainability of health services, increased sense of power in the community to effect change and increased communication and collaboration between community members and health care providers in problem identification and resolution' (372). Paying the consultants for their work is paramount to the success of this model. We do not expect our outreach or health workers, health planners, politicians, or researchers to volunteer their time for identifying community needs and priorities. The community consultants should be treated equally and compensated for their time and effort (Meleis 1992; Pivik 1997; Pivik and Weaver 1997). This suggestion is echoed in Abelson et al.'s (1995) study of community participation in health planning, where citizens indicated that payment for their time would motivate their involvement. Hiring is expected to increase continuity of citizen involvement. One of the greatest obstacles to promoting citizen involvement in health planning is with the idea that, in order to provide meaningful input, you must volunteer your time. This restricts the involvement of those individuals who must work to support themselves or their families. Understandably, concerns for basic needs will supersede volunteering efforts. Only those individuals who are not concerned with ensuring continual sustenance tend to participate in citizen-involvement positions long term, such as retired individuals (Weaver and Pivik 1997a), and, consequently, the representativeness of the individuals participating in citizen-involvement mechanisms is skewed towards a particular group and hence a particular perspective. Research has shown that one's role can influence what information is attended to and given priority (Canter 1977). Further, Howell et al. (1998) find that job creation appears to be the most effective way to involve citizens. In fact, many of the individuals hired

Facilitating Citizen Involvement in Planning 327

through their community-empowerment initiative had previously been welfare recipients or in community health and social service programs. The success of this initiative provides another motivation for the use of community consultants, and, by hiring the community consultants, we address the challenge of representativeness and access to underserved populations. Community consultants would work collaboratively with community consultation coordinators, who would be responsible for coordinating their activities. This may involve identifying current concerns and methods for acquiring information, providing information from the community health services to the consultants and the public, training consultants, organizing involvement opportunities such as community forums, networking with outreach workers and health planners, and collecting and forwarding the data from the community consultants to the Community Health Services Department. Experience from the Healthy Start Program (Howell et al. 1998), a large American health initiative, has indicated that the task of mobilizing the community and coordinating citizen input is labour-intensive and requires adequate staffing resources to succeed. Providing Educational and Training Support It is recommended that regional health authorities provide tangible support for educating and training all individuals involved in health planning. Increased education and training is another necessary component for all individuals involved in health planning. For example, further education and training has been suggested for communication skills (written, verbal, and listening), process issues associated with health planning and evaluation, interview techniques, procedures for meetings, partnership building (Brotchie and Warm 1993; Lilley 1993), group-facilitation techniques, conflict resolution, community building, process issues (Lilley 1993), community networking, research skills related to CP, stress management and assertiveness training (Brotchie and Warm 1993). How-to guides that focus on contacting the community, needs-assessment mechanisms, priority setting, and evaluation can supplement the training and be a useful resource (Dwyer 1997). Continued effort by outreach workers and health planners in assessing needs and priorities will provide a multilayered assessment opportunity (data triangulation). After an evaluation of the proposed model, outreach workers, health planners, and community consultants may

328 Jane Renee Pivik

find that duplication of activities is unnecessary or, conversely, that it provides a broader base of citizen input. In summary, recommendations for regional health authorities focus on the development of collaborative health networks that involve formal partnerships with social services. These networks would include community consultants who are dedicated to the role of obtaining citizen input in health planning. Choosing community consultants based on their experiential knowledge will provide more meaningful information since it is expected that they will have better access to marginalized populations, thereby improving representativeness in health planning. Hiring and training those individuals is also expected to improve continuity of citizen involvement. Combining the information of community consultants working on the same topic from different communities (e.g., youth suicide) may provide a greater breadth of knowledge and so facilitate the accumulation of more representative data. This model may be considered radical by those who believe that only 'well-educated professionals' should be involved in obtaining public input, or that people who have had difficulties in the past may not be up to such an important role. One of the main challenges associated with CP in health planning is the issue of sharing power. Power differentials occur when one group (historically, health professionals) believe that their expertise is the only element necessary to make informed decisions. Fortunately, a paradigm shift is now occurring, as health professionals come to the realization that the experiential knowledge of the consumer, user, or citizen is an invaluable and necessary component for effective and valid decision making. When more and more organizations and agencies incorporate collaborative and participatory approaches in their policies and procedures (where all participants are valued for their specific knowledge and expertise), power struggles will start to dissipate. Along with power struggles, our greatest challenges to gaining meaningful input in health planning are ensuring representation, reaching marginalized populations, and maintaining involvement. Evaluation of the Multi-Modal Consumer Involvement Model by regional health authorities will determine whether it addresses these challenges. Community and Institutional Recommendations Community and institutional recommendations focus on the mechanisms involved in achieving CP in health planning at the local level. Although citizen involvement in health planning typically takes place

Facilitating Citizen Involvement in Planning 329 within communities and institutions (whereas consultation strategies are mainly used federally and provincially), the principles and strategies involved in the following recommendations can apply at all levels. These recommendations are based on a Conceptual Framework for Community Involvement in Health Planning that was derived from an extensive literature review which synthesized the wisdom and opinion of health professionals, governments, and consumers who have had experience and expertise related to CP/ involvement in health planning (Pivik and Weaver 1997). Information describing the techniques, strategies, and recommendations for facilitating CP came from journal articles, occasional papers, and reports from consumer organizations, health care institutions, and governments. It is recommended that community organizations and institutions consider the following four elements, when addressing CP in health planning: a CP-supportive climate; process issues; knowledge issues; and financial and organizational support. For ease of use, the recommendations have been consolidated into a checklist format (see Appendix A). Nurturing a Supportive Climate The first stage in facilitating CP is nurturing a supportive climate. This involves mobilizing the community, fostering respect and trust, achieving an attitude shift among health professionals, and utilizing a partnership approach. Concrete strategies identified for mobilizing the community include utilizing communication tools and techniques such as press releases and conferences, public meetings, and written materials (Rice 1993), drawing upon existing networks (Gott and Warren 1991; Weaver and Pivik 1997b; Siler-Wells 1988), providing public education about community needs, community dynamics, status and the availability of health care participation (Meleis 1992; Siler-Wells 1988), ensuring that information is clearly understood (Lord and Farlow 1990; O'Neill 1992), and providing motivation and rewards for participation (Meleis 1992). Fostering Respect and Trust Fostering respect and trust between health care professionals and community members can be done in a variety of tangible ways, which include really listening to each other, building on an individual's capacities instead of deficiencies (Lord and Farlow 1990), ensuring that participation is not tokenism and that meetings are inclusive of all partners, providing equal information to all participants, acknowledg-

330 Jane Renee Pivik

ing the different perspectives, values, skills, and experiences of all partners (Brotchie and Warm 1993; Consumers' Health Forum of Australia 1990; and Lord 1989), and recognizing that the building of trust takes time (Lord 1989; Singer 1995). Minimizing Power Differentials

Also recommended for nurturing a climate for CP is the minimization of power differentials between community members and health professionals (Lord 1989; Lord and Farlow 1990; Meleis 1992; Sen 1994; Siler-Wells 1988; Weaver and Pivik 1997b). These authors have suggested that health professionals take off the hat of expert, view themselves as partners and learners, and accept that community members have a right to identify their own needs and solutions by promoting methods that are more inclusive, empowering and participative. Partnership Approach

Related to an attitude and role shift is the use of a partnership approach when participating in health planning. Factors identified by Audet and Rostami (1993) for promoting effective partnerships include a positive attitude by all participants, shared values, good communication skills, the capacity to evolve, shared commitment, trust, and a balanced and qualified team. Concrete strategies for encouraging partnerships include effective communication of different realities, values, and assumptions (Lord 1989; Meleis 1992), agreement between partners on the purpose and vision of the task (Gott and Warren 1991; Meleis 1992), and the holding of meetings in community locations (Siler-Wells 1988). Citizen-Involvement Processes Once a climate conducive to CP is established, the mechanisms for conducting citizen involvement can be initiated. Citizen involvement may include working as community consultants or by serving as members of boards or committees in provincial health councils, regional health agencies, local boards of health, hospital advisory committees, hospital boards of directors, or specific task forces (Lilley 1993). Proponents of CP in health care planning stress the importance of focusing on process issues in a partnership context (Bastian 1996; Bichmann et al. 1989; Brotchie and Warm 1993; Consumers' Health Forum of Australia 1990; Lilley 1993; Lord 1989; Meleis 1992; Siler-Wells 1988). Examples of pro-

Facilitating Citizen Involvement in Planning 331

cess issues important to partnerships include determining whom to involve, the reason for their involvement, their role and responsibilities, and how decisions will be made. Criteria for Defining Partners

Identifying the composition of the group to work on a health issue is one of the first process issues to be addressed. Meleis (1992) identifies such life experiences as language, culture, location, and experience with health concerns as important considerations for citizen involvement. Further, she suggests increasing the awareness of the diversity of the community in question and ensuring equitable representation of underserved or vulnerable groups. One strategy identified for ensuring community representation is to recruit community members from organizations that are primarily consumer-based (i.e., individuals who come together voluntarily because of common experiences or conditions) (Bastian 1996; Brotchie and Wann 1993; Consumers' Health Forum of Australia 1990, 1996; Gott and Warren 1991; Lilley, 1993; Pivik 1997; Siler-Wells 1988; Sullivan and Scattolon 1995). These authors identify various advantages in utilizing existing community networks or organizations, including: • • • • • •

support in terms of knowledge and skills from other members; information sharing; accountability; strong mandate from the community; greater representation; and sense of community for consumers.

Shared Vision

Once a group of individuals is organized to work on the problem at hand, the next stage in health care planning is to ensure that all partners share in the vision and goals of the project. Specifically, Meleis (1992) suggests that participants develop common or agreed-upon definitions of purpose and strategies for implementing that purpose. Similarly, participants in a series of health care forums, whose purpose was to explore the public's attitudes and values in the area of health care and resource allocation, noted that the criteria for successful participation were a shared sense of purpose and commitment and active participation (Gott and Warren 1991).

332 Jane Renee Pivik Clear Roles and Expectations

The Consumers' Health Forum of Australia (1996) recommends that participants have a clear understanding of the roles and reasons for consumer participation. One method for facilitating this understanding is the development of a partnership agreement, which incorporates the parameters and levels of involvement of partners in relation to the implementation and continuation of the action plan (McComas and Carswell 1994; Meleis 1992; Pivik 1997). Along with a partnership agreement, several authors have noted the importance of strong leaders for facilitating community participation (Brotchie and Warm 1993; Lord 1989; O'Neill 1992). As either facilitators or chairpersons, these individuals can assist with identifying different values and assumptions and clarify points of disagreement. Decision-making Process

Having an open and transparent decision-making process will facilitate effective citizen involvement and promote a partnership approach to health care planning (Consumers' Health Forum of Australia 1996). Singer (1995) has identified a series of questions that can assist in determining the decision-making process. These include: • Who in the community should be the representative decision-making body? • What information is necessary for effective decision making? • Is a consultation role sufficient or does community participation require decision-making power in relation to resource allocation? • What types of decisions should community groups be responsible for? • To what extent is a decision affected group membership? Assessing Community Participation

The final process consideration is measuring the impact of this partnership approach. The Consumers' Health Forum of Australia (1996) developed a series of questions in order to access the impact of consumer involvement. Answers to the following questions can assist in measuring the process as well as the outcome and satisfaction of consumer participation: • Where is community participation taking place? • What strategies have been used to encourage consumer involvement?

Facilitating Citizen Involvement in Planning 333

• Have individuals with specific disadvantages been involved? • Have there been identified changes in the behaviour/attitudes of health professionals? • What are the benefits of community participation identified by the health professionals? • Have the attitudes of the consumers changed from the beginning of involvement to the one-year point? • What are the benefits the consumers have identified? • What barriers are affecting community participation and how can they be resolved? • What levels of resources were given for encouraging and supporting community participation? (33). At a systemic level, perhaps the most widely known approach to measuring citizen participation was developed by Bichmann, Rifkin, and Shrestha (1989). Based on a review of 200 case studies, these authors identified the following five factors as important to community participation: management, needs assessment, leadership, organization, and resource mobilization. Using a series of questions (e.g., How were needs identified? Was the community involved in needs identification and assessment?), they gave a score for each of the components of community participation. This approach allows a quantifiable measurement of the process of community participation at different times or between different programs, planners, or participants. Knowledge Knowledge is a powerful and necessary component for both community members and health professionals in understanding the value and process of community participation (Bastian 1995). Methods for increasing knowledge include information provision and exchange, education, and training. Information is integral to understanding needs, potential solutions, and the overall process for affecting change in health care planning. Along with the necessary information to do the job, information exchange is important for all participants of this process, that is, between health care professionals and the community, between participants in the health care planning process, between community members and their community organizations, and within health care agencies.

334 Jane Renee Pivik Information Provision

Lack of access to adequate information has been identified as a barrier to effective citizen participation (Brotchie and Wann 1993; Consumers' Health Forum of Australia 1990,1996; Lilley 1993; Sullivan and Scattolon 1995). Health planners need to ensure that their community partners have full access to information and also to individuals capable of offering valuable advice. Various techniques for information provision recommended by Human Resources Development Canada (1994), Siler-Wells (1988), and Brotchie and Wann (1993) include developing or providing • • • • •

guides describing agency or governmental policy; background information on the consumers' roles; background information on the decision-making process; consultation material related to the subject being addressed; and information on effective programs, policies, and models that describe effective leadership.

