Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National Health Informatics Conference (HIC 2012) [1 ed.] 9781614990789, 9781614990772

Citation preview

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

HEALTH INFORMATICS: BUILDING A HEALTHCARE FUTURE THROUGH TRUSTED INFORMATION

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Studies in Health Technology and Informatics This book series was started in 1990 to promote research conducted under the auspices of the EC programmes’ Advanced Informatics in Medicine (AIM) and Biomedical and Health Research (BHR) bioengineering branch. A driving aspect of international health informatics is that telecommunication technology, rehabilitative technology, intelligent home technology and many other components are moving together and form one integrated world of information and communication media. The series has been accepted by MEDLINE/PubMed, SciVerse Scopus, EMCare, Book Citation Index – Science and Thomson Reuters’ Conference Proceedings Citation Index. Series Editors: Dr. O. Bodenreider, Dr. J.P. Christensen, Prof. G. de Moor, Prof. A. Famili, Dr. U. Fors, Prof. A. Hasman, Prof. E.J.S. Hovenga, Prof. L. Hunter, Dr. I. Iakovidis, Dr. Z. Kolitsi, Mr. O. Le Dour, Dr. A. Lymberis, Prof. J. Mantas, Prof. M.A. Musen, Prof. P.F. Niederer, Prof. A. Pedotti, Prof. O. Rienhoff, Prof. F.H. Roger France, Dr. N. Rossing, Prof. N. Saranummi, Dr. E.R. Siegel, Prof. T. Solomonides and Dr. P. Wilson

Volume 178

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Recently published in this series Vol. 177. B. Blobel, P. Pharow and F. Sousa (Eds.), pHealth 2012 – Proceedings of the 9th International Conference on Wearable Micro and Nano Technologies for Personalized Health Vol. 176. T. Kotwicki and T.B. Grivas (Eds.), Research into Spinal Deformities 8 Vol. 175. S. Gesing, T. Glatard, J. Krüger, S.D. Olabarriaga, T. Solomonides, J.C. Silverstein, J. Montagnat, A. Gaignard and D. Krefting (Eds.), HealthGrid Applications and Technologies Meet Science Gateways for Life Sciences Vol. 174. B. Blobel, R. Engelbrecht and M.A. Shifrin – Large Scale Projects in eHealth – Partnership in Modernization – Proceedings of the EFMI Special Topic Conference 18–20 April 2012 Moscow, Russia Vol. 173. J.D. Westwood, S.W. Westwood, L. Felländer-Tsai, R.S. Haluck, R.A. Robb, S. Senger and K.G. Vosburgh (Eds.), Medicine Meets Virtual Reality 19 – NextMed Vol. 171. P.B. Cerrito, Data Mining to Determine Risk in Medical Decisions Vol. 170. G.J.E. De Moor (Ed.), Transatlantic Cooperation Surrounding Health Related Information and Communication Technology Vol. 169. A. Moen, S.K. Andersen, J. Aarts and P. Hurlen (Eds.), User Centred Networked Health Care – Proceedings of MIE 2011 Vol. 168. D.P. Hansen, A.J. Maeder and L.K. Schaper (Eds.), Health Informatics: The Transformative Power of Innovation – Selected Papers from the 19th Australian National Health Informatics Conference (HIC 2011)

ISSN 0926-9630 (print) ISSN 1879-8365 (online)

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Health Informatics: Building a Healthcare Future Through Trusted Information Selected Papers from the 20th Australian National Health Informatics Conference (HIC 2012)

Edited by

Anthony J. Maeder University of Western Sydney, School of Computing, Engineering and Mathematics, Sydney, Australia

and

Fernando J. Martin-Sanchez

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

University of Melbourne, Faculty of Medicine, Dentistry and Health Sciences, Melbourne, Australia

Amsterdam • Berlin • Tokyo • Washington, DC

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

© 2012 The authors and IOS Press. All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without prior written permission from the publisher. ISBN 978-1-61499-077-2 (print) ISBN 978-1-61499-078-9 (online) Library of Congress Control Number: 2012942559 Publisher IOS Press BV Nieuwe Hemweg 6B 1013 BG Amsterdam Netherlands fax: +31 20 687 0019 e-mail: [email protected]

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Distributor in the USA and Canada IOS Press, Inc. 4502 Rachael Manor Drive Fairfax, VA 22032 USA fax: +1 703 323 3668 e-mail: [email protected]

LEGAL NOTICE The publisher is not responsible for the use which might be made of the following information. PRINTED IN THE NETHERLANDS

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Health Informatics: Building a Healthcare Future Through Trusted Information A.J. Maeder and F.J. Martin-Sanchez (Eds.) IOS Press, 2012 © 2012 The authors and IOS Press. All rights reserved.

v

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Preface A record of twenty years of productive scientific output is a substantial achievement in any field of ICT, and especially in the youthful domain of health informatics. The Health Informatics Society of Australia, with the active support and involvement of the Australasian College of Health Informatics, is justifiably proud to have sustained the annual Australian National Health Informatics Conference (HIC) series of meetings through to this stage of maturity. As the largest national event of this type in Australia and New Zealand, with a dedicated scientific stream on health informatics, HIC provides a valued platform for academic and research contributions and interchange. It also consolidates the Australasian profile of strong and current research contributions in health informatics, marking our place on the world stage. The HIC 2012 theme of “Health Informatics: Building a Healthcare Future Through Trusted Information” emphasises the importance of assuring the integrity and security of health data and communications. At a time when large scale health information flows are being initiated, through both the inauguration of the personally controlled electronic health record (PCEHR) and the National Broadband Network rollout, these issues need to remain prominent in the minds of designers and developers as well as the priorities of administrators and executives. They are naturally a major concern for consumers, and an immutable responsibility for vendors. A more trusted environment for managing and using health information will help to consolidate and accelerate the use of health informatics solutions as change mechanisms to drive the establishment and adoption of new models of care and new technology-oriented healthcare processes. Papers in this volume will be found to both align with this theme, and to add colour to the landscape within which it resides by addressing peripheral but related topics. As always, it is a strength of HIC that a wide diversity of work is presented, and that a set of papers has been collected here that ranges from deeply theoretical to intensely practical. The careful reader will be rewarded with exposure to much diversity, and many elements of contemporary health informatics research endeavours. The double blind peer review process established for HIC 2011 in a previous volume has been continued and augmented. All papers were reviewed by either 3 or 4 experts in the field of health informatics, selected as prominent academic and industry specialists. The assistance of the Australasian College of Health Informatics in supporting this processes through the voluntary efforts of a number of their Fellows is gratefully acknowledged, as is the similar contribution made by many senior members of the Health Informatics Society of Australia. This phase of reviewing resulted in the provisional acceptance of 38 papers from a much expanded submission field of 61 (compared with 39 for HIC 2011). The Scientific Program Committee, composed of several previous Scientific Program Chairs of past HIC conferences, then undertook a validation process for all such papers that were resubmitted in amended form, to ensure that reviewers’ recommendations were appropriately addressed or rebutted. This resulted in 35 papers finally being included for publication in this book. Anthony J. Maeder Fernando J. Martin-Sanchez Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

This page intentionally left blank

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

vii

Acknowledgements

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

The Editors and Scientific Program Committee wish to thank the following people for their efforts in reviewing the papers submitted for HIC 2012. Prof. Anthony Maeder, University of Western Sydney Prof. Fernando J. Martin-Sanchez, University of Melbourne Dr. Jim Basilakis, University of Western Sydney Vicki Bennett*, Australian Institute of Health and Welfare Dr. Jen Bichel-Findlay, University of New South Wales Heidi Bjering, University of Western Sydney Neville Board, Australian Commission on Safety and Quality in Health Care Dr. Andy Bond*, National E-Health Transition Authority Kerryn Butler-Henderson, Curtin Health Innovation Research Institute, Curtin University Assoc. Prof. Joanne Callen, University of New South Wales Paul Clarke, JamPac Health Informatics Management Consulting Prof. Enrico Coiera*, Centre for Health Informatics, University of New South Wales Prof. Peter Croll*, Southern Cross University Bernard Crowe, Bernard Crowe & Associates, IHE Australia Dr. Joanne Curry, University of Western Sydney Dr. Karen Day, University of Auckland Cathy Doran, Justice Health & Forensic Mental Health Network Joan Edgecumbe, eHealth Education Dr. Juanita Fernando, Monash University Sebastian Garde, Ocean Informatics Dr. Andrew Georgiou, Centre for Health Systems and Safety Research, University of New South Wales Janette Gogler, Austin Health Assoc. Prof. Heather Grain, eHealth Education Pty Ltd Prof. Andrew Grant, University of Sherbrooke Dr. Kathleen Gray, University of Melbourne Dr. Isla Hains, University of New South Wales Dr. Leif Hanlen, National ICT Australia Dr. David Hansen*, The Australian e-Health Research Centre, CSIRO ICT Centre Jon Hilton, Precedence Health Care Prof. Evelyn Hovenga*, eHealth Education Pty Ltd Prof. Sabine Koch, Karolinska Institute Dr. Michael J. Lawley, The Australian e-Health Research Centre, CSIRO ICT Centre Professor Michael Legg*, Michael Legg & Associates and University of Wollongong Dr. Hugo Leroux, The Australian e-Health Research Centre, CSIRO ICT Centre Dr. Farah Magrabi, University of New South Wales Prof. Jose Luis Oliveira, University Of Aveiro Dr. Paula Otero, Hospital Italiano Dr. Chris Paton, University of Auckland

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

viii

Prof. Jon Patrick*, University of Sydney Dr. Christopher Pearce, Melbourne East General Practice Network Derek Ritz, ecGroup Dr. Vitali Sintchenko*, Sydney Medical School, University of Sydney Assoc. Prof. Jeffrey Soar, University of Southern Queensland Prof. Siaw Teng-Liaw*, University of New South Wales Donna Truran, University of Wollongong Assoc. Prof. Paul Turner, University of Tasmania Adj. Assoc. Prof. Klaus Veil, University of Western Sydney Dr. Amol Wagholikar, The Australian e-Health Research Centre, CSIRO ICT Centre Prof. Jim Warren*, University of Auckland Dr. Patricia Williams, Edith Cowan University Dr. Ping Yu*, Faculty of Informatics, University of Wollongong

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

*Scientific Program Committee Member

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

ix

Contents Preface Anthony J. Maeder and Fernando J. Martin-Sanchez Acknowledgements Understanding Doctors’ Perceptions of Their Prescribing Competency and the Value They Ascribe to an Electronic Prescribing System Melissa T. Baysari, Johanna I. Westbrook and Richard O. Day Privacy with Emergency Medical Information Used in First Response Peter R. Croll and Kimberly M. Ambrosoli

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

The Cradle Coast Personally Controlled Electronic Health Record Evaluation Research Plan Elizabeth Cummings, Colleen Cheek, Winifred van der Ploeg, Peter Orpin, Heidi Behrens, Sharon Condon, Linda Jaffray, Isabelle Ellis, Barbara Ringeisen Arnold, Robyn Brogan and Timothy Skinner

v vii

1 7

14

An Australian Roadmap for ICT Research and Development for Ageing? Lessons from a European Union Initiative Elizabeth Cummings and Patrick Nixon

20

An Information Management System for Patients with Tuberculosis: Usability Assessment with End-Users Jonathan Darby, Jim Black, David Morrison and Kirsty Buising