Information Exchange

Information was the key element employed in the people-centred approach to improving health services in Wales. Wilson (1992) describes how information was used to: 1) identify people's views of their own health needs and those of their families and their community; 2) show the health care agencies' response to the identified needs; 3) provide a description of how the data were analysed; 4) describe current health care services; and 5) inform on the short- and long-term options of current health care plans. This information, given in a series of seminars, resulted in the development of local planning groups, allowed future networking among participants, demonstrated that the community's opinion was valued, and encouraged a collaborative approach between health care providers and their communities. However, how the information is presented is important. Lubalin and Harris-Kojetin (1999) recommend that information be clear, easily understood, and include examples. It is also important that community representatives have effective mechanisms in place for communicating with their organizations about their progress and for receiving advice and support (Brotchie and Wann 1993; Consumers' Health Forum of Australia 1996; Lilley 1993; O'Neill 1992; Sullivan and Scattolon 1995). Sullivan and Scattolon (1995) recommend the use of newsletters or electronic-mail systems for

Facilitating Citizen Involvement in Planning 335

promoting information sharing between consumers. Another suggestion is a 'skills exchange' program, where consumers could access names of community members with specific skills, such as in planning, development, research, and evaluation, for advice or assistance (SilerWells 1988). Education and Training

Increasing education and skills training for both consumers and health care professionals has been suggested as a means of facilitating community participation in health planning (Brotchie and Warm 1993; Church and Reville 1989; Lilley 1993; Sen 1994). Checkoway, O'Rourke and Bull (1984) identify training as one of the most important factors differentiating agencies who have effective CP from those who do not. For Community Members: For community members, Lilley (1993) suggests educating the public about health issues, current health policies and programs, procedures, health planning, and evaluation. Similarly, training needs for consumers identified by Brotchie and Warm (1993) focus on communication skills and updated information sessions on policy and process issues. In relation to communication skills, it was determined that an education module related to the ability to listen, understand, and communicate other people's views would be most useful. Checkoway, O'Rourke, and Bull (1984) recommend that consumers on boards have skill training and that initiatives be conducted in the community that promote leadership, organization, and public awareness of health planning. Research has shown that consumer involvement is enhanced with training, resulting in individuals who attend more meetings, take on greater leadership and communication duties, and perceive themselves as more influential (Checkoway 1979). Also recommended for successful CP is ongoing support and educational opportunities (Zakus and Lysack 1998). Methods suggested for disseminating knowledge to consumers include the use of workshops and the development of easy-to-read manuals (Sullivan and Scattolon 1995). For Health Care Professionals: Lilley (1993) recommends that health professionals be provided the opportunity for education and skills development in group-facilitation techniques, partnership building, conflict resolution, and community building, as well as training on the value and process of community participation. Brotchie and Warm (1993), in

336 Jane Renee Pivik

their study of training needs, also recommended training in community networking as well as enhancing communication skills (verbal, written, and listening), research skills related to community participation, and personal-development training (stress management and assertiveness). In order to balance power differentials between consumers and health professionals, it has been recommended that both groups be provided with information on ways that power hierarchies can undermine community involvement (Sullivan and Scattolon 1995; Valentine and Capponi 1989). Support The final and perhaps most necessary criterion for ensuring citizen participation in health planning is the availability and provision of concrete support. Areas where support could influence citizen involvement include the provision of funds for community involvement and organizational support from health organizations. Financial Support

As indicated earlier, it takes time, resources, and effort for community members to participate in health planning. If not part of their employment, participation has often meant voluntary work on off- hours and costs associated with travelling, parking, and child care. In consideration of these issues and with the belief that community members' participation is an important component of health planning that needs to be respected and valued, resources for their participation must be provided. This includes payment for participation (e.g., consultation) as well as reimbursement for out-of-pocket expenses such as child care, travel, parking, and time off work (Brotchie and Warm 1993; The Consumers' Health Forum of Australia 1996; Lilley 1993). It has also been recommended that community members be provided with resources to obtain, analyse, and present information to support their views (Lilley 1993). This may involve funds necessary to access pertinent information or to cover the costs of educational or training courses relevant to community participation. Organizational Support

Support from the organizations that are responsible for health care planning and delivery is integral to facilitating effective CP. Organizational support covers many areas, such as the development of a CP mandate, the creation of a climate where CP is encouraged, and the

Facilitating Citizen Involvement in Planning 337 provision of adequate human resources as well as the resources needed to include community partners in the decision-making process. Specifically, Weaver and Pivik (1997a) find that organizational support for involving community members, recommended by health care staff, includes the following factors: Policy and procedure • clear organizational policies; • clear procedural guidelines; • a framework for involvement (e.g., recruitment methods), and • a framework for evaluation Human resources • adequate staff; and • sufficient time-frames Financial resources and support • in-service training; • funds for training community members; • public education funds (to describe the mandate of the agency, the issues and policies, the constraints and options, the research methods and the impact of decisions); • resources for involving community members (e.g., transportation/ travel expenses, child-care expenses, available meeting rooms, meals and refreshments, and monetary reimbursement for participation); • materials and information; and • valid and reliable measurement tools Conclusion The most important consideration when conducting CP in health planning is to follow guiding principles and values that are inclusive, participatory, and equitable. Citizens should be respected for their experiential knowledge and expertise. Valuable information can be obtained about the best way to obtain citizen input into health planning by simply asking them. Recommendations for federal involvement in facilitating CP in health planning involves support for the development of research initiatives focused on access to equitable health for Canadians. National surveys targeting citizens and health planners are recommended to identify to how best facilitate CP. Provincial-level recom-

338 Jane Renee Pivik

mendations focus on developing policies that address the social impacts of health and the issue of representation and accessibility. Regional-level recommendations include developing and implementing structures that address these policies and establishing informational and educational venues for promoting informed and considered citizen input. The Multi-Modal Consumer Involvement Model has been developed to facilitate this process and involves formalized community health networks and community-based individuals dedicated to collecting citizen input. Lastly, a Conceptual Framework for Community Involvement has been developed to provide strategies for facilitating CP within communities. Appendix A: Conceptual Framework for Community Involvement in Health Planning JAYNE R. PIVIK AND LYNDA WEAVER

Health Information Partnership, Eastern Ontario Region, 1997 Create a Climate for Citizen Participation Mobilize the Community • Develop a sense of 'community' among affected citizens; • Clarify community needs, community dynamics, and community status; • Inform citizens of opportunities to participate in health care planning; • Provide information about the community using existing local networks, advertising in public places, word of mouth, door-to-door information sheets, newsletters, and a community-based information office; • Ensure that information is accessible to all citizens (plain language, interpretation, creative visuals such as photographs describing a story, videos, posters, and dramas); • Involve other community agencies to reach families, neighbourhoods, and social groups to stimulate interest in community action around health issues; • Ensure that all participants are aware of the time commitment, energy, and effort required prior to their involvement; and • Motivate citizens to participate; and reward the efforts required to participate

Facilitating Citizen Involvement in Planning 339 Foster Respect and Trust • Really listen to people; • Focus and build on individuals' capacities rather than on their deficiencies; • Ensure that participation is not tokenism; • Conduct meetings that facilitate inclusion of all participants; • Recognize the unique perspectives, skills, and experiences of community members; • Treat community members as equals by providing equal information and heeding advice equally; • Acknowledge different perspectives and values; • Ensure language is understood by all; and • Recognize that building trust requires time Shift Attitudes • Encourage health professionals to consider themselves as partners, helpers, learners, enablers, coaches, and supporters rather than as experts or advisers; • Minimize power differences between professionals and community members; • Allow community members to determine their own needs and solutions; and • Allow consumers to play a strong role in health care planning through self-determination and empowerment Partnership Approach • Match the type of partnership arrangement with the level of participation required for the decision at hand; • Ensure equal power among all parties in decision making and collaborative partnership for all stages of health care planning (planning through to implementation); • For advisory partnerships, ensure all that all parties have the capacity to influence decisions; • Select individuals for consultative partnerships on the basis of their knowledge of the issue concerned, and give them opportunities to provide information; • Stress that all partnerships require these fundamental elements: • a positive attitude by all participants; • shared values; • good communication skills;

340 Jane Renee Pivik

• capacity to evolve; • commitment; • trust; and • balanced and qualified partnership team; • Discuss and clarify differences between values (personal opinions) and assumptions (underlying facts or circumstances to be considered); • Agree on a common purpose, agreement or commitment; and • Enhance empowerment by holding meetings in community environments or locations where the majority of participants are most comfortable Process Issues Define Partners

• Recruit community members who have life experiences to share, such as language, culture, location and experience with health concerns; • Aim to maintain equitable representation of underserved or vulnerable groups; and • Recruit from existing consumer-based groups (e.g., self-help, interest groups, advocacy groups), networks (e.g., volunteer agencies, seniors-helping-seniors groups), community organizations (e.g., community associations, business associations), and consumers' rights groups (e.g., local branch of the Consumers' Association of Canada)

Develop a Common Vision

• Work with all partners to develop a common or agreed upon definition of the purpose of the team and a plan to meet that purpose; and • Foster a sense of shared commitment and active participation from all partners

Clarify Roles and Expectations

• Collaboratively decide on each partner's role, involvement, expectations and, tasks; • Develop a partnership agreement that formalizes the roles and expectations of partners in relation to stated goals and objectives; and

Facilitating Citizen Involvement in Planning 341

• Elect/select a strong leader who can assist the group in acknowledging partners' different values and assumptions and in resolving disagreements Define a Decision-Making Process • Have an open and transparent decision-making process; and • Consider these decision-making issues: • Who in the community should be the representative decisionmaking body? • What information is necessary for effective decision making? • Is a consultation role sufficient or does community participation require decision-making power in relation to resource allocation? • What types of decisions should community groups be responsible for? • To what extent is a decision is affected by group membership? (Singer 1995) Assess Community Participation • Remove or reduce barriers and constraints to community participation by considering these questions: • Where is community participation taking place? • What strategies have been used to encourage consumer involvement? • Have individuals with specific disadvantages been involved? • Have there been identified changes in the behaviour/attitudes of health professionals? • What are the benefits of community participation identified by the health professionals? • Have the attitudes of the consumers changed from the beginning of involvement to the one-year point? • What benefits have been identified by the consumers? • What barriers affect community participation and how can they be resolved? • What level of resources were given to encouraging and supporting community participation? (Consumers' Health Forum of Australia 1996) • Consider the following areas for measuring community participation: management, needs assessment, leadership, organization, and resource mobilization.

342 Jane Renee Pivik Knowledge

Information • Ensure that all partners have full access to information necessary for them to fulfil their roles; • Ensure that all partners have full access to advisers, representative bodies, etc.; • Develop or provide the following items for all partners: • guides describing agency or governmental policy; • background information on roles and responsibilities; • background information on the decision-making process; • consultation material related to the subject being addressed; and • information on effective programs, policies, and leadership models • Share information between all partners in a format that is accessible to all; and • Develop a registry of community members who have skills to share, such as those interested in planning, development, research, or evaluation. Education and Training - For Community Members • Educate the public on health issues, health policies, and health programs; • Educate the public on procedures for meetings, on planning, and on evaluation processes; • Provide education sessions on communication skills, such as the ability to listen, understand, and relate to other people's views; • Discuss ways that power hierarchies can undermine community involvement; and • Use workshops or easy-to-read manuals for education purposes Education and Training - For Health Care Professionals • Educate professionals in the following areas: • group facilitation; • partnership building; • conflict resolution; • community development and networking; • community participation; • procedures for conducting meetings • communication skills (verbal, written, and listening);

Facilitating Citizen Involvement in Planning 343

• research skills related to community participation activities; and • personal development (stress management, assertiveness training) • Discuss ways that power hierarchies can undermine community involvement Support Financial Support • Budget for payment for community members' involvement; • Reimburse community members for out-of-pocket expenses such as child care, travel, parking, time off work; • Ensure that pertinent information or resources are purchased; and • Cover costs of pertinent training or educational sessions necessary for meaningful involvement Organizational Support • Create an organizational climate that favours community involvement in health care planning; • Have clear and widely-known mandates, policies, procedures, and guidelines for community participation; • Have a framework for evaluating community participation; • Ensure adequate staffing and time-frames; • Ensure adequate training for staff and community members; • Supply adequate materials and information; • Provide good evaluation measurement tools; and • Provide adequate financial supports Political Support • Lobby government to support community participation through policies and guiding principles, and support the development of regional coalitions; and • Request that government establish mechanisms to involve community members in health care decision making at the regional level.

REFERENCES

Abelson }., et al. 1995. Does the community want devolved authority? Results of deliberative polling in Ontario. CMAJ 153(4): 403-12.

344 Jane Renee Pivik Aronson, J. 1993. Giving consumers a say in policy development: Influencing policy or just being heard? Canadian Public Policy 19(4): 367-78. Association of District Health Councils of Ontario. 1991. Planning Ontario's health care services in the 1900s: Guide to DHC planning. Toronto: Ontario District Health Council. Audet, B.A., and J. Rostami. 1993. Partnership strategies for community investment: Findings of national consultants. Ottawa: Conference Board of Canada. Babwin, D. 1998. The patient as partner. Hospitals & Health Networks, 5 Nov., 24. Bastian, H. 1995. The power of sharing knowledge: Consumer participation in the Cochrane collaboration. Oxford: UK Cochrane Centre. - 1996. Raising the standard: Practice guidelines and consumer participation. International Quality in Health Care 8(5): 485-90. Bichmann, W., S. Rifkin, and M. Shrestha. 1989. Toward the measurement of community participation. World Health Forum 10: 467-72. Bracht, N., and A. Tsouros. 1990. Principles and strategies of effective community participation. Health Promotion International 5(3): 199-203. Brotchie, J., and M. Warm. 1993. Training for lay participation in health: Token voices or champions of the people? London: The Patients Association. Canadian Health Services Research Foundation. 2001. Listening for direction: A national consultation on health services and policy issues. Ottawa. Canter, D. 1977. The psychology of place. London: The Architectural Press. Charles, C, and S. DeMaio. 1993. Lay participation in health care decisionmaking: A conceptual framework. Journal of Health Politics, Policy and Law 18(4): 881-904. Checkoway, B. 1979. Citizens on local health planning boards: What are the obstacles? Journal of Community Development Society 10(2): 100-16. Checkoway, B.T., W. O'Rourke, and D. Bull. 1984. Correlates of consumer participation in health planning agencies: Findings and implications from a national survey. Policy Studies Review 3(2): 296-310. Church, J., and P. Barker. 1998. Regionalization of health services in Canada: A critical perspective. International Journal of Health Services 28(3): 467-86. Church, K. 1996. Beyond 'bad manners': The power relations of 'consumer participation' in Ontario's community mental health system. Canadian Journal of Community Mental Health 15(12): 27-44. Church, K., and D. Reville. 1989. User involvement in the mental health field in Canada. Canada's Mental Health 37(2): 22-5. Consumers' Association of Canada. 1993. The role of the consumer in health: A discussion paper. Ottawa: Consumers' Association of Canada. Consumers' Health Forum of Australia. 1990. Guidelines for consumer representatives: Suggestions for consumer or community representatives working on public committees. Canberra: Consumers' Health Forum of Australia.

Facilitating Citizen Involvement in Planning 345 - 1996. Working with consumers: A guide for divisions of general practice. Lyons, Australia: Consumers' Health Forum of Australia. Deville-Almond, J. 1998. Power points. Nursing Times 94(36): 32-4. Dwyer, S. 1997. Improving delivery of a health-promoting-environments program: Experiences from Queensland health. Australian and New Zealand Journal of Public Health 21(4): 398-402. Edwards, P. 1998. Question time. Health Services Journal, 23 July, 10. Epp, J. 1986. Achieving health for all: A framework for health promotion. Ottawa: Minister of Supply and Services Canada. Feather, J., V. McGowan, and M. Moore. 1994. Planning health needs assessment: The basic choices. Occasional paper no.l. Saskatoon: Prairie Region Health Promotion Research Centre, University of Saskatchewan. Gamm, L.D. 1998. Advancing community health through community health partnerships. Journal of Healthcare Management 43(1): 51-67. Godbout, J. 1981. Is consumer control possible in health care services? The Quebec case. International Journal of Health Services 11(1): 151-67. Gott, M., and G. Warren. 1991. Neighbourhood health forums: Local democracy at work. World Health Forum 12: 413-18. Government of Quebec. 1988. Rapport de la Commission d'enquete sur les services de sante et les services sociaux. Quebec: Les Publications du Quebec. Higgins, J.W. 1999. Closer to home: The case for experiential participation in health reform. Canadian Journal of Public Health 90(1): 30-4. Howell, E., et al. 1998. Back to the future: Community involvement in the healthy start program. Journal of Health Politics, Policy and Law 23(2): 291317. Human Resources Development Canada. 1994. Facilitating consumer involvement in policy planning and program development. Proceedings from the Focus Group Discussion. Halifax: Dalhousie University. Jick, T.D. 1979. Mixing qualitative and quantitative methods: Triangulation in action. Administrative Science Quarterly 24: 602-11. Jordan, J., et al. 1998. Health needs assessment: Whose priorities? Listening to users and the public. BMJ 316:1668-70. Kieffer, C. 1984. Citizen impowerment: A developmental perspective. Prevention in Human Services 3: 9-35. Kushner, C. 1996. Consumers and health policy: Confessions of a guarded optimist. International Journal of Quality in Health Care 8(5): 479-84. Labonte, R. 1992. Community health responses to health inequalities. North York, Ont: North York Community Health Promotion Research Unit. Labonte, R., and J. Feather. 1996. Handbook on using stories in health promotion practice. Ottawa: Health Canada. Laverack, G., and R. Labonte. 2000. A planning framework for community

346 Jane Renee Pivik empowerment goals within health promotion. Health Policy and Planning 15(3): 255-62. Lilley, S. 1993. Making it work! Community participation in health planning in Nova Scotia. Discussion paper for the Strengthening Health Partnership of Nova Scotia. Nova Scotia: Dalhousie University. Lomas, J. 1997. Reluctant rationers: Public input to health care priorities. Journal of Health Services Research and Policy 2(1): 103-11. Lord, J. 1989. The potential of consumer participation: Sources of understanding. Canada's Mental Health, 15-17 June. Lord, J., and D.M. Farlow. 1990. The study of personal empowerment. Health Promotion, fall, 2-8. Lubalin, J.S., and L.D. Harris-Kojetin. 1999. What do consumers want and need to know in making health care choices? Medical Care Research and Review 56(1): 67-102. McComas, J., and A. Carswell. 1994. A model for action in health promotion: A community experience. Canadian Journal of Rehabilitation 7(4): 257-65. MacCormack, C.P. 1983. Community participation in primary health care. Tropical Doctor 13(2): 51-4. MacDermid, R., et al. 1993. The limits and possibilities of democratic participation: A review of selected literatures. North York, Ont: York University. Mattessich, P.W., and B.R. Monsey. 1992. Collaboration: What makes it work, a review of research literature on factors influencing successful collaboration. St Paul, Minn.: Amherst H. Wilder Foundation. Meleis, A.I. 1992. Community participation and involvement: Theoretical and empirical issues. Health Services Management Services 5(1): 5-16. Mhatre, S.L., and R.B. Derber. 1992. From equal access to health care to equitable access to health: A review of Canadian provincial health commission reports. International Journal of Health Services 22(4): 645-68. Nelson, G., J. Lord, and J. Ochocka. 2001. Shifting the paradigm in community mental health. Toronto: University of Toronto Press. Nova Scotia Royal Commission on Health Care. 1989. Towards a new strategy. Halifax: Royal Commission on Health Care. O'Neill, M. 1992. Community participation in Quebec's health system: A strategy to curtail community empowerment? International Journal of Health Services 22(2): 287-303. Ontario Ministry of Health. 1989. Deciding the future of our health care: An overview of areas for public discussion. Toronto: Queen's Printer for Ontario. - 1993. A healthier Ontario: Progress in the '90s. Toronto: Ontario Ministry of Health. Ontario Ministry of Health and Long-Term Care. 2002. Community Health Centres (12 April 2002).

Facilitating Citizen Involvement in Planning 347 Pivik, J.R. 1997. Facilitating collaborative research: Strategies from the perspective of health researchers. Ottawa: University of Ottawa. Pivik, J.R., and L. Weaver. 1997. Facilitating community participation in health care planning: Recommendations from consumers and health professionals. Kingston, Ont: Health Information Partnership. Premier's Commission on Future Health Care for Albertans. 1989. The rainbow report: Our visions for health, vol.1. Edmonton's Queen Printer for Alberta. Proenca, E.J. 1998. Community orientation in health services organizations: The concept and its implementation. Health Care Management Review 23(2): 28-38. Raphael, D. 2000. Health inequities in Canada: Current discourses and implications for public health action. Critical Public Health 10(2): 193-216. Regional Municipality of Halton Health Department. 1993. Community consultation resource guide. Halton, Ont.: Halton Health Department. Regional Municipality of Ottawa-Carleton. 1995. Public consultation policy: Health department implementation guide. Ottawa: Ottawa-Carleton Health Department. Research to Action Forum. 1999. Recommendations and summary of findings. Halifax: Dalhousie University. Rice, J.A. 1993. Community health assessment: The first step in community health planning. Chicago: American Hospital Association. Rifkin, S.B. 1996. Paradigms lost: Toward a new understanding of community participation in health programmes. Ada Tropica 61: 79-92. Robertson, A., and M. Minkler. 1994. New health promotion movement: A critical examination. Health Education Quarterly 21(3):, 295-312. Rodney, M., et al. 1998. Three evaluation methods of a community health advocate program. Journal of Community Health 23(5): 371-81. Rowan, S. 1998. Provider and consumer perceptions of allied health service needs. Australian Health Review 21(1): 88-97. Sabatier, P. 1987. Knowledge, policy-oriented learning, and policy change: An advocacy coalition framework. Knowledge: Creation, Diffusion, Utilization 8: 649-92. St Francis Xavier University, Extension Department. 1999. People's school on health. Antigonish, N.S. Saltman, R.B. 1994. Patient choice and patient empowerment in Northern European health systems: A conceptual framework. International Journal of Health Sciences 24(2): 201-29. Saskatchewan Commission on Directions in Health Care. 1990. Future directions for health care in Saskatchewan. Regina: Saskatchewan Health, Public Affairs Branch. Sen, R. 1994. Building community involvement in health care. Social Policy, spring, 32-43.

348 Jane Renee Pivik Shiell, A., P. Hawe, and J. Seymore. 1997. Preferences and values are not necessarily the same. Health Economics 6: 515-18. Siler-Wells, G. 1988. Public participation in community health. Health Promotion, summer, 7-12. Singer, M. 1995. Community participation in health care decision making: Is it feasible? CMAJ 153(4): 421-4. Sullivan, M.J., and Y. Scattolon. 1995. Health policy planning: A look at consumer involvement in Nova Scotia. Canadian Journal of Public Health 86: 317-20. Turner, T. 1990. Healthy localities. Nursing Times 86(17): 59-61. Valentine, M.B., and P. Capponi. 1989. Mental health consumer participation on boards and committees: Barriers and strategies. Canada's Mental Health 37(2): 8-12. Weaver, L., and J.R. Pivik. 1997a. Community participation among members of the Health Information Partnership. Waterloo, Ont: Health Information Partnership, eastern Ontario. - 1997b. Out of the mouths of experts: A community participation reader. Waterloo, Ont.: Health Information Partnership, eastern Ontario. White, A.T. 1982. Why community participation? A discussion of the arguments. Assignment Children 59:17-34. Whitehead, M. 1992. The concepts and principles of equity and health. International Journal of Health Services 22(3): 429^5. Wilson, B. 1992. Establishing a people-centred health service in Gwent, Wales. Journal of Health Quarterly 14(3): 32-5. World Health Organization. 1978. Primary Health Care, Report of the International Conference on Primary Health Care, Alma Ata, USSR, 6-12 September. Health for all series, no.l. Geneva: World Health Organization. - 1988. Guidelines for rapid appraisal to assess community needs. Geneva: WHO/ SHS/NHP/88.4. Zakus, J.D., and C. Lysack. 1998. Revisting community participation. Health Policy and Planning 13(1): 1-12.

11 The Responsiveness of the Canadian Health Care System towards Newcomers ANITA J. GAGNON

There remains a great deal of variation in the health status of those living in Canada in spite of the existence of the Canada Health act. The Act was developed with a focus on 'equalizing' health status by 'equalizing' access to services, once thought to be the prime determinant of health. Today, people with particular health care needs include newcomers to Canada, specifically those with a precarious immigration status such as asylum seekers and those with a high-risk profile such as refugees arriving in Canada from conflict-laden areas. Specific newcomer needs are often not examined on a systematic basis. This paper describes variation in health care needs according to immigration classification and addresses the question: How well does the Canadian health care system respond to the needs of newcomers? 'Canadian health care system' refers here to services provided or supported by any level of government; while the term 'newcomers' refers to the following mobile populations: permanent residents, those with employment authorization, international students, refugees, and asylum seekers. The paper's findings are not meant to be conclusive; there was insufficient time to permit extensive verification of results. Further, the paper is not intended to be an exhaustive review of the literature or of the services offered; rather, it is meant to present enough detailed information to allow general comparisons of service provision to newcomers across Canadian provinces and territories and five countries (Australia, Netherlands, New Zealand, Sweden, and United Kingdom). These comparisons will offer a point of departure for a discussion of the extent to which Canada is responding to the needs of this marginalized group.