26

We Are Not Educating the Future Clinical Health Professional Workforce Adequately for E-Health Competence: Findings of an Australian Study Ambica Dattakumar, Kathleen Gray, Kerryn-Butler Henderson, Anthony Maeder and Helen Chenery Influencing Factors for Adopting Personal Health Record (PHR) Karen Day and Yulong Gu Exploring the Role of Pathology Test Results in the Prediction of Remaining Days of Hospitalisation Blanca Gallego, Oscar Perez-Concha, Frank Lin and Enrico Coiera Exergames for the Elderly: Towards an Embedded Kinect-Based Clinical Test of Falls Risk Jaime A. Garcia, Karla Felix Navarro, Daniel Schoene, Stuart T. Smith and Yusuf Pisan Managing Collaboration Across Boundaries in Health Information Technology Projects Karin Garrety, Andrew Dalley, Ian McLoughlin, Rob Wilson and Ping Yu

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

33

39

45

51

58

x

An ICU Clinical Information System – Clinicians’ Expectations and Perceptions of Its Impact Isla M. Hains, Nerida Creswick, David Milliss, Michael Parr and Johanna I. Westbrook Use of an Electronic Drug Monitoring System for Ambulatory Patients with Chronic Disease: How Does It Impact on Nurses’ Time Spent Documenting Clinical Care? Antonia Hordern, Joanne Callen, Kathryn Gibson, Louise Robertson, Ling Li, Isla M. Hains and Johanna I. Westbrook Visualising Patient Flow Andrew Jensen, Justin Boyle and Sankalp Khanna Considerations of Electronic Medications Management Systems in Hospital Setting Philip Joyce Early Discharge and Its Effect on ED Length of Stay and Access Block Sankalp Khanna, Justin Boyle, Norm Good and James Lind The Effect of E-Health Contents on Health Science Students’ Attitude Toward the Efficiency of Health ICT in Care Provision Mary K. Lam, Krestina L. Amon, Melanie Nguyen, Andrew J. Campbell and Victoria Neville

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

A Qualitative Study of Australians’ Opinions About Personally Controlled Electronic Health Records Elin C. Lehnbom, Andrew J. McLachlan and Jo-Anne E. Brien

64

71

77

83 92

99

105

A Method for the Semantic Enrichment of Clinical Trial Data Hugo Leroux, Simon McBride, Laurent Lefort, Madonna Kemp and Simon Gibson

111

Electronic Health Information System Implementation Models – A Review Julia Logan

117

Evaluation of Bluetooth Low Power for Physiological Monitoring in a Home Based Cardiac Rehabilitation Program Timothy Martin, Hang Ding, Matthew D’Souza and Mohan Karunanithi

124

The Development of Online Learning Designs for Patients with Type 2 Diabetes Glenn Mason

130

Investigation of Decision Making Issues in the Use of Current Clinical Information Systems Pubudika K. Mawilmada, Susan E. Smith and Tony Sahama

136

Using Australian Medicines Terminology (AMT) and SNOMED CT-AU to Better Support Clinical Research Simon J. McBride, Michael J. Lawley, Hugo Leroux and Simon Gibson

144

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

xi

Classification of Pathology Reports for Cancer Registry Notifications Anthony Nguyen, Julie Moore, Guido Zuccon, Michael Lawley and Shoni Colquist

150

Multi-Layered System Design for Classifying Activities of Daily Living Saif Okour, Anthony Maeder and Jim Basilakis

157

EEG Data Compression to Monitor DoA in Telemedicine Mario E. Palendeng, Qing Zhang, Chaoyi Pang and Yan Li

163

Using Utilisation Data to Estimate Future Demand for Medical Internists: The Impact of Demographic Demand Driver in Thailand Supasit Pannarunothai and Pudtan Phanthunane Computational Recognition of SNOMED CT Codes from ED Case Notes Jon Patrick, Dung Nguyen, Tingxin Wang and Richard Paoloni

175

Evaluating Online Diagnostic Decision Support Tools for the Clinical Setting Marie Pryor, David White, Bronwyn Potter and Roger Traill

180

Align, Share Responsibility and Collaborate: Potential Considerations to Aid in E-Health Policy Development Nouran Ragaban, Karen Day and Martin Orr Understanding Unintended Consequences for EMR: A Literature Review Perdana Rahadhan, Simon K. Poon and Lesley Land

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

169

186 192

Physicians’ Satisfaction with Computerised Physician Order Entry (CPOE) at the National Guard Health Affairs: A Preliminary Study Basema Saddik and Manahil M. Al-Fridan

199

The ENSAT Registry: A Digital Repository Supporting Adrenal Cancer Research Anthony Stell and Richard Sinnott

207

What Do Radiology Incident Reports Reveal About In-Hospital Communication Processes and the Use of Health Information Technology? Michael J. Stewart, Andrew Georgiou, Antonia Hordern, Marion Dimigen and Johanna I. Westbrook The University of NSW Electronic Practice Based Research Network: Disease Registers, Data Quality and Utility J. Taggart, S.T. Liaw, S. Dennis, H. Yu, A. Rahimi, B. Jalaludin, and M. Harris Using the General Practice EMR for Improving Blood Pressure Medication Adherence Jim Warren, John Kennelly, Debra Warren, C. Raina Elley, Kuinileti Chang Wai, Marilyn Manukia, Jennifer Davy, Thusitha Mabotuwana and Elizabeth Robinson Electronic Referrals: What Matters to the Users Jim Warren, Yulong Gu, Karen Day, Sue White and Malcolm Pollock

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

213

219

228

235

xii

Patients’ Perceptions of Web Self-Service Applications in Primary Healthcare Xiaojun Zhang, Ping Yu, Jun Yan, Hongxiang Hu and Niraj Goureia The Impact of OCR Accuracy on Automated Cancer Classification of Pathology Reports Guido Zuccon, Anthony N. Nguyen, Anton Bergheim, Sandra Wickman and Narelle Grayson

242

250

257

Author Index

259

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Subject Index

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Health Informatics: Building a Healthcare Future Through Trusted Information A.J. Maeder and F.J. Martin-Sanchez (Eds.) IOS Press, 2012 © 2012 The authors and IOS Press. All rights reserved. doi:10.3233/978-1-61499-078-9-1

1

Understanding doctors’ perceptions of their prescribing competency and the value they ascribe to an electronic prescribing system

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Melissa T. BAYSARI a, b, Johanna I. WESTBROOK a and Richard O. DAY b, c a Australian Institute of Health Innovation, University of New South Wales b Department of Clinical Pharmacology & Toxicology, St Vincent’s Hospital c Faculty of Medicine, University of New South Wales

Abstract. Resistance to adoption has been identified as one of the major barriers to successful implementation of technological systems in hospitals. Acceptance of an electronic prescribing (e-prescribing) system is expected to occur if prescribers perceive a need for e-prescribing systems to reduce prescribing errors. We set out to examine doctors’ perceptions of their prescribing competency and to identify perceived advantages and disadvantages of using an e-prescribing system, with the objective of determining the value doctors ascribed to the e-prescribing system. This study was conducted at a teaching hospital in Sydney, Australia. Sixteen prescribers participated in a 20-minute semi-structured interview where they were asked to comment on prescribing errors (their own errors and errors they believed to be common) and advantages and disadvantages of the e-prescribing system. Prescribers held the view that they rarely made prescribing errors. Although users recognised advantages and disadvantages of using the e-prescribing system, most preferred paper to electronic prescribing. Prescribers most likely overestimated their prescribing competency and so failed to see the value of an e-prescribing system to reduce prescribing errors. E-prescribing system implementation is a challenging task for any hospital. These results suggest that keeping prescribers informed about their prescribing errors and the quality improvement benefits of eprescribing may lead to greater acceptance of and satisfaction with an eprescribing system. Keywords. electronic prescribing, prescribing errors, technology acceptance

Introduction Electronic-prescribing (e-prescribing) systems in hospitals automate aspects of the medication ordering process and ensure standardised, legible and complete orders [1]. Computerised alerts integrated into e-prescribing systems have the potential to reduce prescribing errors because they warn prescribers about possible risks such as allergies, inappropriate doses, and duplicated orders [2]. Despite these benefits, implementations of e-prescribing systems are often unsuccessful [3]. Resistance to adoption has been identified as one of the major barriers to successful implementation of e-prescribing systems in hospitals [4]. Acceptance of an innovation often begins with the recognition of a problem or need. If users do not believe a problem exists, they are likely to be disinterested in an innovation designed to solve the problem [5].

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

2

M.T. Baysari et al. / Understanding Doctors’ Perceptions of Their Prescribing Competency

Individual reactions to computing technology have been investigated from a variety of theoretical perspectives but the construct within each model shown to be the strongest and most significant predictor of intention to use a technological system is perceived usefulness [6]. This is the extent to which a user believes the application will increase his/her job performance within an organisational context [7]. It follows from this that users will adopt e-prescribing systems and view them positively if they recognise the limitations of paper based prescribing and see electronic systems as useful in addressing some of these limitations. We aimed to examine doctors’ perceptions of their prescribing competency and to identify perceived advantages and disadvantages of using a hospital e-prescribing system, with the objective of determining the value doctors ascribed to the e-prescribing system.

1. Method 1.1. Setting and Details of e-Prescribing System This study was conducted at a 320-bed teaching hospital in Sydney, Australia. At the time of the study, all wards were using the e-prescribing system, MedChart (www.isofthealth.com), except for the emergency department (ED) and the intensive care unit (ICU). MedChart is an electronic medication management system that links prescribing, pharmacy review, and drug administration. The system included basic decision support comprising a “Reference viewer” look-up tool that allowed prescribers to access relevant reference information (e.g. the Australian Medicines Handbook) by clicking on a tab at the top of the prescribing screen, and also a range of automatic alerts (i.e. allergy, therapeutic duplication, pregnancy, locally developed rule-based messages) delivered to prescribers immediately following the selection of a drug.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

1.2. Participants Sixteen prescribers from various specialties were opportunistically recruited (via direct approach, phone or email) to participate in interviews. Of these, three were residents, ten were registrars (at least three years in hospital practice) and three were consultants. Recruitment of prescribers for interviews continued until saturation of content was achieved (i.e. no new content was being elicited). 1.3. Procedure Prescribers participated in a 20-minute semi-structured interview. During the interview, doctors were asked to comment on prescribing errors (their own errors and errors they believed to be common) and factors they believed contribute to prescribing errors. They were also asked whether they preferred electronic or paper prescribing and to list advantages and disadvantages of the e-prescribing system. Three team members independently reviewed the interview transcripts to identify prescriber attitudes towards the e-prescribing system, liked and disliked features, and most importantly, prescriber views about their own prescribing errors (how common they are, what kinds they make) and prescribing errors in general. Ethics approval was

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

M.T. Baysari et al. / Understanding Doctors’ Perceptions of Their Prescribing Competency

3

obtained by the human research ethics committees of the hospital and the University of NSW.

2. Results Both junior and senior doctors held the view that they rarely made prescribing errors. They admitted that they made lots of errors while using the e-prescribing system, like clicking on the wrong frequency or administration time, but they usually noticed these while reviewing the order and corrected them. For example, a registrar (#10) said: “I often make errors of clicking what I want to write but I often, but I always change them. I notice them and change them…I don’t generally decide to give the wrong drug”. Prescribers reported rarely making clinical errors that they didn’t detect and correct (e.g. deciding to prescribe the wrong drug or the wrong dose). A consultant (#2) said: “Error? Never”. Doctors also had great difficulty estimating how often they made an error while prescribing. They explained that this was because they received very little feedback about errors, only those that led to patient harm. A registrar (#7) said: “Probably do make errors, how commonly, I don’t know”. Several doctors could not provide an example of an error they had made recently. A resident said (#3): “I’m not aware of mistakes that I've made recently because I haven't been told about them”. When asked to describe what types of prescription errors were common, most doctors described slips while using MedChart e.g. choosing items above and below the desired option on drop down lists. Factors frequently stated as contributing to prescribing errors included time pressure, a lack of knowledge, interruptions and distractions. Perceived advantages and disadvantages of using the electronic system were fairly consistent across all users (see Table 1). Although most doctors believed that the eprescribing system was easy to use once they’d had an opportunity to familiarise themselves with it, when asked if they preferred paper or electronic prescribing, few (3/16) reported a preference for electronic charts.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Table 1. Perceived advantages and disadvantages of using Medchart Advantage/disadvantage Remote access

Example “But its obviously all in one place and that's probably the best thing, like you can just look on the computer, even if the computer is slow, than to have to run around looking for the yellow chart”.

Ability to view patient history

“…you can see what’s been, you just don’t look at one chart, you see all the previous charts and you can also see what’s happened when people have been discharged, you don’t need to call up old notes”.

Drop down lists

“It means that you can’t get the scale of dosing wrong really because it usually comes up with a suggested scale, so there’s no capacity for adding or taking a zero for instance”.

Reference Viewer

“I like the fact that we can actually check before you prescribe because I find if I was just writing, scribbling something on a paper chart, I would just write it and then say that I need to check it, and sometimes you forget and then it gets given…whereas on Hatrix (MedChart) we usually on the spot just say lets look it up”.

Increased time

“I know some teams find the computer so slow that they actually use

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

4

M.T. Baysari et al. / Understanding Doctors’ Perceptions of Their Prescribing Competency

two computers side by side to check bloods and prescribe at the same time, which seems like a poor use of the infrastructure but I suppose there’s a testament to maybe the computers need to be faster”. Inflexibility

“I think its also very rigid in the way you select things which I suppose is frustrating and you have to think about it…I suppose its not actually difficult, its just frustrating to use because its not as flexible as paper where you can just write the drug and then you can write the dose and…”.

Poor usability

“It’s as if I put my arm around my head to touch my nose instead of just touching my nose directly, it just feels that way”.

Increased errors

“It’s just too fiddly, like you go down to scroll to put weekly and you easily put daily. Errors are easier to make, absolutely”.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

3. Discussion Doctors in this study most likely overestimated their prescribing competency, just as it has been shown that doctors overestimate the appropriate treatment they provide to patients [9-11]. Prescribers (both junior and senior) held the view that they rarely made prescribing errors, other than simple slips while using the e-prescribing system, but a recent audit of medication charts at the study site revealed that patients experienced on average 1.5 prescribing errors per admission [8]. It follows from a belief that errors are not made, that a system like MedChart is not valued for its role in error reduction. Perceived usefulness – the extent to which a user believes the system will increase his/her job performance – is a reliable predictor of intention to use a technological system [7]. The e–prescribing system is currently not perceived to be useful in targeting prescribing errors as doctors believed very few errors were being made. Although prescribers recognised advantages and disadvantages with using the eprescribing system, most preferred paper to electronic prescribing. This is consistent with other studies that have found that some clinicians do not regard e-prescribing as an improvement over written paper charts [e.g. 12]. The perceived advantages and disadvantages of e-prescribing use in this setting are also consistent with those previously reported in the literature [13, 14]. For example, remote access is often reported as the primary beneficial effect of e-prescribing and increased time to prescribe as the primary negative effect [13]. Satisfaction with electronic systems has been shown to be related to the efficiency with which users successfully perform tasks using the system. Research has shown that users place more emphasis on e-prescribing characteristics related to efficiency (e.g. speed) than characteristics related to quality (e.g. fewer medication errors) [15]. This might be because these ‘efficiency’ consequences of system implementation are more visible to prescribers than the ‘quality’ ones. Observability (the degree to which the results of an innovation are visible to users) has been identified as an important factor in facilitating adoption of new technology [5]. The prescribers in this setting were noticeably aware of MedChart’s impact on time to order medications (i.e. prescribing was seen to take longer) but unaware of MedChart’s capacity to reduce prescribing errors. The overall prescribing error rate at the hospital has more than halved following MedChart implementation [8] but some prescribers believed use of the e-prescribing system led to more errors than paper prescribing. This may partly be due to the fact that

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

M.T. Baysari et al. / Understanding Doctors’ Perceptions of Their Prescribing Competency

5

visibility of errors to prescribers is enhanced with e-prescribing systems, for example, when an alert fires. Also at the time of this study, prescribers had not been informed of the significant reduction in prescribing errors associated with system implementation. Prescribers reported receiving very little feedback about prescribing errors. Feedback has been identified as one of the key strategies to change physician behavior [16]. There is also some evidence to suggest that providing prescription-relevant feedback to doctors may lead to changes in prescribing practices [17]. With the shift to computerised prescribing, the process of identifying prescribing errors and feeding this information back to prescribers may become less difficult, costly and time consuming. One of the major benefits of e-prescribing is being able to identify the prescribing physician and to track orders. Feeding prescribing error data back directly to individual prescribers via the e-prescribing system may result in greater awareness of prescribing competency and so to a realisation that the e-prescribing system is useful and needed. 3.1. Limitations This study was conducted at only one site but the system and organisational factors in place are typical of those found in other hospitals. We hope our findings may provide general lessons for other sites.

4. Conclusion

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Prescribers most likely overestimated their prescribing competency and so failed to see the value of an e-prescribing system to reduce prescribing errors. Although all doctors recognised advantages and disadvantages associated with electronic prescribing, on the whole, most preferred paper charts to electronic charts. We suggest that keeping prescribers informed about their prescribing errors and the quality-improvement benefits of e-prescribing may lead to greater satisfaction and acceptance of eprescribing systems.

5. Acknowledgements This research was supported by NH&MRC Program Grant [568612]. The authors would like to thank Dr Darren Roberts for his help with participant recruitment and Katrina Richardson for her help with understanding and describing MedChart.

References [1] [2]

[3]

R. Kaushal, K.G. Shojania, D.W. Bates, Effects of computerized physician order entry and clinical decision support systems on medication safety, Archives of Internal Medicine 163 (2003), 1409-1416. C. Zhan, R.W. Hicks, C.M. Blanchette, D.D. Cousins, Potential benefits and problems with computerized prescriber order entry: Analysis of a voluntary medication error-reporting database, American Journal of Health-System Pharmacy 63 (2006), 353-358. A. Wright, S. Phansalkar, M. Bloomrosen, R.A. Jenders, A. Bobb, J.D. Halamka, G.J. Kuperman, J.W. Payne, A.J. Vaida, D.W. Bates, Best practices in clinical decision support, Applied Clinical Informatics 1 (2010), 331-345.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

6

[4] [5]

[6] [7] [8]

[9]

[10]

[11]

[12] [13]

[14] [15] [16]

J.S. Ash, D.W. Bates, Factors and forces affecting EHR system adoption: report of a 2004 ACMI discussion, Journal of the American Medical Informatics Association 12 (2005), 8-12. R.R. Mackie, C.D. Wylie, Factors influencing acceptance of computer-based innovations. In M. Helander, editor. Handbook of Human-Computer Interaction, Elsevier Science Publishers, NorthHolland, 1988. V. Venkatesh, M.G. Morris, G.B. Davis, F.D. Davis, User acceptance of information technology: Toward a unified view, MIS Quarterly 27 (2003), 425-478. F.D. Davis. Perceived usefulness, perceived ease of use, and user acceptance of information technology, MIS Quarterly 13 (1989), 319-340. J.I. Westbrook, M. Reckmann, W.B. Ruunciman, R. Burke, C. Lo, M.T. Baysari, J.Braithwaite, R.O. Day, Effects of two commercial electronic prescribing systems on prescribing error rates in hospital inpatients: A before and after study, PLoS Medicine 9 (2012) e1001164. doi:10.1371 /journal.pmed.1001164. P. McBride, H.G. Schrott, M.B. Plane, G. Underbakke, R.L. Brown, Primary care practice adherence to national cholesterol education program guidelines for patients with coronary heart disease, Archives of Internal Medicine 158 (1998), 1238-1244. B. Woo, B. Woo, E.F. Cook, M. Weisberg, L. Goldman, Screening procedures in the asymptomatic adult: Comparison of physicians’ recommendations, patients’ desires, published guidelines, and actual practice, Journal of the American Medical Association 254 (1985), 1480-1484. D.A. Leaf, W.E, Neighbor, D. Schaad, C.S. Scott, A comparison of self-report and chart audit in studying resident physician assessment of cardiac risk factors, Journal of General Internal Medicine 10 (1995), 194-198. P.A. Glassman, B. Simon, P. Belperio, A. Lanto, Improving recognition of drug interactions: Benefits and barriers to using automated drug alerts, Medical Care 40 (2002), 1161-1171. Z. Niaskhani, H. Pirnejad, M. Berg, J. Aarts, The impact of computerized provider order entry systems on patient clinical workflow: A literature review, Journal of the American Medical Informatics Association 16 (2009), 539-549. J.S. Ash, L. Fournier, P.Z. Stavri, R. Dykstra, Principles for successful computerized physician order entry implementation, AMIA Symposium Proceedings (2003), 36-40. F. Lee, J.M. Teich, C.D. Spurr, D.W. Bates, Implementation of physician order entry: User satisfaction and self reported user patterns, Journal of the American Medical Association 3 (1996), 42-55. J.M Eisenberg, S.V. Willaims, Cost containment and changing physicians’ practice behavior: Can the fox learn to guard the chicken coop? Journal of the American Medial Association 246 (1981), 21952201. W.W. Rosser, Using the perception-reality gap to alter prescribing patterns, Journal of Medical Education 58 (1983), 728-732.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

[17]

M.T. Baysari et al. / Understanding Doctors’ Perceptions of Their Prescribing Competency

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Health Informatics: Building a Healthcare Future Through Trusted Information A.J. Maeder and F.J. Martin-Sanchez (Eds.) IOS Press, 2012 © 2012 The authors and IOS Press. All rights reserved. doi:10.3233/978-1-61499-078-9-7

7

Privacy with emergency medical information used in first response Peter R CROLL and Kimberly M AMBROSOLI Southern Cross University, Gold Coast, Australia

Abstract. In an emergency there are many stages of Medical response. This paper focuses on the first response stage of an emergency medical incident, for example, a sporting accident. Today’s Information Technology together with mobile devices now permits vital medical information regarding an individual to be available at the scene. Those first in attendance are often not medically trained. The literature shows the importance of appropriate first response in minimising harm and the significant investments being made to educate the public in this regard. It also highlights the privacy concerns that arise from the provision of sensitive health information in electronic form. The method utilised is a Privacy Impact Assessment to ascertain the suitability and compliance of the proposed technology. This approach follows the privacy guidelines specified by the Australian government to include information flow mapping, evaluation, compliance analysis, risk analysis and recommendations to management. The resultant output is a list of specific questions and a set of commendations and recommendations that are matched against the National Privacy Principles. The paper concludes that: the approach is both technical and pragmatically viable; it can meet all the reasonableness tests for privacy concerns; it can adopt standard security measures, and; it discusses its potential to be integrated into Personally Controlled Electronic Health Records.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Keywords. emergency response, mobile devices, information security, privacy