350 Anita J. Gagnon

Background Marginalized Sub-groups in Canada While Canadians and others living in Canada are, on the whole, a healthy group (Organization for Economic Co-operation and Development 2001), there remain important differences in sub-population health even though the application of the Canada Health Act guarantees access to hospital and physician services for most people. If differences remain even with 'universal' access to services, then factors other than service must play a role and this has been clearly outlined in the Ottawa Charter (World Health Organization 1986). Categories of health determinants are those based on the work of the federal, provincial, and territorial health ministers and include income and social status, social support networks, education, employment and working conditions, physical environment, biology and genetic endowment, personal health practices and coping skills, healthy child development, and health services (Huff and Kline 1999; Minister of Supply and Services Canada 1994). Each category affects health in a different way. If the goal of the various levels of government in Canada is to maximize the health of its population, then potential governmental responses must be viewed in this broad framework. With this in mind, marginalized groups can be defined as those disadvantaged in relation to one or several of the identified determinants of health. These then might include the physically or mentally disadvantaged, aboriginals, people living in poverty, or newcomers, to name but a few. Newcomers As a Potentially Marginalized Group One in six people in Canada is a newcomer (Statistics Canada 1996) and the newest migrants to Canada are more likely to be visible minorities, to come from less developed countries, and to be younger than those who arrived previously. Newcomers are a heterogeneous group. They include, for example, those arriving in Canada in the independent class of immigrants, including investors with a great deal of economic and other resources; family members arriving to be reunified with their families and therefore often having a social support structure in place upon arrival; and refugees, who may have significant histories of violence, varied economic and educational resources, and a reduced social support structure. Landed immigrant is a term used interchange-

Responsiveness of the System towards Newcomers 351 Table 11.1 Newcomers to Canada: Data for 2000 Newcomer category

Number

Permanent Residents (landed status)

Family Class Economic Class Live-in Caregivers Provincial Nominees Deferred Removal Order Class Post-determination Refugee Claimants in Canada Class

60,515 132,118 2,784 1,252 297 163

Temporary Workers

86,225

Students

63,618

Landed Refugees (Convention refugees)

Refugee Claimant / Asylum Seekers3

Government-sponsored Privately Sponsored Dependents abroad Landed in Canada

10,661 2,905 3,486 12,978 35,368

Sources: Citizenship and Immigration Canada, 2001 a; Citizenship and Immigration Canada, 2001 b. a Includes temporary residents allowed to remain in Canada on humanitarian grounds those from countries on the moratorium list, those with minister's permits, and those with a refugee history but who have not filed a refugee claim.

ably with permanent resident, the latter being used preferentially in several countries (and consequently used in this text). Table 11.1 summarizes Canadian immigration figures. Each category of newcomer has different resettlement needs and the administrative classification of the migrant determines accessibility and eligibility to government health and social services. Studies of newcomers have been unable to address the effect of this heterogeneity because to do so would require the application of specific sampling strategies to ensure an adequate number of individuals in the various administrative classes of immigrants. Existing national databases are inadequate to address the complexity of health determinants faced by newcomers (Dunn and Dyck 1998). The Canadian Census, for example, provides population figures for mother tongue and language spoken at home, number of landed immigrants

352 Anita J. Gagnon

received (all categories) and when they were received (i.e., = 3 or 5 previous years), non-permanent residents, ethnic origin, members of the 'black' community, and religious affiliation (Gravel and Battaglini 2000). Migration history and associated risks and effects cannot be specified with data aggregated in this way. Birth and death data provide other relevant statistics, but, again, they lack the level of detail needed to examine issues specific to various classes of migrant. The migration histories presented by certain newcomers and variation in service access by immigration classes suggest that an important number of newcomers should be considered marginalized, with those in the refugee and asylum-seeking class with the most precarious health status. 'Refugees' refers to 'Convention refugees' in the vast majority of cases in that they meet the definition of a refugee provided in the 1951 Geneva Convention relating to the Status of Refugees (United Nations High Commission for Refugees [UNHCR]): 'owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership in a particular social group, or political opinion, is outside the country of his nationality, and is unable to or, owing to such fear, is unwilling to avail himself of the protection of that country' (UNHCR 2000). Assignment of an immigration classification of 'refugee' occurs in two main ways in Canada: having been selected overseas through UNHCR channels; and through a successful request for asylum. Requests for asylum are assessed by Immigration Refugee Boards (IRBs). Individuals are classified as 'asylum seekers' until their case has been assessed by an IRB. Requests for asylum can be made at two points. The first is at the border, and these individuals are labelled 'border claimants' or 'port-of-entry claimants.' The second is in the country, and these cases are labelled 'inland claimants.' The distinction between 'refugee,' 'border claimant,' and 'inland claimant' is important because it suggests that different health determinants are operating to various degrees in the lives of these individuals which, in turn, will affect their health status and their response to illness. This administrative classification distinction further defines which health services may be accessed and when. Details of these aspects will be discussed below. Immigration to Canada over the last decades has grown to 17.4 per cent of the total population (Statistics Canada 2000). Of an immigration total of three and one-half million from 1979 to 1999, 531,417 persons were refugees, with an average of 24,000 per year over the last five years (7,000 of whom were selected through UNHCR channels) (Citizenship and Immigration Canada 2000). The source regions for migra-

Responsiveness of the System towards Newcomers 353

tion over this period have been mainly Asia, the Pacific, Africa, and the Middle East. It should be noted that refugees, asylum seekers, and other immigrants concentrate in Canada's major urban centres, and thus the burden of health care falls disproportionately on those areas. The Needs of Newcomers The Health of Newcomers to Canada Differences between the experiences of those in the various immigration classes have been reviewed (Gravel and Battaglini 2000). When examined together, newcomers are multi-ethnic, use a variety of languages and are characterized by a diversity of religious traditions, lifestyles, and political alliances. As opposed to 'refugees/ 'independent' or 'family' immigrants choose to come to Canada and are motivated to leave their country and re-establish themselves in a new country in the hope of a better life. Their departure is planned and they are able to return to their countries of origin if they so choose. On the other hand, 'refugees' are forced to leave their country to ensure their survival and they are not able to return to their country of origin. All immigrants will go through phases of adjustment; however, the permanent, forced nature of the refugee migration experience makes their integration into society more difficult (Sundquist et al. 2000). These variations in migration scenarios suggest that several health determinants related to migration need to be considered (Gagnon, Merry, and Robinson 2002), including whether departure from the country of origin was of a forced nature, whether the individual comes from a rural versus an urban area, experience of war in the source country, camp experience, history of torture or abuse, family separation, existence of a 'like community' in the new country, length of time in the new country, ability to speak either French or English, discrimination experience, and change in social status. Additional bio-psychosocial determinants particularly relevant to migrants include usual versus receiving-country diet, nutritional status, infectious-disease exposure, social isolation, stress/anxiety, stress-management strategies, depression, post-traumatic stress disorder, somatization, services available/received (Western and traditional), and cultural compatibility with the health service provider. Women face additional risks, which suggest consideration of factors including sex and gender-based violence (including rape), post-abortion care, and STI/HIV exposure and pregnancies resulting from rape.

354 Anita J. Gagnon

Immigrants Generally Immigrants to Canada and the United States have been found to be healthier than their North American counterparts although they lose this advantage over time (Chen, Ng, and Wilkins 1996, Parakulam 2000; Pomerleau 1997; Rumbaut 1996; Weeks 2000). Immigrants must meet certain health and other criteria to be accepted into Canada and they must have enough economic resources to complete the process of immigration, which likely results in a 'healthy immigrant' effect. Individuals of various immigration classes are usually not separated out in such reports. Thus, it is likely that the health status of 'economic' or 'family' immigrants overwhelms the data on newcomers in other classes because of the volume of immigrants in that category (90 per cent of all immigrants fall into either the economic class or the family class) (Citizenship and Immigration Canada 2001 a). Refugee Health Canadian studies focusing on the health of refugees or asylum seekers have examined general health (Thonneau, Gratton, and Desrosiers 1990) and mental health (Beiser et al. 1995; Beiser and Edwards 1994; Beiser and Fleming 1986; Beiser and Hyman 1997; Dillmann, Pablo, and Wilson 1993; Stephenson 1995). Two studies in particular are worthy of note. The first, conducted in Calgary, sheds light on the health of government-sponsored refugees in Canada (that is, those selected overseas via UNHCR channels, in the 'refugee' class). Health problems identified most often in the 1,104 government-sponsored refugees who arrived in Calgary in the early 1990s included upper-respiratory tract infections (17.8 per cent), impaired vision (15.4 per cent), dental emergencies (12.8 per cent), ear infections (7.4 per cent), gynaecological problems (6.1 per cent), and obstetrical conditions (5.6 per cent). Twenty-two (5.8 per cent) required emergency hospitalization (Dillmann, Pablo, and Wilson 1993). The second study, conducted in Montreal in 1985-6, sheds light on the health of refugees who seek asylum in Canada at the border or once in the country (as opposed to via UNHCR channels - that is, 'asylum seekers' (Thonneau, Gratton, and Desrosiers 1990). In all, 1,994 applicants received a medical examination at a clinic identified by the Quebec government as the care provider for refugee applicants. Of these, 87 per cent were considered to be in good health, 3 per cent had a major handicap or chronic illness, and 10 per cent were in poor health. Health problems identified included nutritional deficiencies, stunted growth, anaemia, parasitic infections, syphilis, risk for tuberculosis,

Responsiveness of the System towards Newcomers 355 355

inadequate vaccination, and evidence of physical torture. Other more recently arrived refugee groups have been found to have a similar health profile, with malnutrition, dehydration, gastrointestinal problems, scabies, head lice, minor skin conditions, and hepatitis B identified (Kent 2000). A review of the mental health of refugees and immigrants to Canada (Beiser and Edwards 1994), based largely on the experiences of Southeast Asian refugees, showed that exposure to catastrophic stress such as torture and rape may lead to post-traumatic stress disorder, that women are at increased risk of mental disorder over time, and that adolescent and older women may be at higher risk owing to the conflicting expectations of parents versus peers in the case of adolescents and to the greater value placed on youthfulness over wisdom in the case of elderly people (Beiser, Dion, Gotowiec, Hyman, and Vu 1995; Beiser and Edwards 1994). Unemployment was suggested as being an important problem since refugees admitted for compassionate reasons may not have employable backgrounds. Extended separation from family was also suggested as having a harmful psychological impact. On the other hand, resettlement was thought to be facilitated by having adequate sources of material support, the presence of like-ethnic communities (which may provide a psychological sense of community for some ethnic groups), and individual psychological hardiness. Higher-Risk Groups As previously mentioned, female refugees are exposed to even greater health risks because of issues related to childbearing and sexual health. Children are another at-risk subgroup on account of their vulnerability, and the numbers arriving unaccompanied in Canada are growing (Citizenship and Immigration Canada 2001a). Trafficked migrants are yet another high-risk category (United Nations High Commissioner for Refugees 1998). The United Nations estimates that four million individuals are trafficked internationally each year. Of these, 700,000 are women or children. Trafficking is said to occur when a migrant is illicitly recruited and/or moved by means of deception or coercion for the purpose of economically or otherwise exploiting the migrant, under conditions that violate their fundamental human rights (International Organization for Migration 2001). Promises of good jobs from human smugglers or dreams of a better life often lead migrants to purchase irregular or illegal immigration services without knowledge of the risks involved (Gushulak and MacPherson 2000). The number of

356 Anita J. Gagnon

trafficked migrants and the degree of risk in the method of transport may be increasing (International Organization for Migration 2001). Poor living and working conditions stimulate irregular migration and these same factors are associated with poor health. Other higher-risk migrant groups include those placed in dependent positions, such as those with a fiance and a temporary work visa, and those participating in the live-in caregiver program. Canadians' Values about Responding to Newcomers' Needs Canadians' attitudes towards immigration and immigration policy have been compared to those of other country nationals using data gathered from national surveys conducted in each of the countries between 1970 and 1995 (Simon and Lynch 1999). More Canadians favoured increasing immigration than did Americans, yet, at the peak, the number accounted for only 17 per cent of the population. Those preferring a decrease ranged from 32 to 55 per cent, with an average of 42 per cent, lower than the Australia average by over twenty points and the U.K. average by ten points. The U.S. average was closest, with a range of from 33 to 66 per cent, or an average of 49.7 per cent, of respondents preferring a decrease. Responses to a 1993 Canadian question on how the increase in the number of immigrants arriving from Asia, the West Indies, and other mainly Third World countries was viewed, showed only 20 per cent saying that it was a good thing. Determinants of negative attitudes towards immigration have been found to be the result of a complex interrelationship of factors that include unemployment rates and racism (Palmer 1996). These studies may suggest that some Canadians do not support providing services to newcomers, but, in the absence of other data, this cannot be confirmed. If it were determined to be true, there may be variation in degree of support depending on immigration class (humanitarian versus non-humanitarian). Potential Barriers to Access and Use Immigrants have been found to underutilize health services for which they are eligible (Hyman 2001). Barriers to services used by migrants are several but can be divided into three broad categories. The first is a fear of accessing the system, grounded in the belief that they might be considered a burden on the system and accessing care would therefore be detrimental to their immigration applications (personal communication, A. Li); the second relates to culturally inappropriate care and Ian-