Introduction Using Information Technology to manage individual’s medical and health related information has many benefits. This has been recognised by the Australian government through the recent development of the Personally Controlled Electronic Health Record (PCEHR) [1]. The focus to date has been usage with the aged and chronically ill as the main users [2]. Yet the value in quickly obtaining current and accurate information for use in medical emergency situations has been previously emphasised [3]. Emergency or pre-hospital emergency care includes the care and treatment of patients prior to them reaching hospital [4]. This is generally a field for specific emergency services, but in many cases the public can be the ones to provide the first responses. For the purposes of this paper, medical treatment has been divided into three stages. The ‘First Respondents’ are those that are at the scene when the incident occurs. For example, a car accident, an incident in the home or at the beach, during an excursion or at a venue. The ‘Second Respondents’ are the paramedics or ambulance, which if called to the incident, could typically can take from 15 minutes to over an hour. The ‘Third Respondents’ will be the medical team at the emergency ward or treatment centre, which depending on circumstances, can be up to 2 hours later.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

8

P.R. Croll and K.M. Ambrosoli / Privacy with Emergency Medical Information Used in First Response

The focus of this paper is on the ‘First Respondents’ at the scene of the incident. The amount of first-aid or medical training of a first respondent will be varied and dependent on circumstances and luck. For example, on the beach with life savers the respondents will be first-aid trained and on standby whereas with a car accident it could be anyone first at the scene. Yet there are many incidents that occur under controlled conditions where prompt access to medical information would be a valuable life-saver. These include sports events, school outings, club activities, etc. where the individuals are personally known and the first responders more than likely have basic first-aid knowledge. Studies have shown that ‘society spends large resources to inform and educate the public in order to enhance people’s ability to take correct actions’ [5]. Therefore, it can be surmised that there is an expectation for civilians to be more involved at the scene of an incident. This paper will consider the use of information technology to provide knowledgeable carers and group leaders with vital emergency medical information in a timely manner. Important medical information, such as allergies and medical preconditions would normally be kept confidential and only released on a need-to-know basis. Following an accident or other life threatening incident, the rapid access to essential known medical conditions for the individual concerned would be greatly advantageous. There are many examples where such information has been supplied in advance but not available at the scene. That is, the health forms have been completed and filed and are known by certain officers of an organisation; yet no effective method exists to relay this information in those first critical minutes to a first respondent. Often the Privacy concerns [6, 7, 8, 11, 13, 14] prevent this information being more widely available. This paper will, therefore, consider an actual case study that has a viable technological solution; it will then ascertain if privacy and security concerns can be adequately addressed and, finally, if this emergency response information can be integrated into the proposed PCEHR [2]. We acknowledge that privacy issues have been raised by the Australian Privacy Foundation, the Privacy Commissioner, ACHI and other organisations with regard to the PCEHR legislation review, however, the details of these issues are outside the scope of this paper where the focus is on the technology used to retrieve emergency medical information.

1. Implementation Most first respondents will rely on telephone calls to connect with emergency services. The extensive use of mobile phones now allows for information to be communicated directly at the scene of an incident. With the advent of smart phone technologies, such as iPhones, both internet data together with, and even simultaneous with, voice communication is now readily available in a seamless manner. The challenge at the scene of an incident is to provide the simplest possible way to obtain and view emergency medical data for the individual concerned. One solution is the use of 3D Bar charts known as QR codes. The example QR Code shown in figure 1, points directly to the HIC 2012 conference web page. By running a pre-loaded QR Reader Application on a smart phone will automatically launch the appropriate web site on the phone’s web browser to show the relevant information from that web page at the scene.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

P.R. Croll and K.M. Ambrosoli / Privacy with Emergency Medical Information Used in First Response

9

Figure 1. The QR Code, a Student Card and the Web Link with Emergency Medical Information (Note: fabricated for illustration purposes)

These QR Codes are easily generated and printed out. This allows clubs and schools, for example, to print off summary sheets of the class with a photo and associated QR code. This paper summary sheet can be used by the group leader to ensure the emergency medical information is accessible whenever needed.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

2. Privacy Impact The recommended approach endorsed by US, UK, EU, Canada, Australian and New Zealand federal governments is to undertake a Privacy Impact Assessment (PIA) [9]. The Victorian state government of Australia defines a PIA as “an assessment of any actual or potential effects that the activity or proposal may have on individual privacy and the ways in which any adverse effects may be mitigated”. To achieve this requires a detailed analysis of the data flows and architecture of a proposed application to include the details of security measures put in place (both technical and social). This PIA is undertaken by following a step-by-step investigation that takes into account the applicable privacy laws, the international security guidelines, the organisations involved and its governance procedures, the key personnel (including users) and any risk minimisation measure put in place [15].

3. Case Study – Vital One Technologies Vital One Technologies Pty Ltd. have adopted the QR scan technologies and developed a viable and patented Emergency Response system (Patent filed: 11/08/2011, #2011903211). The consultancy undertaken and analysis of their system provided the necessary details in this paper as an actual case study for emergency medical information used by first respondents. The method used to determine the Privacy risks is primarily based on the requirements of a PIA [15]. The supporting documentation issued with the PIA guidelines focusses on the adherence to Privacy Principals and the associated legal framework. The PIA specifies the steps involved and provides a check list of generic questions but does not specify the methods to employ for implementing the analysis [11, 13 & 14]. This case study method involved a qualitative process which included a high level impact analysis; the development of an heuristic checklist derived from PIA guidelines [15]; interviews with key stakeholders and management; and the use of

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

10 P.R. Croll and K.M. Ambrosoli / Privacy with Emergency Medical Information Used in First Response

experts to develop questions appropriate to the application and to drill down where there are perceived problems. Note that larger studies involve a more detailed approach which includes interviews with end-users through focus groups and follow the lines of responsibility through the organisational structure. The approach selected in this case study was to review all the current documentation on the use and operation of the Identity Card; to then undertake a small focus group with the key staff to ascertain the scope and higher levels information flows; then a set of specific question were then drawn up and answered in one-to-one interviews. Based on the answers given, a set of commendations and recommendations were derived for executive reporting based on the NPPs, the legal requirements and experience of known IT issues and common failures. A set of specific question were derived with the key ones listed here to assist the readers in understanding typical implementation and governance issues that arrive from such a study. Privacy Officer: Who is responsible as the Privacy Officer and where is it stated for users to see? Do you have a Privacy Policy and is it readily available for users to view? What happens when someone contacts you to enquire about Privacy Issues? Card Information: Does the card used contain any digital chip, RFID or other hidden digital information? How does the Unique card identifier link to the individual? How do you check the identity of the person on the ICE mobile number on the card? Do you plan to use Blue Tooth or RFID wireless capability on the card in the future? Security/Governance: Is all the emergency medical information stored on a cloud service? Is any of this information encrypted? Is the information being transferred to the Smart Phone (or equiv.) in plain text HTML? Could you make use of SSL encryption technology? Who keeps the passwords, can these be reissued? Do you know which location in terms of States and Countries this emergency medical information in the cloud is actually stored? Can the clubs and schools download a copy of this medical emergency information and store it somewhere else? What happens if your company or the cloud provider goes out of business? What does an individual need to supply in order to set/ reset a password? User Access: Will people provide you with their paper systems containing the emergency medical information for upload and storage purposes? Can users select their own login name or is it assigned? Are the users made aware of the risks involved? Can the users control who accesses what emergency health data? Does your system have an ‘override’ capability (‘break glass’ option)? If there is an override option how is this authorised and audited? Legislation: Will you state the Privacy Laws (or Privacy Principles) that you comply with? If you wish to link to a Personally Controlled Electronic Health Record (PCEHR) are you aware this will require a further Privacy assessment [2]? Are you aware what ‘secondary use’ of data is and what is permitted under legislation? Do you plan to store the Unique Health Identifier (UHI) [10] as only registered health care provides can use this? Are your privacy disclaimers informative to the users about the risks or simply legally defensive? From this initial high-level PIA analysis a number of ‘commendations’ were derived that demonstrated where the organisation was ensuring good privacy practice and which National Privacy Principals apply [6], see examples in Table 1. The resultant output from the PIA is a set of management ‘recommendations’ for minimising the risks of privacy breaches, see examples in Table 2.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

P.R. Croll and K.M. Ambrosoli / Privacy with Emergency Medical Information Used in First Response 11

4. Discussion

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

New technology is providing innovative solutions for rapid access of essential medical information in emergency situations. The wider adoption of smart phone technology allows for critical information at the scene of an incident. The privacy concerns raised needed to be adequately addressed to ensure the medical information is not sought out by adversaries and inappropriate third parties, e.g. pharmaceutical and marketing organisations. The PIA undertaken highlighted some common concerns that could put individual’s at risk and potentially put the commercial operation at risk if the trust of the users is lost through inappropriate data breaches. The analysis showed that it was possible to provide a secure solution giving the right balance between the risk to the individual and community benefits. As with all security and privacy assessments it is necessary to perform ongoing risk audits of the technology on a regular basis. By focusing on the specific aim of supplying emergency medical information reduces the risk of ‘function creep’. The restriction of the scope is essential to ensure that only the minimal emergency information (a minimal clinically useful dataset) is available, thus reducing risk and the personal impact of privacy breach. Although the opportunity to integrate this medical information with the PCEHR has some clear benefits from not replicating information and providing all stages of respondents with the same updated information. That is, should the PCEHR have a medical emergency page that is accessible on a need-to-know basis? This has the advantages for both medical and first-aid respondents whereby they can access the same reliable and up to date information from a common source. The disadvantage is that the PCEHR is not going to be universally utilised and populated, particularly in the case of healthy people which will widely adopt such ID card technology. Also, the added security risk needs to be carefully handled to ensure more sensitive health data is not breached. Further study is needed ascertain the privacy risks that such an integrated solution might introduce. It is recognised that use of IHIs is controlled under the HI Act 2010 and hence integration with the PCEHR may not be so straight forward. Table 1. Example ‘Commendations’ of good privacy practice Commendation The project could involve ‘sensitive’ information. All the staff spoken to in this study demonstrated appropriateness in their approach to dealing with any sensitive information The manner in which medical information is managed and controlled by the users (i.e. the organisations such as schools and sports clubs, plus the individuals who supply their personal emergency medical information) and not by the technology providers is commended.

Reasoning Health Information is regarded as ‘sensitive information’ that requires special handling under the NPPs

Applicable NPPs NPP 10: sensitive information

This minimises the possibility of sensitive information being breached through loss of paper records, etc. and the inappropriate linking of such medical data held in electronic files

NPP 1: Collection NPP 2: Use & Disclosure NPP 3 & 4: Quality and Security NPP 6: Access & Correction NPP 10: Sensitive Information

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

12 P.R. Croll and K.M. Ambrosoli / Privacy with Emergency Medical Information Used in First Response

Table 2. Example ‘Recommendations’ to minimise privacy risks Recommendations Steps need to be put in place that prevent internet searching showing up the emergency details together with a personal identity. This should only normally occur when a person presents their card, i.e. gives consent. Consider steps that prevent access from overseas locations at the user’s request. That is, access from overseas could be blocked by checking the source of the URL when an access request from the Internet is made. It is recommended that the ability to quickly block access to information should be introduced once a card or other linking information is lost or compromised.

Reasoning Note the consent can be regarded as ‘explicit’ - they personally present the card or ‘explicit’ - they are carrying the card and could not be reasonably expected to present e.g. unconscious. Limiting to Australia might be the default option that can be expanded to include overseas (even specific countries for specific periods) when the users set this option. It is important to have quick user/organisation controls to protect access to an individual’s emergency medical information quickly and easily as appropriate.

Applicable NPPs NPP 2: Use /Disclosure NPP 3 & 4: Quality and Security NPP 6: Access & Correction NPP 8: anonymity NPP 9: Transborder data flows

NPP 2:Use/ Disclosure NPP 3 & 4: Quality and Security NPP 6: Access & Correction

5. Conclusions

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

It was found that utilisation of QR codes together with smart phones that can link to web pages of emergency medical information is both technical and pragmatically viable. The high level privacy impact analysis identified areas of concern that were addressed through the recommendations. The organisation concerned implemented the recommended changes to ensure it now meets the reasonableness tests for privacy concerns. Standard tried and trusted security measures were adopted to provide the level of protection necessary. Finally, it was shown that first response medical information has the potential to be integrated into a PCEHR.

Acknowledgements I would like to acknowledge and thank Helen Petaia, CEO and Craig Estwick, Director of Vital One Technologies Pty Ltd. for their assistance and willingness to participate in this publication by way of a case study. They demonstrated a high degree of professionalism with regard to their awareness of the critical issues of Privacy, Security and Confidentiality required for supporting this project. References [1] [2] [3] [4] [5]

Personally Controlled Electronic Health Records (PCEHR) Program Standards Review: Recommendations on Selection of Standards, National E-Health Transition Authority Ver(1) May 2001. Commonwealth of Australia 2011- National Health Reform: The Personally Controlled Electronic Health Record (PCEHR) System: Consumer Booklet, ISBN 978-1-74241-429-4. NEHTA Blueprint: National E-Health Transition Authority, V2.0, 30 September 2011. Elmqvist, C; Brunt, D; Fridlund, B and Ekebergh, M, 2009 Being First on the Scene of An Accident Experiences of ‘Doing’ Pre-Hospital Emergency Care. Larsson, EM; Martenssson, NL and KAE Alexanderson; 2002 “First Aid Training and Bystander Actions at Traffic Crashes – A Population Study, Prehospital and Disaster Medicine, Vol 17, No 3.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

P.R. Croll and K.M. Ambrosoli / Privacy with Emergency Medical Information Used in First Response 13

[6] [7] [8] [9] [10] [11] [12] [13] [14]

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

[15]

ALRC 2008 - “Review of the Australian Privacy Law”, ALRC report 108, Australian Law Reform Commission (http://www.austlii.edu.au/au/other/alrc/publications/reports/108/), May 2008. R Bruyer, ‘Privacy: A Review and Critique of the Literature’ (2006) 43 Alberta Law Review 553, 576. Croll, P.R. Are undue Privacy concerns putting our Health Research at high risk? Privacy Law Bulleting, LexisNexis Butterworths, vol. 2, no.10, April 2006, pp139-140. Office of the Privacy Commissioner - 2006 “Privacy Impact Assessment Guide”, August 2006. Australian Government ComLaw, 2010, Healthcare Identifiers Act, 5 July 2010. Croll, PR, Privacy Impact Assessments – The Organisation Versus the Individual’s Viewpoints. Proceedings of Health Informatics Conference HIC 2008, Melbourne 2008. Office of the Privacy Commissioner, 2006, National Privacy Principles, 14 September 2006. PR Croll and J Croll, Investigating risk exposure in e-health systems . International Journal of Medical Informatics (2007), Volume 76 , Issue 5 - 6 , Pages 460 – 465 PR Croll, Determining the privacy policy deficiencies of health ICT applications through semi-formal modelling. International Journal of Medical Informatics, 8 0 (2011) e32-e38 Privacy Impact Assessment Guide, Office of the Australian Privacy Commissioner, Revised May 2010, www.oaic.gov.au/publications/guidelines/Privacy_Impact_Assessment_Guide.html

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

14

Health Informatics: Building a Healthcare Future Through Trusted Information A.J. Maeder and F.J. Martin-Sanchez (Eds.) IOS Press, 2012 © 2012 The authors and IOS Press. All rights reserved. doi:10.3233/978-1-61499-078-9-14

The Cradle Coast personally controlled electronic health record evaluation research plan

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Elizabeth CUMMINGSa1, Colleen CHEEKb, Winifred VAN DER PLOEGb, Peter ORPINc, Heidi BEHRENSb, Sharon CONDONb, Linda JAFFRAYb, Isabelle ELLISb, d, Barbara RINGEISEN ARNOLDe, Robyn BROGANb and Timothy SKINNERb a eHealth Services Research Group, School of Computing and Information Systems, University of Tasmania b Rural Clinical School, University of Tasmania c Department of Rural Health, University of Tasmania d School of Nursing and Midwifery, University of Tasmania e North West Area Health Service, North West Regional Hospital

Abstract. In 2010 the Federal Government announced funding over two years to create a Personally Controlled Electronic Health Record (PCEHR) for Australians. One of the wave 2 implementation sites is the Cradle Coast in Tasmania. A PCEHR Program Benefits and Evaluation Partner (BEP) has been appointed to undertake evaluation activities with the e-health lead implementation sites. In addition to this implementation a comprehensive research plan has been developed and commenced through the Rural Clinical School at the University of Tasmania. The overarching aim of the research agenda is to evaluate the outcomes of various elements of the 4C project as it evolves and is implemented, from multiple perspectives. The research agenda is important as it expands upon the NEHTA mandated evaluation and provides an holistic overview of the PCEHR implementation process and outcomes for clinicians, patients and family members. This paper will detail the planned evaluation and its progress to date. Keywords. PCEHR, evaluation, advance care planning, Gold Standards Framework

1. Introduction The Federal Government has spent in excess of $450 million to introduce a Personally Controlled Electronic Health Record (PCEHR) for every Australian who wants one. Set for introduction later this year, the PCEHR will provide summaries of patients’ health information; secure access for patients and healthcare providers to their e-health records; and rigorous governance and oversight to maintain privacy [1]. A combination of ‘top down’ national initiatives and ‘bottom up’ lead implementation projects are

1 Corresponding Author: Elizabeth Cummings, eHealth Services Research Group, School of Computing and Information Systems, Private Bag 87, Hobart, Tasmania; Email: [email protected] Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

E. Cummings et al. / The Cradle Coast PCEHR Evaluation Research Plan

15

being used to inform the development of the PCEHR system and its associated infrastructure. In terms of the latter, one such example is the Cradle Coast Connected Care (4C) project based in North West Tasmania, Australia. It emerged from earlier work in the region, through a collaborative body of stakeholders, to use technologies and information systems to create a culture and expectation of health information sharing by establishing secure messaging services between local general practices and allied healthcare providers. As an extension of that work, and with additional funding, the 4C project has the principal objective of optimising the organisation and quality of end-oflife care provided to people in Residential Aged Care Facilities (RACF) across the North West coast using an Australian adaption of the United Kingdom-based Gold Standards Framework (GSF) [2] in RACFs. The system underpinning the project is a shared e-health record that connects RACFs with general practices, local hospitals, after-hours GP services, and allied health providers such as pharmacists. The 4C project will implement a regional PCHER-compliant repository testing national infrastructure and standards set by the National E-Health Transition Authority (NEHTA) whilst also informing the inclusion of residents’ Advanced Care Plans (ACPs) into the PCEHR to assist provision of optimal care for people nearing the end of their life. A PCEHR Program Benefits and Evaluation Partner (BEP) has been appointed to undertake evaluation activities with the e-health lead implementation sites. The focus of the BEP reporting is the adoption of metrics and indicators to monitor the progress of the projects toward realising the potential longer-term clinical benefits. The BEP metrics for the Cradle Coast site were selected based on:

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

x x x x x

magnitude of impact from benefit; attribution and relevance to PCEHR Program; metrics and benefits that are common to multiple capabilities; framework benefits prioritisation; and ease of data collection.

It is anticipated that the BEP will draw on a range of data sources, including jurisdictional and national data, for the purposes of establishing a baseline and setting targets for the various metrics [3]. In addition to this implementation a comprehensive research plan has been developed and commenced through the Rural Clinical School at the University of Tasmania. The overarching aim of the research agenda is to evaluate the outcomes of various elements of the 4C project as it evolves and is implemented, from multiple perspectives. The research agenda is important as it expands upon the NEHTA mandated evaluation and provides an holistic overview of the PCEHR implementation process and outcomes for clinicians, patients and family members. This paper will detail the planned evaluation and its progress to date.

2. Method Four research questions were identified and a methodology for evaluating the impact of the 4CEHR in relation to each of these RQs was determined. Each research question and its related methodology will be briefly described in the following section.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

16

E. Cummings et al. / The Cradle Coast PCEHR Evaluation Research Plan

Participants for each stage of this research will be recruited from residents of the participating residential aged care facilities [n=60], family members [n=20], carers [n=20], RACF staff [n= 30], primary care practitioners [n=30] and Department of Emergency Medicine staff [n=30].

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

2.1. Question 1: What role does an electronic Advanced Care Plan have with the dilemmas of treatment decisions for residents in aged care facilities? This research question aims to examine the efficacy of the 4CEHRs in assisting clinicians with decision-making when there is a change in the condition of a resident of the aged care facility. It is limited to examining the role of the availability of the 4CEHRs in regard to the dilemmas of treatment decisions in relation to residents of aged care facilities. There is no intention to examine any treatment initiated. Data will be collected using four different methods. Firstly, examination of the RACF and hospital records for those RACF residents who have required non-routine medical care or experienced a deterioration in their medical condition during a 3 month period prior to the introduction of the 4CEHR. This will provide the baseline data in relation to the use of ACPs for decision-making. A further set of records will be examined, over a subsequent 3 month period, following the 4CEHR implementation in the RACFs. Examination of these records will be undertaken by a clinical team member with experience in the examination of hospital records in previous research projects. This stage will provide an understanding of whether the 4CEHRs are being used at the dilemma points of treatment decisionmaking. Next, log files from the 4CEHR will be used to assist with identification of the points at which the 4CEHR has been accessed. These data will be used as an indicator for the interview questions. The third method involves conducting semi-structured interviews with consenting members of the key stakeholder groups to understand their impressions of the 4CEHRs and their usefulness. Finally, individual cases where the 4CEHRs have been available will be identified. The clinical staff involved in the care of these residents will be interviewed regarding their decision-making processes and how the 4CEHR may have influenced the decisions made. 2.2. Question 2: What is the process for developing individual electronic Advanced Care Plans and how is it experienced by the individual, their family and health care providers? The evaluation of this question will adopt a pragmatic multi-method approach, as the question requires a description of the development phase and an exploration of the experience of the participants. Pragmatism is a form of research that matches the data collection with the requirements to answer the question. It synthesises different types of data to gain a picture of the complex social environment as well as describing the process elements in detail. The first stage of this evaluation involved the conduct of a desk-top analysis of the design and construction process for the 4CEHR being developed for the 4C Project. This included accessing the meeting minutes of the design team, workshop notes and the build log of the software engineers.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

E. Cummings et al. / The Cradle Coast PCEHR Evaluation Research Plan

17

Following the desk-top audit a series of interviews and focus groups will be conducted. Interviews will be conducted with health care professionals and focus groups with residents and family members. Interviews will be conducted one-on-one with participants at a place and time of their preference. Focus group sessions will be conducted at the RACF premises. All interviews and focus group sessions will be audio taped and transcribed verbatim. Data will be analysed utilising a thematic analysis approach, developing a system of codes with supportive labels, definitions, descriptions and examples. Qualitative software management tool, NVivo, will be used to manage and analyse the data.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