Responsiveness of the System towards Newcomers 357

guage barriers; the third concerns difficulties navigating the system. There is a large body of literature describing the effects of culturally inappropriate care across various multicultural groups. In general, it is suggested that health care providers learn about their individual patient's cultural concepts, family structure, and practices at the outset of developing a working relationship (Hamilton 1996). The professional can then offer culturally acceptable explanations of healthrelated matters. Developing a trusting relationship can be further hampered when patient and provider do not speak a common language (Bowen 2000; Davies and Yoshida 1981). In these cases, a interpreter may be required, which compromises the patient's privacy; often a family member is sought to provide this service, resulting in altered power dynamics within the family. Counselling services are extremely difficult to manage in these situations. Delays in seeking care, reduced comprehension, reduced compliance, and harmful health outcomes are associated with language barriers. Informed consent for care is difficult to obtain under these conditions. Difficulties navigating 'the system' are not unique to newcomers but are compounded by language and cultural barriers and the newcomer's expectations, which are usually based on experiences with the system in the country of origin. Another 'system' barrier is the underrepresentation of immigrant and racial minorities, especially women, in health care professions (with foreign-earned credentials an issue) and on boards of directors of hospitals, universities, and other major institutions that set health care policies (Simms 1996). 'Best Practice' Guidelines for Health Services Delivery for Newcomers The Canadian Council for Refugees (CCR) has published Best Settlement Practices, in which guidelines for best practices are outlined (Canadian Council for Refugees 1998). Services should: be accessible to all who need them; be offered in an inclusive manner, respectful of, and sensitive to, diversity; empower clients; respond to needs as defined by users; take account of the complex, multifaceted, interrelated dimensions of settlement and integration] be delivered in a manner that fully respects the rights and dignity of the individual; be delivered in a manner that is culturally sensitive; promote the development of newcomer communities and newcomer participation in the wider community, and develop communities that are welcoming of newcomers; be deliv-

358 Anita J. Gagnon

ered in a spirit of collaboration; be made accountable to the communities served; be oriented towards promoting positive change in the lives of newcomers and in the capacity of society to offer equality of opportunity for all; and be based on reliable, up-to-date information. Although not drafted specifically for health services delivery, these guidelines offer a framework by which the Canadian health care system can be examined to determine how well the needs of newcomers are being met. National standards developed in the United States may also be useful (U.S. Department of Health and Human Services 2001). Summary of Newcomers' Needs The health of newcomers likely varies by migration history, although databases are inadequate to answer this question directly. Migrants not forced out of their country of origin are expected to have a lower health-risk profile than those forced out. Newcomers are known to underutilize services and some Canadians likely prefer that limited access to services be available to newcomers. Barriers to accessing care for which they are eligible include fear of jeopardizing immigration applications by seeking care, language difficulties, culturally inappropriate care, and problems in navigating the system. Optimal care is thought to be based on core values of access, inclusion, empowerment, user-defined services, holism, respect, cultural sensitivity, community development, collaboration, accountability, orientation towards positive change, and reliability (Canadian Council for Refugees 1998). Governmental Responses to the Health Needs of Newcomers Methods Information on access to services from all provinces and territories (regardless of volume of immigration) were sought, collated, and tabulated for comparison. Certain countries were selected for comparison purposes based on similarity of immigration patterns, economic systems, and health care structures. The vast majority of information was obtained through extensive searches of governmental and non-governmental websites and their cited documents, and by subsequently contacting individuals and organizations via e-mail for missing information. Information was sought from ministries of health, government health-insurance agencies, ministries of immigration, regional health and social service agencies, main municipal, provincial, territorial, and

Responsiveness of the System towards Newcomers 359

national offices, and non-governmental organizations (NGOs) including both those that act as umbrella organizations for other NGOs and those directly serving migrants. Google, Yahoo, and Alta Vista search engines were employed. Electronic databases of literature were also used to supplement information. Bibliographic references from identified texts were sought for review. The CCR list-serve was used to identify case examples of access issues for newcomers. Access to Services Tables 11.2-11.7 (in Appendix) provide details on access to services. The legend for the tables is presented first and the bibliography of documents used to compile information found in the tables follows the final table. In terprovincial/Territorial Comparisons

The Canada Health Act guarantees access to hospitalization and physician services for Canadians (Romanow 2002). Other health services, including professional services (e.g., nurses), preventive care, home care, long-term care, dental care, and pharmacare (out of hospital), vary by province. The extent of coverage is detailed in Table 11.2 and shows that some non-medical professional services are covered in nine provinces/territories, some preventive services in five, home care in five, long-term care in two, some dental care in eleven, and some pharmacare in eight. Access to these services varies by province according to the immigration class of the individual. These classes include permanent residents, those with employment authorizations, students, student's spouses, refugees, and asylum seekers. Provincial/territorial variation in access to services by immigration status is detailed in Table 11.3. As can be seen, landed immigrants have access to the same services as Canadians. Those with employment authorizations have the same access depending on the duration of the authorization; in all but three provinces, the authorization must be for six or twelve months. Students and their spouses are specifically not covered in four provinces. Either the provincial plans or the Interim Federal Health Program (IFHP) (described below) variously cover refugees. Asylum seekers are not covered under provincial plans but rather under the IFHP. The point in time during the immigration process at which newcomers are covered for these services also varies considerably and is detailed in Table 11.4. The time at which coverage begins for all immi-

360 Anita J. Gagnon

gration classes except refugees and asylum seekers is up to three months for all provinces, except Nova Scotia where those with an employment authorization have access after six months and students after twelve months. Time to access health plans for provinces and territories that offer refugees coverage varies, with the shortest being in the Northwest Territories, where access is upon arrival. There is a twoto three-week delay in Quebec, a three-month delay in British Columbia and Ontario, and in Manitoba the timing depends on the employment authorization. The least amount of information is available for asylum seekers. However, reports from British Columbia suggest a one- to four-week delay; for Quebec, same day to ten weeks; and for Ontario, four weeks to six months. Variation in access for individuals in these latter two immigration classes suggests a greater dependence on the IFHP. Care Assured Directly by the Canadian Government The IFHP was put in place for humanitarian reasons to allow refugee claimants (asylum seekers), Convention refugees, and others in humanitarian-designated classes to receive essential health care (Citizenship and Immigration Canada 1998). It is not meant to replace provincial health plans and does not provide the same extent of coverage as that for permanent residents. Eligibility for this program is determined by a demonstrated lack of funds, which is assessed by an immigration officer, and this only once eligibility to make a refugee claim has been approved and if the claimant is not covered by a private or public health plan. IFHP benefits are limited to essential health services for the treatment and prevention of serious medical/dental conditions, essential prescription medications, contraception, prenatal and obstetrical care, and the Immigration Medical Examination (limited to identification of active tuberculosis, syphilis, and acute psychiatric disorders [personal communication, B.D. Gushulak]). Prior approval is required for, among other things, a complete physical examination, diagnostic services (unless short-term complications are seen), ongoing psychiatric care, and psychotherapy/counselling. Alternative and over-the-counter medications and root canals, among other treatments, are expressly not covered. Even with the IFHP, certain asylum seekers have no health insurance for some periods of time owing primarily to delays in processing their requests to apply for refugee status. Thus, those who are awaiting eligibility determination to make a refugee claim have no coverage. Health

Responsiveness of the System towards Newcomers 361 and social services provision during this waiting period varies according to services available through local NGOs and community health centres. Inland claimants have the most difficulty. According to one respondent to a CCR list-serve question regarding variation in service availability, refugees in Ontario can wait up to one year to obtain coverage (personal communication, South Etobicoke Community Legal Services, Toronto), although the usual delay in Montreal is reported to be three weeks (personal communication, D. Isaacs). As recently as 8 April 2002, in Toronto, people were said to be lining up overnight in order to submit their applications for refugee status (personal communication, E.F. Khaki). Port-of-entry asylum seekers could receive the necessary forms as they enter the country or up to one month thereafter. If there are problems with their 'Notification of Intention to Make a Refugee Claim' form, it can be returned to them and further delays occur in obtaining IFHP coverage. Concerns with regards to the IFHP have been summarized by the CCR based on reports from twenty-seven of its members and from A Report on the Experience of Sponsors of Kosovar Refugees in Ontario, published by the Centre for Refugee Studies (CRS) and the Centre of Excellence for Research on Immigration and Settlement (CERIS) (personal communication, R. Hogue). These include: difficulty in accessing IFHPcovered services by those eligible due, in large part, to a lack of willingness on the part of potential providers to process IFHP forms or a lack of knowledge about it (difficulty in accessing services is exacerbated in areas where there is a shortage of health professionals); processing difficulties at the IFHP office, including pre-authorization for certain conditions and late reimbursement to providers; and absence of coverage for certain conditions, dental care being cited most often. These difficulties result in fewer service options available for a needy population (personal communications, P. Dongier; A. Li). This feedback on IFHP must be interpreted cautiously given the apparent absence of a representative survey; however, it suggests that health care coverage for refugees and asylum seekers and services received by them are not uniform, and the rate at which they have access to services differs from one province to the next. No such difference in health care coverage can be seen for other classes of immigrants or for people born in Canada. International Comparisons Coverage of health services in Australia, the Netherlands, New Zealand, Sweden, and the United Kingdom is detailed in Table 11.5.

362 Anita J. Gagnon

The most complete coverage seems to be available in the Netherlands. Four of the five countries cover hospitalization, while three completely cover physician services and some other professional services. Preventative care is widely covered in two countries. The greatest variation by country is seen for home care, long-term care, dental care, and outof-hospital pharmacare. Coverage for dental care, for example, ranges from no coverage in Australia to complete care and treatment for a specified time period in the United Kingdom, and to full preventative care and treatment in the Netherlands. Access to these services varies by country according to the individual's immigration class. Access to services by immigration class in selected countries is detailed in Table 11.6. Residents have the greatest access and students the least. Refugees have immediate access to the same health care as nationals in every country. Asylum seekers have access to the same health care as nationals in the Netherlands, New Zealand, and the United Kingdom. Information describing the point in time during the immigration process at which newcomers are covered for these services was more difficult to obtain. Available information is detailed in Table 11.7. Nonhumanitarian immigrants have a waiting period ranging from three months in Sweden to two years in Australia. Refugees have immediate coverage in three of the countries examined. Asylum seekers theoretically have immediate access, although processing time for their applications results in documented delayed access to services in at least New Zealand and the United Kingdom. Summary of Access to Services

Canada has responded to certain needs for access to care by implementing the IFHP. However, there remain serious difficulties with accessing services through this program, including delays in immigration-processing times, errors in completing forms, limitation to coverage of essential services only, and the administrative burden placed on service providers. Other comparable countries have generally not created a separate funding mechanism for health care for refugees and asylum seekers, but rather have limited access to available services according to immigration class and length of time in the country. 'Best Practice' Examples In ter-Provincial/Territorial 'Best practice' examples from various provinces are too numerous to

Responsiveness of the System towards Newcomers 363

describe here. Advocacy and networking are the activities of groups focusing on immigrant-resettlement issues (Mosaic 2002) and the umbrella organizations representing them (Table de concertation des organismes au service des personnes refugiees et immigrantes [TCRI] 2001; Ontario Council of Agencies Serving Immigrants 2002; The Affiliation of Multicultural Societies and Service Agencies [AMSSA] 2002. Several specialized service agencies are in existence. These include agencies focusing on culturally appropriate mental (Kirmayer 2002; Regie regionale de la sante et des services sociaux de Montreal-Centre 1996) and physical health services (Lorion and Dion 1999) and interpreter databanks (Regie regionale de la sante et des services sociaux de Montreal-Centre 2002). The concerns of female immigrants have been highlighted in some areas (Maritime Centre of Excellence for Women's Health 2000; Riverdale Immigrant Women's Centre 2002), and guidelines for practices have been published (Ministere de la sante et des services sociaux 1999). Umbrella and direct service agencies are variously supported and often rely on government funds to some extent. Federal/National

There are some 'best practice' examples at the federal level. Among these is the Metropolis Project, which was awarded the Public Service Award for Excellence in Policy Development in 1999-2000 (Canadian Heritage Multiculturalism 2000). This project brings together policy makers from three levels of government, NGOs, and researchers in an effort to examine broadly the effects of migration on Canadian cities. Several federal departments fund the project. Four research 'centres of excellence' (in Toronto, Vancouver, Montreal, and the prairies) have their own partnerships with local NGOs and policy makers (Metropolis Canada 2002). There is also an international component of the project involving several countries. Another initiative at the national level that appears promising is the Multilingual-Health-Education Network (Multilingual Health Education Net 2002). This network is funded by Health Canada and Heritage Canada with the goal of improving standards and procedures for translating patient education and instructional material. It is a nonprofit alliance of Canadian health agencies focused on improving service delivery by making translated documents easily available to providers and the public. The Canadian Council for Refugees, the Canadian Network for the Health of Survivors of Torture and Organized Violence (RIVO), the National Organization of Immigrant and Visible Minority Women