2.3. Question 3: How does the Gold Standards Framework and the electronic Advanced Care plan affect the “culture of care” in residential care homes? The first step in evaluating this research question is to analyse transcripts from after death audits (ADAs). This will be undertaken as a component of the Gold Standards Framework Care Homes (GSFCH) process in order to construct a baseline measure of the pre-GSFCH implementation culture of care and clinical practice in participating RACFs. This baseline will in turn allow the tracking of future changes associated with interventions under the 4C project. After death audits are a key component of the introduction of the GSFCH [3]. Five audits will be undertaken prior to the commencement of the implementation in each RACF, and five in its final stages. Each audit consists of an audio-recorded de-briefing session with staff involved in the final end of life care of a resident who has recently died. Their purpose is to allow staff to reflect on both the care process and their own experience in order to inform future improvements in the quality of end of life care for residents. While the GSFCH has undergone an extensive development and evaluation process in the UK where it originated [4,5], it has neither been fully evaluated in the Australian setting nor in the most recent ‘high intensity facilitation’ format being instituted in the 4C format. Currently, four RACFs have conducted preimplementations ADAs and analysis of the transcripts from these will provide a valuable baseline picture of the current end of life care culture and processes and form a critical part of a extensive, longer-term research project to explore the effect of the GSFCH and 4CEHR process on the culture of care with participating RACFs. In addition to the ADAs and to more deeply explore the issues contained within this broader research question a number of other data collection techniques will be undertaken. These include a broader document audit to examine documents about how process implemented. These documents will demonstrate the staffing levels involved, training provided, and the methods of communication with staff, GPs, and residents and their families. This broader audit will provide documentary evidence of what is occurring during the change management process and will also form the basis for further interview and focus group questions. Interviews will be conducted with a range of participants including staff, residents and family members. These interviews are aimed at gaining understanding the implementation processes from a broader perspective so the involvement of staff will provide an understanding of how the processes have changed. Whereas the interviews with residents and family members will provide an understanding of how well they understand the process, how it is adhered to and whether it has made any difference. Additionally surveys will be conducted with the local primary health care providers and the wider RACF staff population. Given their importance in the

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

18

E. Cummings et al. / The Cradle Coast PCEHR Evaluation Research Plan

development, implementation and sharing of 4CEHRs and their time constraints the use of surveys is most appropriate for use with primary health care providers to gain a wide range of views from as many as possible. These surveys will contain both open and closed questions and elicit an understanding of how well the goals of the project have been communicated, implemented and achieved. Finally, focus groups will be conducted with staff, family members and carers. These focus groups will provide a forum for small groups to discuss the process of the introduction, development and use of the 4CEHRs and provide information about whether there is a perceived benefit in their use. 2.4. Question 4: What additional value does the community place on an electronic advanced care plan compared to current practice?

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

The process involved in seeking to establish the additional value that the community places on an electronic advanced care plan compared to current practice must also be informed by an assessment of what cost is involved. The 4C project team has proposed that the combination of the GSF and the 4CEHR will reduce avoidable hospital admissions and over burdensome treatment. The first part of this process will be to capture the change in hospitals from before the intervention to the 12 months following – thus measuring the change in the number of presentations to the Emergency Department, the lengths of stay and costs of interventions. It will also measure the impact on the costs of RACFs. The second part proposed, contingent on an agreed approach and endorsed by ethics review, is to involve two surveys to establish the benefits of the changes from the perspective of (1) people who have recently been bereaved; and (2) a random sample of the population. In the latter instance a form of contingent valuation (willingness to pay) will be used to allow an estimation of the monetary valuation of the change in benefits where it is assumed there are potential benefits to three groups: (1) the dying; (2) the relatives of the dying; and (3) the bereaved.

3. Discussion Exploration of these research questions is in the formative stages and will continue past the evaluation report by the BEP. So although some overlap exists between the four research questions and the BEP metrics, much of this research will be specific to the Cradle Coast and therefore will not be involved in the national evaluation. In order to minimise the impact on data capture and reporting for sites a collaborative approach between the BEP and the University of Tasmania in ensuring a degree of alignment between the research questions and the BEP evaluation metrics for Cradle Coast. Therefore, a broader comprehensive mixed method approach to the evaluation will provide an holistic evaluation that can be used to demonstrate benefits to a broad range of stakeholders. Four RACFs have commenced implementing the GSF and are working with the project team to ensure that it is possible to implement the 4CEHR by the end of June 2012. It is anticipated that the privacy review and consent model for the 4CEHR project will be approved by April 2012 and that recruitment of people will begin by May.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

E. Cummings et al. / The Cradle Coast PCEHR Evaluation Research Plan

19

The process of implementing the GSF is quite time consuming for an organisation. The implementation of the GSF in the UK raised a lot of issues that may also be reflected in Australia. These included: pre-existing knowledge of practice, standards of teamwork, relationships with GPs, management support for training, suitability of media, timing of conversations, staff turnover, and the cultural change that is implied by introducing an e-health record. At 4CEHR sites, the GSF is taught to the RACF staff through a series of 4 workshops conducted 3 months apart. In between the workshops, the RACFs are visited by the Specialist Nurse Facilitator to help embed the changes and provide support for the key tasks of each workshop. It is anticipated that through the coordinated and comprehensive research and evaluation agenda any benefit or issues relating to the overall development and implementation of the 4CEHR will be examined. These results will then be available to inform the overall PCEHR development, implementation and evaluation. Through the combination of the four evaluation research questions and the metrics involved in the BEP evaluation it is anticipated that feedback from all possible stakeholder groups will be made available to both the PCEHR development and implementation teams nationally and also the wider public. This methodological approach will not only evaluate the success of the local PCEHR implementation but also provide evidence on the issue of end of life care planning.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

4. Conclusions This paper has provided a description of the 4CEHR evaluation plan in relation to the PCEHR Program Benefits and Evaluation Partner and also the specific research evaluation of the 4CEHR project. The 4CEHR is specifically related to the development of advance care plans, using the Gold Standard Framework, and the electronic sharing of the developed ACPs to assist in maintaining the dignity of residents of aged care facilities at the end of life. This paper has also described the methodological approach to the evaluation in relation to the described research questions.

References [1] [2] [3] [4] [5]

NEHTA, 2010, http://www.nehta.gov.au/ehealth-implementation/pcehr-lead-sites Accessed 01/01/2012. Gold Standards Framework Centre, 2011, Overview of the Gold Standards Framework in Care Homes Training Program, www.goldstandardsframework.org.uk/ Accessed 11/11/2011. PWC, 2011, Cradle Coast site analysis report for the Personally Controlled Electronic Health Record (Draft Report). J. Watson, J. Hockley, and S. Murray, Evaluating effectiveness of the GSFCH and LCP in care homes. End of Life Care, 2010. 4(3): p. 42-9. F. Badger, C. Clifford, A. Hewison, and K. Thomas, An evaluation of the implementation of a programme to improve end-of-life care in nursing homes. Palliative Medicine, 2009. 23: p. 502-11.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

20

Health Informatics: Building a Healthcare Future Through Trusted Information A.J. Maeder and F.J. Martin-Sanchez (Eds.) IOS Press, 2012 © 2012 The authors and IOS Press. All rights reserved. doi:10.3233/978-1-61499-078-9-20

An Australian roadmap for ICT research and development for ageing? Lessons from a European Union initiative Elizabeth CUMMINGSa1 and Patrick NIXONa eHealth Services Research Group, School of Computing and Information Systems, University of Tasmania

a

Abstract. The Bridging Research in Ageing and ICT Development (BRAID) project has developed a comprehensive Research and Technological Development (RTD) roadmap for active ageing in European Union countries. This paper provides an overview of the BRAID project, including its methodological approach and outcomes, and suggests that a similar approach could be undertaken to consolidate research and development in relation to ICT and ageing in Australia. Keywords. ICT and ageing, research strategy, Research and Technological Development

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Introduction It has been proposed that information and communications technology (ICT) can help elderly individuals to improve their quality of life, stay healthier, live independently for longer, and counteract the reduced capabilities more prevalent with age. In addition ICT can support older people to remain active at work or in their community [1]. Two concepts that align with the use of ICT are independent living and ageing well. Independent living is the ability for older people to manage their own life styles in their preferred environment. This entails maintaining a high degree of independence and autonomy, enhancing their mobility and quality of life, improving their access to age-friendly ICTs and personalised integrated social and health care services. Ageing well also includes being able to continue active participation in social life and work as desired [1]. However, to date much of the effort regarding ICT & Ageing has been focused on provision of assistance to older persons in that phase of life when a reduction of physical or mental capabilities is observed. Recently some studies have begun to recognise the importance of considering a broader perspective in the ageing process and so there is a growing recognition that the elderly population should not be considered a burden on the society but instead an asset that needs to be properly considered. Research has identified that a critical challenge for society, in respect to the active ageing and ageing well process [2-4], is to identify new organisational structures, 1 Corresponding Author: Elizabeth Cummings, Private Bag 87, Hobart, Tasmania; Email: [email protected] Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

E. Cummings and P. Nixon / An Australian Roadmap for ICT Research and Development for Ageing? 21

approaches, and mechanisms so that elderly citizens not only get better support in their daily needs, but also do not feel excluded, and have the chance to use their knowledge and expertise in making valued contributions to the communities where they live. It has been proposed that ICTs offer the opportunity to: create new ways of working; facilitate social interaction; and increase personal control through the reduction of limitations imposed by location, time and physical and mental conditions. The complexity of the social and political environment, combined with the lessons learned with the limited success of past experiences and the risk of continuously developing technology that is not taken-up by target users, suggests the need for a careful analysis and a better planned approach towards new developments. Roadmapping can play a fundamental role in the identification of a strategic research agenda and prioritisation of needed actions within this context [5]. In this context the European Union Framework 7 funded Bridging Research in Ageing and ICT Development (BRAID) project has developed a comprehensive Research and Technological Development (RTD) roadmap for active ageing. This has been done initially by consolidating existing roadmaps and by describing and launching a stakeholder co-ordination and consultation mechanism. The roadmap characterises key research challenges and produces a vision for a comprehensive approach in supporting the wellbeing and socio-economic integration of increasing numbers of senior citizens in Europe. Researchers from the University of Tasmania have been partners in this research project. This paper will provide an overview of the BRAID project including its methodological approach, the outcomes and suggest that a similar approach could be undertaken to consolidate research and development in relation to ICT and ageing in Australia.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

1. Background As technology progresses in the world of ICTs several major themes are becoming more evident. There are a number of themes that can be considered as basic needs for the effective use of ICTs in an ageing population. Understanding these themes can assist in paving the way for all stakeholder groups in developing programmes to increase the use and availability of ICTs and determine what technology might be useful to an ageing population. The themes, forming the basis of the conceptual framework for the proposed research, are: x x x x x

People want to focus on what they can do, not what they can't. Ageing in place means more than just staying at home. Health is not an objective quality; it is defined collaboratively and culturally. People mark the progression of ageing by watershed events. Healthy ageing is inextricably linked to social participation.