364 Anita J. Gagnon

(NOIVMWC), and the Canadian Council on Multicultural Health (CCMH) are agencies offering important resettlement service or policy work nationally. International 'Best practice' examples from various countries and regions of the world have been identified. Of specific interest to Canada are: the New Zealand language-law and resettlement program, the Australian translation services, and the European Union Networks on Reception, Integration and Voluntary Repatriation (EU Networks). New Zealand law states that every individual has the legal right to an interpreter for health services to ensure informed consent and this service must be provided by the treating agency (New Zealand Ministry of Health 2002). Refugee Health Care: A Handbook for Health Professionals has been developed by the New Zealand Ministry of Health in conjunction with several community and professional organizations (New Zealand Ministry of Health 2001). Approaches to care for refugees and asylum seekers are discussed, including care provided at the Mangere Refugee Resettlement Centre. Here, refugees spend six weeks upon their arrival; they receive health screening and care, English lessons, a basic orientation to New Zealand, and money to pay for essentials. There are currently six health care interpreter Service (HCIS) centres covering the State of New South Wales in Australia (New South Wales 2002). This service consists of professional interpreting services provided free of charge in more than fifty languages to people using public health services (which are the majority) in areas around Sydney. In metropolitan areas, these services are available twenty-four hours a day, seven days a week, depending on urgency. In some areas, they are also available for telemedicine consultations. Other country-specific 'best practice' programs in the European Union have been described elsewhere (ECRE Task Force on Integration 2000). EU Networks is a project financed by the European Union that focuses on improving the integration of refugees in member countries (EU Networks on Integration of Refugees 2002). It aims to develop issue-oriented networks on relevant themes such as education, language and vocational training, health, reception, and voluntary return. Activities have included maintaining an inventory of integration activities, developing good-practice guidelines, and publishing reports of interviews with refugees from all fifteen member countries. The International Rehabilitation Council for Torture Victims (IRCT)

Responsiveness of the System towards Newcomers 365

is an independent, international organization of health professionals which promotes and supports the rehabilitation of torture victims and the prevention of torture worldwide (International Rehabilitation Council for Torture Victims 2002). Working with two hundred rehabilitation programs internationally, including several in Canada, it monitors torture globally and intervenes in conflict situations. It is at least partially supported by the European Union, the United Nations Voluntary Fund for Victims of Torture, and the Oak Foundation. Summary of 'Best Practice' Examples Across Canada and at the national level, there are several special programs for newcomers that respond to many of the 'best practice' guidelines. 'Best practice' examples from other countries that Canada may wish to consider include New Zealand's language law requiring interpreter services for everyone in need and Australia's extended interpreter services, as well as multidisciplinary resettlement centres in New Zealand that provide a range of services to newly arrived refugees on a regular basis. Endeavours spanning more than one country that may be of interest to Canada include the EU Networks, which address common resettlement issues, and ICRT, in which several Canadian agencies already participate.

Comment: Remaining Dilemmas, Challenges, and Choices Service Gaps The existence of a separate financing structure for health services for refugees and asylum seekers (the IFHP) offers humanitarian-class immigrants coverage for basic services; however, owing to the heavy administrative burden placed on providers, there are no built-in incentives for service providers to accommodate the needs of these individuals. Other barriers include periods of time without access to care on account of immigration-processing delays and lack of coverage for care not included in the IFHP. Language was repeatedly cited as a barrier throughout the country, and although the MultilingualHealth-Education Network has been set up to help lower these barriers through written documentation, a more active approach is needed. The notion that care might be provided in the absence of truly informed consent as a result of language difficulties is of some concern.

366 Anita J. Gagnon

Inadequate Information and Data Sources Information available on the Internet describing health services directed towards newcomers to Canada is limited; for most provinces, there is no mention of 'immigrant' on websites designed to provide information to users about accessing services for which they are eligible. Federal and provincial health databases are currently inadequate to assess the extent to which immigration class and migration history affect health outcomes and services use, because data on immigration class and migration history are not included in databases at a level of detail that would allow such comparisons. Thus, because of the invisibility of newcomers within existing databases, the effect of any changes in policy or structure of the health care system will be difficult to evaluate among this population group. Furthermore, there is little population-based research in this area. Public Opinion on Immigration A portion of the Canadian population may believe that Canada is already providing too many benefits to newcomers. When contrasted against Canada's commitment to the 1951 Convention as well as care provided by other countries, however, the level of care offered by Canada to its newcomers at greatest risk appears moderate. One approach to meeting the needs of both the public and newcomers might be to decrease immigration-processing time as much as possible such that those not meeting refugee criteria could be identified rapidly, while simultaneously covering those at greatest risk for health decline during the period in which they are awaiting an IRB decision on their application. Recommendations Issue 1: Administrative burden on health care professionals providing care under the IFHP Potential Actions: • reduce cumbersome processing requirements for reimbursement of services provided; • develop a health care card system such as the one that exists for the provinces as a reasonable alternative to the current system;

Responsiveness of the System towards Newcomers 367 • develop and distribute pamphlets and other information resources on the IFHP to health care providers; and • create financial or other incentives for health care providers to provide care to individuals covered by the IFHP Issue 2: Eligibility delays for the IFHP Potential Actions: • monitor length of time for IFHP eligibility for the IFHP from the date of entry into the country until the date of eligibility to determine bottlenecks and permit benchmarking; and • reduce processing time for acknowledging reception of applications for asylum so that individuals eligible for the IFHP can access health services as soon as possible Issue 3: Limited service coverage under the IFHP Potential Actions: • expand services covered under the IFHP, with special consideraion to mental health services (given the psychological and emotional implications of the migration histories of individuals covered by this program) and dental services; and • expand coverage directly to specialized centres (e.g., for victims of trauma or violence), especially in urban areas Issue 4: Lack of access to appropriate care and to informed consent among people who do not speak an official language Potential Actions: • establish translation in health care as a right for newcomers; • set up local and/or national telephone interpreter services; • form alliances with NGOs dedicated to providing translation services; • fund training in the development of national professional standards of practice to increase cultural sensitivity towards newcomers and to reduce language and ethnocultural barriers; • develop resource materials on ethnocultural care for professionals; • finance ethnocultural liaison staff in hospitals and community health centres to work with various ethnic groups and to help newcomers 'navigate' the system; and

368 Anita J. Gagnon

• form an umbrella organization in North America (similar to the EU Networks in Europe) which would be designed to support the work of NGOs and governmental agencies in providing services to immigrants of all classes (this could reduce duplication of services and offer the benefit of sharing effective resettlement strategies) Issue 5: Paucity of data upon which to build sound policies and practices Potential Actions: 9

include information in federal and provincial databases that reflect immigration class, migration history, visible minority status, and official-language proficiency so that implications of health-risk status and access to care can be more readily quantified; • require documentation of gender and age effects of both Citizenship and Immigration Canada and Health Canada procedures to assess the effectiveness of our responses to newcomers at greatest risk; and • collect and regularly report on data regarding job skills, Canadian certification in specialities, and time to employment in area of speciality, considering the importance of gainful, meaningful employment as a key determinant of health Issue 6: Paucity of easily available information on eligibility of newcomers to services Potential Actions:

• have information about exact coverage of health services for newcomers readily available on relevant national and provincial websites and in other written and electronic resources; and • create an information system (e.g., 1-800 number), for newcomers covered by the IFHP and elsewhere, on the most efficient way to access various specific services

Issue 7: Health care financing structure not taking into consideration the urban-rural discrepancy of service provision to newcomers Potential Actions:

• Restructure financing for health services according to the geographic distribution of newcomers; and • offer more services in urban areas

Responsiveness of the System towards Newcomers 369 Conclusion

Canada has been moderately responsive to the health needs of newcomers. Yet there are several approaches that Canada could take to improve the responsiveness of the system even further. These include lightening the administrative burden of the IFHP on health care providers and reducing the time to access the program, offering interpretation services widely, and improving the content of health databanks to allow the examination of newcomers' health and health care according to immigration class. Appendix Legend for Tables 11.2-11.7 Code A B C D F G H I J N O P R S T U V X Y Z

Definition Subject to means testing Including spouse and dependants Authorization must be valid for a minimum of six months Authorization must be valid for a minimum of twelve months Covered by the Interim Federal Health Program Group-specific coverage Covered if provided in hospital Information incomplete/unclear Immediately once granted this status Not applicable Spouse and dependants must submit individual requests to be included in a claimant's application Partially Retroactive Surgical procedures Varies depending on length of time for claimant status application to be processed Upon arrival Variable No Yes Certain services covered

Table 11.2 Health care services covered for Canadians Health care system components Hospitalization Physician services Other professional services Preventative care Home care Long-term care Dental care Pharmacare (out of hospital)

AB

BC

MB

NB

NF& Labrador

NWT/ Nunavut*

NS

ON

PEI

QC

SK

Yukon

Y Y

Y Y

Y Y

Y Y

Y Y

Y Y

Y Y

Y

Y Y

Y Y

Y Y

Y Y

ypbQC

yd

I I I

Y9 I I I PS

I I I I HPS

I lp I I

Zk Yr I

HSy

YeGf I Y I HPS

P'QJ

YG° Yu

I Is I Pv I GddpS

P' YG 1 P 1 P

I

I

Gee

I

P

P

Y Gx I

Yh q

G I I

I Y I

SzGaa

HGbb

HPS Gcc HPS

G"

Ggg

Qphh

yw

GP"

ymGn

Y

1 1 1

Gji

NWT health care coverage is in effect in Nunavut. Excludes health care coverage provided by federal government to certain groups (Aboriginals, veterans, and inmates) Chiropractic services, podiatry, physical therapy, respiratory therapy, occupational therapy, speech pathology services, some oral and maxillae-facial services from oral surgeons. Optometry services covered for those nineteen and under or sixty-five and over. Surgical, psychiatric, ophthalmologic, diagnostic, and anaesthesia services when referred by physician. Surgical, diagnostic, and chiropractic services when referred by physician. Ambulance services not covered. Optometry services covered for those nineteen and under or sixty-five and over. Community mental health services. Ambulance, chiropractic, and optometric services not covered. Podiatric, chiropractic, and osteopathic services. Ophthalmologic services available every two years for those twenty to sixty-four years of age and once a year for those under twenty or sixty-five and over.

Table 11.2 (Concluded) k

When referred by physician. Physiotherapy, occupational therapy, ambulance, optometric, and mental health services. m Mental health services. n Glasses and eye examinations for children of low-income families under the Children's Drug and Optical Program (CDOP). 0 Children covered for immunization programs. p Yearly physicals not covered. q The Optometric Program provides preventative routine vision care for children ten and under, and for seniors sixty-five and over. r Covered if provided in an accredited facility. s Smoking-cessation medication covered. ' Screening mammography for women aged fifty to sixty-nine. u Home respite and palliative care. v Light and heavy housekeeping, meal preparation, and errand running. w If provided in an accredited facility. x Some services covered for seniors. y Related to severe congenital abnormalities. z Services specific to injuries related to jaw injury or disease. aa Seniors sixty and over. bb Coverage for children up to the age of ten. cc Children qualifying for the Children in Need of Treatment Program. dd Children ten and under and those receiving financial assistance. ee For seniors, children in care, and those with certain diseases/conditions. ff Seniors sixty and over and those with diseases/conditions covered by the extended health benefits plan. 99 The Senior's Pharmacare program covers adults sixty-five and over. hh For those eligible for the Ontario Drug Benefits Program. " Drug-cost assistance schemes available to those with specific diseases/conditions. ji Seniors sixty-five and over and their spouse if sixty and over. '

Table 11.3 Health services coverage (refer to Table 11.2) as per immigration status: Canada Categories of immigration status Permanent residents People with employment authorization Students Spouses of students Refugees Refugee claimants/ asylum seekers

AB

BC

MB

NB

Labrador

NWT/ Nunavut*

NS

ON

PEI

QC

SK

Yukon

Y

Y

Y

YB

Y

Y

Y

Y

Y

YBO

Y

I

YBD YD

YBD X X FY9

YBODa

YbBD

X X Y

X X I

YD Y I I

Y Y I Y

YBC I I Y

YC X X I

YBOC YdBO

Y Y

YB Y Y Ff

YeO YBO

YBC Y I FYh

I I I I

F

F

F

F

F

F

F

F

F

F

F

F

NF&

* NWT health care coverage is in effect in Nunavut. Those married to or a dependant of an eligible NB resident are also entitled to coverage. b Employment authorization must be issued before coming to Canada for a named Newfoundland employer for a specific job within the province. c If spouse does not arrive within the twelve-month period following the arrival of the person with employment authorization, coverage differs. d Students from countries not included in the 'social security agreement' (entente de securite sociale) with Quebec are not covered. e Spouses of students from countries not included in the 'social security agreement' (entente de securite sociale) with Quebec are not covered. f Eligible for supplementary benefits including physiotherapy, massage therapy, chiropractic and naturopathic services, and surgical and non-surgical podiatry with a combined annual limit of ten visits per year. 9 Covered by IFHP until the time employment authorization is granted. h May be eligible if in possession of valid employment authorization, student authorization, or a minister's permit issued by Citizenship and Immigration Canada. a

Table 11.4 Time period between arrival and eligibility for health services coverage as per immigration status: Canada Categories of immigration status

NWT/

SK

Yukon

3 months Uc

Up to 3 months

1st day of 3rd month

I

After 6 months Rd

3 months U

Up to 3 months

1st day of 3rd month

I

U

1st day after 13th month6

I

N

U'

1st day of 3rd month

I

N

I

I

I

N

U°

N

I

I

N

J

I

3 months' I

2-3 weeks

N

I

T

T

T

T

Vk

V

T

T

BC

MB

NB

Ua

Up to 3 months

Ub or the 1st day of the 3rd month

People with employment authorization

3 months

Up to 3 months

Students

3 months

Spouses of students

Refugees

Permanent residents

NF & Labrador

QC

AB

Nunavut*

NS

ON

3 months I

U

I

U or the 1st day of the 3rd month

3 months I

U

Up to 3 months

N

N

N

3 months

Up to 3 months

N

N

I

3 months

Vh

PEI

Min.