Frequently older people have a combination of social and healthcare needs and it has long been argued that a more integrated approach is needed. The emerging concept of long-term care services incorporating social, health and, sometimes, housing components is seen as having particular relevance in this regard and ICT solutions are not just for healthcare but can be valuable for ageing in general [6, 7]. This conceptual framework has been tested through the BRAID project.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

22 E. Cummings and P. Nixon / An Australian Roadmap for ICT Research and Development for Ageing?

2. Methodology This project was undertaken using a staged approach with eight inter-related work packages as portrayed in figure 1.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Figure 1. BRAID work package linkages

Each work package was led by a different partner organisation but all contributed to many of the work packages. The work packages were: x x x x x x x x x

WP1 - Taxonomy and Trends, Global Security Intelligence: Global Security Intelligence (GSI), United Kingdom WP2 - Stakeholder identification and needs analysis, VDI/VDE Innovation + Technik GmbH, Germany WP3 - Stakeholder co-ordination, Trilateral Research & Consulting LLP, United Kingdom WP4 - Scenario development and consolidated Vision, University of Amsterdam, the Netherlands WP5 - Engaging Stakeholders, CSSC, Italy WP6 - ICT and Ageing Roadmap, UNINOVA, Portugal WP7 - Dissemination and Impact Creation, University of Tasmania, Australia WP8 - Project Management, Queen’s University Belfast, United Kingdom The NETWELL Centre in Dundalk contributes across the work packages.

The arrows on the diagram show the progression of the project over time and the way in which each work package related to the others. Those work packages on the left Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

E. Cummings and P. Nixon / An Australian Roadmap for ICT Research and Development for Ageing? 23

commenced first, whilst those on the right were the later work packages that were fed by the initial work. The work packages 7 and 8 were conducted in collaboration with the other packages and were responsible for co-ordination of the outputs from the project. An initial baseline consolidation exercise was conducted through analysis of the current state of the art in ICT and ageing. This was primarily through a synthesis of contemporary research results complemented by a technology trends analysis along with an examination of socio-organisational models relevant to ageing well and effective inclusion of seniors. This stage was initially undertaken in mid 2010 and repeated twelve months later. Stakeholders formed a critical segment of this project. There were a number of work packages relating to the identification and engagement of stakeholders. Core to these was the examination of stakeholder engagement mechanisms and the development of a stakeholder co-ordination mechanism. This co-ordination mechanism offers direction in how organisations can ensure that currently un-met stakeholder needs are addressed. The identified strengths and weaknesses of the different types of stakeholder co-ordination mechanisms demonstrated that some organisational types are better suited to performing certain tasks and for integrating particular types of stakeholders [8]. A selection of common and critical scenarios was developed and these assisted in the building of the initial visions for ICT supported ageing. The visioning and initial roadmapping was then tested through a series of stakeholder workshops conducted in a range of European Union countries over a period of 10 months in 2011. This was an iterative approach with the results of each of the first four workshops being used to revise the vision and roadmap. The final workshop was used to confirm the draft vision and roadmap prior to the release of the final versions. Once the vision and roadmap were finalised an implementation plan and internationalisation plan were developed to assist with the translation of the project outcomes into policy and practice.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

3. Results The BRAID project has produced a range of key documents and has engaged wide range of stakeholder groups in Europe and internationally. Through the dissemination and stakeholder engagement activities the BRAID project has managed to ensure that information about, and participation in, the project has been able to reach stakeholders in at least 26 nations. A total of 113 individuals from different organisations participated in the BRAID workshops. The final roadmap for ICT research and technology development [9], combined with the strategic research agenda were released in November 2011 [10]. The guidelines for implementation of the roadmap were published in January 2012 [11]. The finalisation of the project will occur in May 2012 with an international conference.

4. Discussion The BRAID project has been a successful collaboration between researchers, independent companies and stakeholder groups from the United Kingdom, Ireland, Germany, Netherlands, Portugal, Italy and Australia. The outcome of the project have

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

24 E. Cummings and P. Nixon / An Australian Roadmap for ICT Research and Development for Ageing?

provided the strategic direction for policy and research in the area of ICT supported ageing for the European Commission. The outcomes have significant potential for other regions also. The researchers consider that it would be possible to use similar techniques and test the appropriateness of the final roadmap and recommendations for the Australian environment, with the potential to expand to other countries at a later date. So consideration has been given to what would be required to test this for Australia. It is suggested that this could be undertaken in the following six steps. Baseline consolidation of the research into the Australian context through analysing the current state of the art in ICT and ageing, through a synthesis of contemporary research results, complemented by a technology trends analysis. Socioorganisational models relevant to ageing well and effective inclusion of seniors will also be analysed. Confirmation of the critical scenarios and vision developed by the BRAID project. Recent dramatic changes in the socio-economic system, such as the current economic crisis, are likely to have a deep impact on the living conditions of senior citizens. The consideration of various such critical scenarios, together with the identification of key drivers and relevant trends, is an important basis for defining a vision of a desirable future supported by technology and an appropriate socio-organisational mechanism. Confirmation of the strategic plan of RTD actions. Induced by the analysis of the gap between current baseline and the established vision, complemented by a Strengths, Weaknesses, Opportunities, and Threats (SWOT) analysis, a strategic plan of RTD actions will be examined, which may lead to the development of new technological solutions and more adequate socio-organisational structures. Policy recommendations accompany this plan. Proposals elaborated in previous roadmapping projects will be analysed, reviewed, consolidated and extended to ensure their relevance to the Australian context. Consensus building. Previous steps, taken in terms of the traditional scientific method, correspond to the background observation and hypothesis formulation. This step corresponds to the hypothesis validation and revision through extensive consultation with stakeholders. This will be undertaken through interviews with experts and collaborative workshops. Proposing implementation mechanisms. Based on the revised vision and plan of actions, implementation modalities, timing and needed resources are defined for each action. Recommendations regarding policies and novel socio-organisational models are elaborated hand-in-hand with the strategic research agenda. Raising awareness and promoting stakeholder engagement. Once completed, both the vision and the roadmap need to go through extensive dissemination, using appropriate channels for each group of stakeholders, in order to promote acceptance and engagement in its implementation.

5. Conclusion This paper has provided an overview of the Bridging Research in Ageing and ICT Development (BRAID) project and its development of a comprehensive roadmap and strategic agenda for research and technology development to support active ageing. It has suggested a method by which this European based project could be tested and possibly implemented in an Australian context. There is a degree of duplication of

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

E. Cummings and P. Nixon / An Australian Roadmap for ICT Research and Development for Ageing? 25

effort and funding that is occurring in the area of ICT research and development, particularly in relation to supporting ageing. It is important that the findings from the BRAID project are used to address ICT and ageing globally. However, the issue of cultural differences needs to be understood. These may influence the types of technologies that are acceptable to different people in different nations. Therefore the research undertaken within an international context should include inter-disciplinary teams who can understand the cultural aspects as well as the technological aspects of the developments. Different nations are in different stages of development and have varying levels of skills and infrastructure. These differences need to be considered when implementing the BRAID roadmap. Developments should cover a range of possible users not be homogeneous. It is evident that collaborative networks should be developed and they should build upon current activities, particularly where R&D is being undertaken well. It is not always essential to repeat investigations when introducing concepts or technology to a new country.

Acknowledgements The authors would like to acknowledge the contributions of the partner organisations from the BRAID project. BRAID is an ͒EU FP7 Support Action funded within the specific programme "Cooperation" and the research theme "ICT" of the 7th European Framework Programme (Objective ICT-2009.7.1 ICT & Ageing).

References

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

[1]

European Commission. (2010) Overview of the European strategy in ICT for Ageing Well. Brussels: European Union, Accessed 20/04/2011, http://ec.europa.eu/information_society/activities/ einclusion/docs/ageing/overview.pdf [2] Camarinha-Matos, L., and Afsarmanesh, H. (2009) Collaborative mechanisms for a new perspective on active ageing. In: DEST 2009 - 3rd IEEE Int. Conference on Digital Ecosystems and Technologies Istanbul, Turkey: IEEE. [3] HSBC Insurance. (2007) The future of retirement – The new old age. Accessed 20/04/2011, hsbc.com/1/PA_esf-ca-app-content/content/assets/retiremet/2007_for_report.pdf [4] Wilson, P., Johnson, K., and Buller, W. (2008) Ageing well in a connected world: Smarter care, smarter participation. CISCO Internet Business Solutions Group. Accessed 20/04/2011, http://s3.amazonaws.com/connected_republic/attachments/9/Aging_Well_POV_REV1130.pdf [5] Camarinha-Matos, L., and Rosas, J. (2011) D6.1 Interim Roadmap for ICT and Ageing. BRAID Project. Accessed 01/03/2011, http://www.braidproject.eu/?q=publications [6] Wadhwa, K. (2010) D1.2 Technology Baseline & Trends. BRAID Project. Accessed 05/04/2011, http://www.braidproject.eu/?q=publications [7] Empirica and Work Research Centre. (2010) ICT and Ageing: European Study on Users, Markets and Technologies: Final Report. ICT and Ageing. Accessed 21/06/2011. ec.europa.eu/information.../studies/docs/ict_ageing_final_report.pdf [8] Finn, R., and Wright, D. (2011) Mechanisms for stakeholder co-ordination in ICT and ageing. Journal of Information, Communication & Ethics in Society 9, 265-286. [9] Camarinha-Matos LM, Ferrada F, Oliveira AI, Rosas J, (2011), BRAID D2.1 Consolidated Roadmap for ICT and Ageing, http://www.braidproject.eu/?q=publications [10] Camarinha-Matos LM, Ferrada F, Oliveira AI, Rosas J, (2011), BRAID D2.2 Strategic research agenda for ICT and Ageing, http://www.braidproject.eu/?q=publications [11] Camarinha-Matos LM, Ferrada F, Oliveira AI, Rosas J, Afsarmanesh H, Lynch U, Hadjri K, Bond R, Brielmann M, (2011), BRAID D6.3 Guidelines for roadmap implementation, http://www.braidproject.eu/?q=publications

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

26

Health Informatics: Building a Healthcare Future Through Trusted Information A.J. Maeder and F.J. Martin-Sanchez (Eds.) IOS Press, 2012 © 2012 The authors and IOS Press. All rights reserved. doi:10.3233/978-1-61499-078-9-26

An information management system for patients with tuberculosis: Usability assessment with end-users

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Jonathan DARBYa,b,1, Jim BLACKb, David MORRISONc and Kirsty BUISINGa,d a Infectious Diseases Department, St Vincent’s Hospital, Melbourne b Nossal Institute for Global Health, The University of Melbourne c Department of Epidemiology & Preventive Medicine, Monash University d Victorian Infectious Diseases Service, Royal Melbourne Hospital

Abstract. Introduction: Information systems with clinical decision support (CDS) offer great potential to assist the co-ordination of patients with chronic diseases and to improve patient care. Despite this, few have entered routine clinical use. Background: Tuberculosis (TB) is an infection of public health importance. It has complex interactions with many comorbid conditions, requires close supervised care and prolonged treatment for effective cure. These features make it suitable for use with an information management system with CDS features. In close consultation with key stakeholders, a clinical application was developed for the management of TB patients in Victoria. Methods: A formal usability assessment using semi-structured case-scenario based exercises was performed. Subjects were 12 individuals closely involved in the care of TB patients, including Infectious Diseases and Respiratory Physicians, and Public Health Nurses. Two researchers conducted the sessions, independently analysed responses and discrepancies compared to the voice record for validity. Results: Despite varied computer experience, responses were positive regarding user interface and content. Data location was not always intuitive, however this improved with familiarity of the program. Decision support was considered valuable, with useful suggestions for expansion of these features. Automated reporting for correspondence and notification to the Health Department were felt worth the initial investment in data entry. An important workflow-based issue regarding dismissal of alerts and several errors were detected. Conclusion: Usability assessment validated many design elements of the system, provided a unique insight into workflow issues faced by users and hopefully will impact on its ultimate clinical utility. Keywords. usability assessment, hospital information systems, clinical decision support systems, communicable diseases