Refugee claimants/ asylum seeker T *

a

b c

1-4 weeks' T

T

NWT health care coverage is in effect in Nunavut If application is received within the three months following arrival, coverage is effective upon arrival. For applications received after the required time, the effective date is determined at the time of registration. Or as soon as immigration status is confirmed, for within-country applicants. Those establishing residence in PEI may be eligible on the first day they become resident of the province.

Table 11.4 (Concluded) d

Those with employment visas: after residing six months retroactive coverage back to the date of arrival provided they have not been absent for more than thirty-one days from NS and are intending to be employed for the following six months. Those with employment authorization: eligible to apply for coverage on the first day of the seventh month provided that they have not been absent from NS for more than thirty-one days. 8 Provided they have not been absent from NS for more than thirty-one consecutive days. Coverage begins on first day of seventh month if student is working as a teaching or research assistant at a university in NS. 1 Students from countries not included in the 'social security agreement' (entente de security sociale) with Quebec are not covered. 9 Spouses of students from countries not included in the 'social security agreement' (entente de security sociale) with Quebec are not covered. h IFHP coverage in effect until employment authorization is issued, at which time provincial coverage takes effect. ' Eligible three months after the landing fee is paid. ' May be delayed by sudden influx of claimants who are detained (priority cases). Port-of-entry claimants are covered immediately. k Four to six weeks if claimed at port of entry. Otherwise, four to six months. 1 Port-of-entry claimants are covered upon arrival. May take one to ten weeks for inland claimants depending on completeness of their application.

Table 11.5 Health care services covered for citizens according to country Health care system component

Canada3

Australia

Netherlands

New Zealand

Sweden

United Kingdom

Hospitalization Physician services Other professional services Preventative care Home care Long-term care Dental care Pharmacare (out of hospital)

Y Y V V V V V V

Y Y YG9 I I I X PG U

Y Y Y Y k G' Y° Y Yq Yv

Y

yc

PA I Gr

PG b PG f PG 1 Gn Gp Y PG S

QW

pXQV

a

pdQe

PG h

Qm

Y YG' Y A A HG* G2

Refer to Table 11.2. Those under twenty are exempt from fees, and medical expenses are capped. c Includes medication, dental treatment, and optical services within hospital. Those with low income are entitled to travel assistance to and from hospital. d Most health care provided free of charge with the exception of primary health care, for which a fee-for-service exists. e Children under six, those with a 'community services card' (low income), those with a 'high-use health card' (more than twelve visits per years to a GP for a particular condition), and families who require more than twenty subsidized prescriptions per year are exempted from service fees. f Those under twenty are exempt from fees, and medical expenses are capped. 9 Ambulance services covered for those on government pension and those with low income. h Children under six, those with a 'community services card' (low income), those with a 'high-use health card' (more than twelve visits per years to a GP for a particular condition), and families who require more than twenty subsidized prescriptions per year are exempted from service fees. ' Those under twenty are exempt from fees, and medical expenses are capped. ' Those over sixty, children under sixteen, those under nineteen who are in full-time education, those on income support, and those with specific ophthalmologic conditions (or a family history of these conditions) are entitled to optical care. Hearing aids and their maintenance are also covered. b

Table 11.5 (Concluded) k

Covers genetic testing for hereditary conditions and thrombosis preventative services. ' Hepatitis B testing in pregnant women, PKU testing of neonates, and immunization of children. m Children are covered for immunizations. Children, those over sixty-five, and those with specific chronic medical conditions are covered for influenza vaccination. n Children under 'school age' are provided with vaccinations, health checks, consultations, and certain types of treatment free of charge at children's clinics. Health education provided in schools. Prenatal care covered. 0 Aids, medical devices, maternity care (including domestic services), and home dialysis services are also covered. p Elderly and disabled are provided with medical service, nursing, and technical aids. q Includes preventative maintenance. r Children are eligible for free dental care until their eighteenth birthday. s Those twenty and under are covered. ' Children under nineteen, pregnant women (up to twelve months post-birth), those on low-income support, disabled persons, and those receiving working-family tax credits are entitled to free dental services. u Costs for those with concession cards (students, elderly, and disabled) and for families are capped. v With a limit to reimbursement. w Children under six, those with a "community services card" (low income), those with a "high-use health card" (more than twelve visits per years to a GP for a particular condition), and families who require more than twenty subsidized prescriptions per year are exempted from service fees. x Total annual costs are capped. y Insulin is free for children under eighteen. 2 Children under sixteen, or under nineteen if in full-time education, and those sixty and over are covered. Pregnant women with exemption certificates, and individuals on income support are entitled to prescriptions.

Table 11.6 Health services coverage (refer to Table 11.5) as per immigration status: According to country Categories of immigration status

Canada3

Australia

Netherlands

Permanent residents People with employment authorization Students Refugees Refugee claimants / asylum seekers

Y V V Y

YbB Y X Y YnG°

a

ym

New Zealand

Sweden

United Kingdom

Yc

Yd

Ye

1 X Y Y

h Z9G

X X Y

YfB YB Yk YB YB

XGJ Y1 Y

PpGq

Refer to Table 11.3. Australia has reciprocal health care arrangements with some countries and residents from these countries are covered for basic health care services in Australia. c The Netherlands has reciprocal health care arrangements with some countries and residents of these countries are also covered for basic health care services in the Netherlands. d New Zealand has reciprocal health care arrangements with some countries and residents of these countries are also covered for basic health care services in New Zealand. e Sweden has reciprocal health care arrangements with some countries and residents of these countries are also covered for basic health care services in Sweden. f Nationals of the European Economic Areas are variably covered according to country of origin, and nationals of countries with which the United Kingdom has a reciprocal health care agreement are also covered. 9 Anyone suffering from an accidental injury is covered. h Teachers funded by the Ministry of Education's Foreign Language Teaching Assistantship Scheme are eligible for all services. Children of eligible persons have access to all services. Those with a valid temporary residence permit for two or more years are entitled to access public health services. ' Includes unpaid workers in certain voluntary organizations, and students and trainees who are required to work for twelve weeks during their first year. ' Students funded by the Ministry of Foreign Affairs' Official Development Assistance Programme, their partners, and their children are eligible for all health services. b

Table 11.6 (Concluded) k 1 m n 0

p q

Short-term (less than six months) students who are from countries that do not have a health care agreement with the United Kingdom are not entitled to coverage. Including multidisciplinary trauma counselling. Under the Interim Federal Health Program (IFHP). Must hold a valid visa with work rights in order to be covered. Claimants without visas are given financial assistance to cover health care and counselling services under the Asylum Seeker Assistance Scheme (ASA). Medical, dental, and specialist treatment is available for unauthorized immigrants in detention centres as medically required. Adult asylum seekers qualify for subsidized emergency medical and dental care; prescription medications and costs are capped. Asylum-seeking children are entitled to same health care as Swedish children. Maternity care, preventative child and antenatal care, and dental care for children under eighteen are free.

Table 11.7 Time period between arrival and eligibility for health services coverage as per immigration status: According to country Categories of immigration status

Canada3

Australia

Netherlands

New Zealand

Sweden

United Kingdom

Permanent residents People with employment authorization Students Refugees Refugee claimants / asylum seekers

V V V V V

2 years 2 years N J 6 months6

I I N J U

I I

3 mos. N N J U

lb U U Vd V9

a

N Vc Vf

Refer to Table 11.4. A person living in the United Kingdom for a settled purpose for more than six months is covered. c Length of time for obtaining the Community Services Card, which entitles refugees to coverage, may vary. d There is a registration process for determining free entitlement to services and this may delay access to coverage. For exemption from standard charges, there is also an application process, which is available only in English, is lengthy, and requires renewal after six months. e It takes six months for eligible claimants to be given financial assistance under the Asylum Seeker Assistance Scheme (ASA). There is immediate coverage for those in detention centres and for those with valid visas with work rights. ' Length of time for obtaining the Community Services Card, which entitles asylum seekers to coverage, may vary. Eligible only once they have lodged a claim and are awaiting a hearing. 9 There is a registration process for determining free entitlement to services and this may delay access to coverage. For exemption from standard charges, there is also an application process, which is available only in English, is lengthy, and requires renewal after six months. b

380 Anita J. Gagnon NOTE I would like to thank Claude-Catherine Lemoine, Carla Pascoe, and Robert Sewell for their diligence and perseverance in searching for, and collating, much of the information needed to produce the tables included in this paper.

REFERENCES

Literature Cited The affiliation of Multicultural Societies and Service Agencies (AMSSA). 2002. Available at www.amssa.org/ (March 2002). Beiser, M., and I. Hyman. 1997. Refugees' time perspective and mental health. American Journal of Psychiatry 154 (7): 996-1002. Beiser, M., R. Dion, A. Gotowiec, I. Hyman, and N. Vu. 1995. Immigrant and refugee children in Canada. Canadian Journal of Psychiatry / Revue Canadienne de Psychiatric 40(2): 67-72. Beiser, M., and R.G. Edwards. 1994. Mental health of immigrants and refugees. New Directions for Mental Health Services 61: 73-86. Beiser, M., and J.A. Fleming. 1986. Measuring psychiatric disorder among Southeast Asian refugees. Psychological Medicine 16(3): 627-39. Bowen, S. 2000. Language barriers in access to health care. Ottawa: Health Canada. Canadian Council for Refugees. 1998. Best settlement practices: settlement services for refugees and immigrants in Canada. Montreal: Canadian Council for Refugees. Canadian Heritage Multiculturalism. 2000. Annual report on the operation of the Canadian Multiculturalism Act, 1999-2000. Ottawa: Minister of Public Works and Government Services Canada. Chen, J., E. Ng, and R. Wilkins. 1996. The health of Canada's immigrants in 1994-95. Health Reports, 7(4): 33-45. Citizenship and Immigration Canada. 2000. Facts and figures 1999. Ottawa: Minister of Public Works and Government Services Canada. - 2001a. Facts and figures 2000: Immigration overview. Ottawa: Minister of Public Works and Government > Services Canada. - 2001b. Facts and Figures 2000: Statistical overview of the temporary resident and refugee claimant population. Ottawa: Minister of Public Works and Government Services Canada. Citizenship and Immigration Canada. 2002. Interim federal Health (IFH) Program. Available at www.cic.gc.ca/manuals%2Dguides/english/ om%2Dweb/1998/ip/ip98%2D16.html (March)

Responsiveness of the System towards Newcomers 381 Davies, M., and M. Yoshida. 1981. A model for cultural assessment of the new immigrant. Canadian Nurse 77(3): 22-3. Dillmann, E., R. Pablo, and A. Wilson. 1993. Patterns of health problems observed among newly arrived refugees to Canada, in R. Masi, L. Mensah, and K.A. McLeod, ed., health and cultures: Exploring the relationships, 2nd edition. Oakville, Ont.: Mosaic Press. Dunn, J., and I. Dyck. 1998. Social determinants of health in Canada's immigrant population: Results from the National Population Health Survey. Ottawa: Statistics Canada. ECRE Task Force on Integration. 2002. Good practice guides on the integration of refugees in the European Union. Available at www.refugeenet.org/ documents/volg.php3?ID=146 (March). EU Networks on Integration of Refugees. 2002. EU Networks on reception, integration and voluntary repatriation project. Available at www.refugeenet.org/ about/index.html (March). Gagnon, A.J., L. Merry, and C. Robinson. 2002. A systematic review of refugee women's reproductive health. ReFuge 21(6). Gravel, S., and A. Battaglini. 2000. Culture, sante et ethnicite: vers une sante publique pluraliste. Montreal: Regie regionale de la sante et des services sociaux. Gushulak, B.D., and D.W. MacPherson. 2000. Health issues associated with the smuggling and trafficking of migrants. Journal of Immigrant Health 2(2): 67-78. Hamilton, J. 1996. Multicultural health care requires adjustments by doctors and patients. CMAJ 155(5): 585-7. Huff, R.M., and M.V. Kline. 1999. Promoting health in multicultural populations: A handbook for practitioners. New York: Sage Publications. Hyman, I. 2001. Health policy working paper series. Ottawa: Health Canada. International Organization for Migration. 2001. Migrant trafficking and human smuggling in Europe: A review of evidence. Europe: Author. International Rehabilitation Council for Torture Victims. 2002. About the IRCT. Available at www.irct.org/about_the_irct_0.htm (March). Kent, H. 2000. BC's Chinese migrants a healthy lot, MDs find. CMAJ 162:256. Kirmayer, L.J. 2002. Culture and mental health. Available at www.mcgill.ca/ psychiatry/transcultural/cmhru.html (March). Lorion, J., and C. Dion. 1999. Front line health and social services ...for a human mosaic. Annual report. Montreal: CLSC Cote-des-Neiges. Maritime Centre of Excellence for Women's Health. 2000. Keeping Canadian values in health care symposium. Moving Towards Women's Health 9. Metropolis Canada. 2002. The Metropolis project-An international forum for research and policy on migration, diversity and changing cities. Available at http://canada.metropolis.net/frameset_e.html (March).

382 Anita J. Gagnon Minister of Supply and Services Canada.1994. Strategies for population health: Investing in the health of Canadians. Ottawa: Health Canada. Ministere de la Sante et des Services sociaux, Ministere de la Famille et de 1'Enfance. 2002. Avec les families immigrantes: guide d'intervention Available at ftp://206.167.52.6/acrobat/f/documentation/1998/98-265.pdf (March). Mosaic. 2002. Family programs and settlement services. Available at www. mosaicbc.com. (March). Multilingual Health Education Net. 2002. Multilingual Health Education Net: A non-profit alliance of Canadian health agencies. Available at www.multilingualhealth-education.net (March). New South Wales. 2002. Multicultural health services and programs in NSW. Available at www.internal.health.nsw.gov.au/health-public-affairs/mhcs (March). New Zealand Ministry of Health. 2001. Refugee health care - A handbook for health professionals. Auckland, N.Z.: Folio Communications. - 2002. 'About the New Zealand health and disability system. Available at www.moh.govt.nz/moh.nsf (March). Ontario Council of Agencies Serving Immigrants. 2002. About OCASI. Available at www.ocasi.org (March). Organisation for Economic Co-operation and Development. 2002. OECD health data 2001. Available at www.oecd.org/EN/statistics/0,,EN-statistics-12nodirectorate-no-no--12,00.html (March). Palmer, D.L. 1996. Determinants of Canadian attitudes toward immigration: More than just racism? Canadian Journal of Behavioural Science 28(3): 18092. Parakulam, G. 2000. Health status of Canadian born and foreign born residents. Canadian Journal of Public Health 83(4): 311-14. Pomerleau, J. 1997. The relationship between place of birth and some health characteristics in Ontario. Canadian Journal of Public Health 88(5): 337-45. Regie regionale de la sante et des services sociaux de Montreal-Centre. 1996. Organisation des services aux clienteles refugiees, revendicatrices de statut, migrantes et immigrantes pour les CLSC de la region de Montreal-Centre. Montreal: Regie regionale de la sante et des service sociaux de Montreal-Centre. - 2002. Inter-regional interpreters bank 2000-2001. Available at www.santemontreal.qc.ca (March). Riverdale Immigrant Women's Centre. 2002. Who we are... Available at www.a4s.org/RIWC/ (March). Romanow, R.J. 2002. Shape the future of health care. Interim report. Canada: Commission on the Future of Health Care in Canada. Rumbaut, R. 1996. Unravelling a public health enigma: Why do immigrants

Responsiveness of the System towards Newcomers 383 experience superior perinatal health outcomes. Research in the Sociology of Health Care 13B: 337-91. Simms, G.P. 1996. Aspects of Women's health from a minority/diversity perspective. Ottawa: Health Canada. Simon, R.J., and J.P. Lynch. 1999. A comparative assessment of public opinion toward immigrants and immigration policies. International Migration Review 33(2): 455-67. Statistics Canada. 1996. Immigration and citizenship -1996 census. Ottawa: Statistics Canada. - 2002. Health indicators: Proportion of immigrants. Available at www.stat can.ca/english/freepub/82-221-XIE/tables/htmltables/c4t05.htm (March). Stephenson, P.H. 1995. Vietnamese refugees in Victoria, B.C.: An overview of immigrant and refugee health care in a medium-sized Canadian urban centre. Social Science & Medicine 40(12): 1631^12. Sundquist, J., L. Bayard-Burfield, L.M. Johansson, and S.E. Johansson. 2000. Impact of ethnicity, violence and acculturation on displaced migrants: Psychological distress and psychosomatic complaints among refugees in Sweden. Journal of Nervous & Mental Disease 188(6): 357-65. La Table de concertation des organismes au service des personnes refugiees et immigrantes (TCRI). 2001. Rapport d' activites 2000/2001. Montreal: TCRI. Thonneau, P., J. Gratton, and G. Desrosiers. 1990. Health profile of applicants for refugee status (admitted into Quebec between August 1985 and April 1986). Canadian Journal of Public Health 81(3): 182-6. United Nations High Commissioner for Refugees. 1997. Refugees and others of concern to UNHCR - 1997 statistical overview. Available at www.unhcr.ch (March). - 2002. Who is a refugee? Available at www.unhcr.ch (March). U.S. Department of Health and Human Services, OPHS Office of Minority Health. 2001. National standards for culturally and linguistically appropriate services in Health Care. Washington, D.C.: IQ Solutions. Weeks, J.R. 2000. Infant mortality among ethnic immigrant groups. Social Science & Medicine, 33(3): 327-34. World Health Organization. 1986. Ottawa charter on health promotion. Copenhagen: World Health Organization Regional Office for Europe. Available at the Canadian Public Health Association. Personal Communications P. Dongier, physician, CLSC Cote-des-Neiges, Montreal, Quebec. 2002. Request for information on providing services to refugees and asylum seekers. Telephone conversation.

384 Anita J. Gagnon Etobicoke Community Legal Services Toronto. 2002. Request for information on delays in refugees accessing health care coverage (Toronto). Internet communication. B.D. Gushulak, medical director, Citizenship and Immigration Canada. 2002. Request for information on the Interim Federal Health Program. Internet communication. R. Hogue, Canadian Council of Refugees, CCR LISTSERV . 2002. IFHP paper. Internet Communication. D. Isaacs, refugee outreach worker, Juan Moreno, Montreal, Quebec. 2002. Request for information on delays in refugees accessing health care coverage (Montreal). Telephone conversation. E.F. Khaki. 2002. Request for information on refugees making claims. Internet communication. A. Li, Chinese Canadian National Council. 2002. Request for information on health care access. Internet communication. Sources for Tables 11.2-11.7 Canadian Comparisons Alberta Ministry of Health and Wellness, Alberta. Information request on when eligibility for coverage commences. Internet communication, April 2002. - On-line with Alberta Health and Wellness. Available at www.health.gov.ab.ca (April 2002) E. Skoropad, Ministry of Health and Wellness, Alberta, Request for information on provincial health insurance. Internet communication, April 2002. British Columbia Ministry of Health Services, British Columbia. 'Health Services.' Available at www.gov.bc.ca/healthservices (April 2002). M. Van der Meij, Victoria Immigrant and Refugee Centre Society. Information request on access to health care for immigrant populations. Internet communication, April 2002. Manitoba Government of Manitoba. Manitoba Health. Available from www.gov.mb.ca/ health/index.html April 2002. U. Wendt, Ministry of Health, Government of Manitoba. Request for information on public health for people not born in Canada. Internet communication, (April 2002).

Responsiveness of the System towards Newcomers 385 Newfoundland Department of Health and Community Services, Newfoundland and Labrador. Health and Community Services. Available at www.gov.nf.ca/deptnew. htm (March 2002). K. Dyer. Ministry of Health, Government of Newfoundland. Access to health care for immigrant populations in Newfoundland. Internet communication, (March 2002). New Brunswick Department of Health and Wellness, New Brunswick. Health and Wellness. Available at www.gnb.ca/0051/index-e.asp (April 2002). P. MacDonald, Ministry of Health, Government of New Brunswick . Request for information on health services for people not born in Canada. Internet communication, March 2002. Northwest Territories Department of Health and Social Services, Northwest Territories. Health and Social Services. Available at www.hlthss.gov.nt.ca (March 2002). Nova Scotia Department of Health, Nova Scotia. Department of Health. Available at www.gov.ns.ca/health (March 2002). Nunavut Department of Health and Social Services, Nunavut. Department of Health and Social Services. Available at www.gov.nu.ca/hss.htm (March 2002). Prince Edward Island Department of Health and Social Services, Prince Edward Island. Health and Social Services. Available at www.gov.pe.ca/hss (March 2002). B. MacLure, Department of Health and Social Services, Government of Prince Edward Island. Information on provincial health insurance. Internet communication, 2002. Quebec Portail Quebec. Regie de 1'assurance maladie du Quebec. Available at www.ramq.gouv.qc.ca (March 2002). Regie de 1'assurance maladie du Quebec. Ce a quoi votre carte vous donne acces: le regime d'assurance maladie. Available at www.ramq.gouv.qc.ca/ cit/acces/regime.htm (February 2002).

386 Anita J. Gagnon S. Reichhold, Director, La Table de concertation des organismes de Montreal pour les refugies. Personal communication, 2 April 2002. Saskatchewan Saskatchewan Health. March 2002. Health. Available at www.health.gov.sk.ca (March 2002). C. Gordon, Ministry of Health, Government of Saskatchewan. Health benefits eligibility. Internet communication, April 2002. Yukon Department of Health and Social Services, Yukon. Yukon Health and Social Services. Available at www.hss.gov.yk.ca (March 2002). International Comparisons Australia Australian Red Cross. Available at www.redcross.org.au Department of Immigration and Multicultural and Indigenous Affairs, Australia. Assistance for asylum seekers in Australia. Available at www.immi.gov.au (March 2002). - Seeking asylum within Australia. Available at www.immi.gov.au (March 2002). - Immigration detention. Available at www.imini.gov.au (March 2002). - Overseas students in Australia. Available at www.immi.gov.au (March 2002). - Temporary protection visas. Available at www.immi.gov.au (March 2002). Health Insurance Commission, Australia. Visitors travelling to Australia. Available at www.hic.gov.au/yourhealth/services_for_travellers/vtta.htm (March 2002). - About Medicare. Available at www.hic.gov.au/yourhealth/our_services/ am.htm (March 2002). - About the Pharmaceutical Benefits Scheme (PBS). Available at www.hic.gov.au/yourhealth/our_services/apbs.htm (March 2002). Kelsall, L., C. Kiro, U. Antipas, W. Cook, J. Calveley, A. Mortensen, K. Boyle, S. Wells, M. Gardner, and R. Walker. Health needs assessment 2001 (Australia: Auckland District Health Board, 2001). Refugee Resettlement Advisory Council. Help for people settling in Australia. Available at www.immi.gov.au/tis/index.htm (March 2002). UNHCR. Resettlement handbook: Chapter country - Australia. Available at www.unhcr.org (March 2002).

Responsiveness of the System towards Newcomers 387 Canada Citizenship and Immigration Canada. Immigrant Services: Health and income security. Available at www.cic.gc.ca/english/newcomer/fact_04e.html (February 2002). - Interim Federal Health Program. Available at www.fasadmin.com/english/ partl.html (February 2002). Netherlands European Council on Refugees and Exiles (ECRE). The Netherlands. Available at www.ecre.org/drc/netherlands.shtml (February 2002). Ministry of Health, Welfare and Sport, Health insurance in the Netherlands. IE (The Hague: International Publication Series, 2001). UNHCR. Resettlement handbook: Chapter country - Netherlands. Available at www.unhcr.org (March 2002). New Zealand H. Kizito, Refugee health care - A handbook for health professionals (New Zealand: Folio Communications, 2001). New Zealand Ministry of Health. An overview of the health and disability sector in New Zealand Available at www.moh (March 2002). - About the New Zealand health and disability system. Available at www.moh.govt.nz/moh.nsf (March 2002). UNHCR. Resettlement handbook: Chapter country - New Zealand. Available at www.unhcr.org (March 2002). Sweden Centre for Evaluation of Social Services, National Board of Health and Welfare. Different conditions - different health: A study of immigrants from Chile, Poland, Iran and Turkey. Available at www.sos.se/sos/publ/refereng/ 0003003E.htm (March 2002). Department of Health Policy and Services. Health care systems in transition: Sweden. Available at www.euro.int/documents/e72481.pdf (March 2002). European Council on Refugees and Exiles (ECRE). Sweden. Available at www.ecre.org/drc/sweden.shtml (March 2002). Migration Board, Sweden. The costs of refugee reception. Available at www.migrationsvekert.se (March 2002). - Refugee reception. Available at www.migrationsvekert.se (March 2002). - Swedish residence permits for EU/EEA Citizens. Available at www.migrationsvekert.se (March 2002). Swedish Institute. The health care system in Sweden. Available at www.se.si/ docs/infos weden/engelska/fs63.pdf (March 2002).

388 Anita J. Gagnon - Immigrants in Sweden. Available at www.se.si/docs/infosweden/engelska/ fs63.pdf (March 2002). - Social insurance in Sweden. Available at www.se.si/docs/infosweden/ engelska/fs5z.pdf (March 2002). UNHCR. Resettlement handbook: Chapter country - Sweden. Available at www.unhcr.org (March 2002). United Kingdom British Medical Association. Access to health care for asylum seekers. Available at www.asylumsupport.info (March 2002). Department of Health Publications. NHS booklet HC11 - Help with health costs. Available at www.doh.gov.uk (March 2002). European Observatory on Health Care Systems. Health care systems in transition (United Kingdom: Nation Health Science, 1999). Home Office. Full and equal citizens (United Kingdom: IND Communications Team and Home Office Communication Directorate, 2000). Immigration and Nationality Directorate. Asylum in the U.K.: Once you are here. Available at www.ind.homeoffice.gov.uk (March 2002). National Health Service. The NHS explained. Available at www.nhs.uk (March 2002). - National Health Service hospital charges to overseas visitors. Available at www.doh.gov.uk/overseasvisitors/patientguide.htm (March 2002). Refugee Council. Support and entitlements. Available at www.refugeecouncil. org.uk (March 2002).