Introduction Technological advances have created new opportunities to help manage healthcare data and ultimately improve patient care. These systems may have several overlapping functions, such as the viewing of investigation results, provision of clinical decision support systems (CDS), composition of an electronic health record (EHR) to collate 1

Corresponding Author: Dr Jonathan Darby, Department of Infectious Diseases, St Vincent’s Hospital, Melbourne, 41 Victoria Parade, Fitzroy, Victoria, 3065; Email: [email protected]. Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

J. Darby et al. / An Information Management System for Patients with Tuberculosis

27

notes, enabling disease surveillance and epidemiology, and provision of auditing and research capability. Many electronic information systems for healthcare have been produced, but overall adoption by clinicians and other end-users has generally been slow. The potential reasons for this are manifold, however problems with usability are likely to be central to poor uptake [1-3]. In the design and development of systems for clinical use, it is important to keep the needs of the end-user a priority to ensure that the content and functionality is relevant and can be integrated into routine clinical workflow. Information management systems are particularly suited to chronic conditions that require coordinated care by several health professionals over prolonged periods of time. They are also suited to complicated treatment protocols where medication dosing and drug interactions need to be individually tailored and carefully monitored, with potential for rule-based alerts to be utilised. Clearly such systems also facilitate organisation of data for public health surveillance, performance auditing and research. Tuberculosis (TB) is an infectious disease with a worldwide distribution that can be transmissible and although Victoria is an area of low endemicity, it remains of public heath importance [4]. Control of TB disease and transmission requires adequate antibiotic treatment and effective tracing of contacts. Case numbers have increased over the last twenty years, largely caused by the breakdown of social infrastructure in many countries and fuelled by the HIV epidemic [5]. Patient care is complicated by prolonged treatment required, particularly with rising global drug resistance, extending treatment from a standard of 6 months to 18-24 months [5]. Thus, TB provides a good example for the development of an electronic system to assist with the clinical care of patients with a chronic infectious disease.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

1. Background There are limited published experiences with applications related to Tuberculosis. Those that have been described deal with managing data on patients from an operational sense, or assisting with logistics such as medication supply and procurement, rather than being specifically focused on the total clinical encounter [6, 7]. Until recently most hospitals in Australia have been using paper-based records. EHR offer many advantages over traditional patient files, but few are tailored to patients with specific diseases. The benefits of disease specific systems include an ability to customise fields, create relevant alerts and CDS, produce reports (e.g. letters, visit records) that are relevant to health professionals involved at different time points, and in the long term, by standardising data collection to improve audit and research capability. Our aim was to analyse the usability of a newly developed TB Clinical Application [8, 9] with health professionals of different backgrounds and varied levels of seniority, who all have a stake in the routine care of patients with TB in Victoria.

2. Methods A Computerised Management System for Tuberculosis management was developed by the authors (Infectious Diseases Physicians), in collaboration with VPAC and the Victorian Department of Health. The application was built with a .Net web-based

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

28

J. Darby et al. / An Information Management System for Patients with Tuberculosis

interface utilising a MS SQL database and linked to an integration engine that incorporates patient administration, radiology and pathology data via Health Level Seven (HL7) messages. The exercises were conducted on a PC running Microsoft Windows 7 and Internet Explorer 8, with the program installed on the hospital server. Two research staff conducted each exercise: one leading the tasks and the other observing. Both researchers recorded and independently analysed observations about the process. Triangulation of findings was achieved by verification against the voice record where required. Twelve participants were tested; clinicians (Infectious Diseases and Respiratory) within the public health sector, and Public Health TB Nurses from the Department of Health. Each set of exercises required approximately two hours. This project was approved by the St Vincent’s Hospital Human Research Ethics Committee. Prior to the exercises, subjects were asked about their expectations and current difficulties in managing information. A brief orientation was then provided outlining the structure, content and functionality of the system. Each participant (doctor or nurse) was led through four patient-based scenarios, each highlighting different potential workflow pathways. The system was pre-populated with mock data with varying amounts of information provided depending on the task. Following the exercises, structured feedback was sought regarding interface impressions, content suggestions, alert functionality and research and quality assurance activities.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

3. Results Prior to exposure to the program, most participants were limited in their expectations. Their focus was on presentation of data in a clear, organised manner and viewed the TB Clinical Application as a static repository or database. They had no expectations regarding more sophisticated functionality such as decision support, creating reports or electronic communication. A strong theme from all participants was the need to have the medication history clearly documented, which is consistent with the importance placed on medication documentation and adherence for TB patients. Few voluntarily mentioned the use of this system to guide their decision-making, provide reminders, or the future use for auditing or research purposes. 3.1. Functionality/User Interface Participants had varied prior experience of computerised patient systems, however most found the interface easy to navigate after some initial familiarisation. The ‘Patient Summary’ across the top of the screen displaying a snapshot of the most important information was viewed favourably by all users, as were the tabs used to organise each patient’s data into sections. Some tab titles were not completely clear to users with respect to information within them, but locating data improved with use. In general, buttons were felt to be intuitive, as was the use of definitions when the cursor hovered over an item. Scrolling down the screen to view all relevant information and use of the ‘Edit/Save’ toggle button was initially difficult for some users, but improved with use. Participants appreciated workflow that protected unintentional changes being made. Several issues were identified for suggested improvement. This included the ambiguity of the ‘Delete’ function, interpreted variably as ‘to cease’ or ‘permanently delete’ an item, or an item that was added in error. Useful suggestions were to add a warning “Are you sure?” or “This will delete the medication from the record entirely”.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

J. Darby et al. / An Information Management System for Patients with Tuberculosis

29

A consistent theme was the dislike of hidden information. This included results tables and a medication table that initially defaulted to a nominated small number of recent entries to minimise the use of screen space. The user’s preference was almost always for a default setting of ‘View All’ rather than pre-determined limits being set. Most commented that the full treatment history needed to be visible to provide a reminder, or “a crucial map of where the patient had been”, as stated by one user. Another consistently mentioned concern regarded results tables that required horizontal scrolling to see all content. Regarding order of display, participants preferred most recent results at the top, conforming to familiar programs currently in use. 3.2. Content

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Participants generally appreciated the utility of the data collected. Some commented on the time required to enter the large volume of data collected for an initial patient encounter, but most users believed this was balanced by the time saved by generating automated reports, notifications, correspondence letters etc. with this information. We discovered that a ‘Date of Diagnosis’ was difficult to consistently define, and our surrogate field for this, ‘Date of Treatment Commencement’ needed to be refined. Examples were proffered, “Where the patient may not accept treatment, where treatment was delayed for a particular reason, or where treatment was never administered”. Suggestions included the addition of a new field for ‘Date of Diagnosis’ where this may default to ‘Date of Treatment Commencement’ where appropriate, but where these dates are different this may be stated clearly. A new field was proposed, ‘Diagnosis Based On’ with options ‘Clinical, Culture, PCR, Radiological, Histology’. Several participants had useful suggestions for making existing fields more prominent. This included adding information to the patient banner, such as ‘Sputum Smear status’ (an indicator of overall infectivity) and a more detailed description of ‘Drug Resistance’ (patients requiring more intensive treatment for longer durations). Additional fields that were suggested included a radiological categorisation of chest Xray findings, and a separate module for latent TB (a related dormant form of TB). 3.3. Utility Time management: Speed of use increased during the course of the four patient-based exercises. There was a general belief in the benefit of time spent entering data once you saw what the system could produce. The best example of this is the legislative requirement to notify the State Health Department in writing of a case of TB. The document takes some time to complete and may be difficult to locate, in paper or electronic form. Participants liked the ability of the system to produce and automatically fax this form to the relevant health authority. This enthusiasm will have to be tested in a busy clinical environment away from the test setting. Alerts: Most users found that the built-in alerts were for useful indications. However, an important workflow issue was discovered whereby the item being alerted could be dismissed without being viewed properly. Many users had useful suggestions to improve the logic of the system so this was avoided. A universal opinion was the misunderstanding of the word ‘Action’ when used to turn off a specific alert. Some felt this to mean a particular task had been undertaken, while another “expected to be taken to a screen to perform a specific action”. Alternatives such as ‘Acknowledged’ or

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

30

J. Darby et al. / An Information Management System for Patients with Tuberculosis

‘Noted’ were offered as better terms, with the intention to turn the alert off. Participants agreed that an audit trail would be useful to track this function. Suggestions for additional alerts included such items as ‘Pathology Out of Range’, ‘Hospital Admission’, and ‘Multi-Drug Resistant TB’. After experiencing the system, several subjects began to link data fields in a decision support framework. Examples included the prescription of the TB medication, Ethambutol and linking this to reminders for accepted clinical screening such as monitoring of renal function and visual side effects. Another suggestion was that the prescription of a corticosteroid drug be linked to baseline screening for Hepatitis B and Strongyloides. Our exercises identified that medical staff were more likely to be interested in clinical indicator alerts, whereas public health nursing staff were more interested in alerts regarding epidemiology that might reflect potential for transmission, nonattendance to appointments or medication non-adherence. One participant helpfully commented, “As long as you didn’t get alerted to death … because it would be ... crying wolf to a certain extent if an important one was buried in there somewhere”. Communication: Most users identified that the system would enhance communication between health providers in several ways; between doctors within a specific clinical unit, between doctors and nurses working in the Health Department, and between doctors and the patient’s other health care providers. Reporting functions such as a clinic visit record, population of letter to the patient’s local doctor and notification of TB form were anticipated to be very useful.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

3.4. Research Although few users volunteered ideas about how the system may contribute to future research, when prompted all identified its utility. Users identified potentially overcoming inconsistency of data collection and difficulties in extraction of information from paper-based records. Links between hospitals were proposed to broaden the patient base available for research. Quality improvement activities were also generally seen as a potential benefit, such as an analysis of treatment outcomes, attendance rates to clinic and monitoring frequency of various drug toxicities. Comments made following the exercises were generally positive. Participants envisaged the system to be useful and time-saving when seeing a new patient for the first time, as it provided a good snapshot of their overall progress to date.

4. Discussion and Implications Important lessons were learned during the process of undertaking this usability exercise. Many design features that the authors felt offered advantages were reinforced by participants and seen to be beneficial. However, several major issues were highlighted, and without conducting this exercise these may not have been discovered, potentially leading to adverse clinical consequences. Examples of this include the reordering of workflow for alert dismissal, and the full display of results. Many computerised information systems have been built by groups disparate from the clinical environment and without an emphasis on improving patient care. By utilising decision support features in varied form, the authors have kept the clinical encounter with the patient at the forefront of development. Multiple methods have been employed to ensure medical, nursing and public health priorities have been addressed.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

J. Darby et al. / An Information Management System for Patients with Tuberculosis

31

Several suggestions from participants have already been implemented and there are further enhancements scheduled for inclusion with the next phase of software development. This project has significant scope to further integrate with existing systems. A hospital wide EHR has recently been implemented at this institution and work is underway to automatically export reports from the TB management system. Given the importance of medication adherence to TB treatment success [10] further possibilities to be explored include the integration of electronic prescribing, electronic reconciliation with pharmacy dispensing records, and incorporating an interaction with the patient. This may take the form of an interactive patient viewer portal, where patients can contribute observations about their health, communicate with health care providers and source trusted health information. Although designed with TB as the focus, the development group maintained flexibility and built the system with a generic ‘back-end’ database to potentially suit a broader audience and enable adaptation to other disease states [11]. It was critical to obtain the view of end-users at all stages of development of this system, from initial concept through to the final testing of the end product. Thorough training and support will be required as this program is implemented in the clinic setting, to ensure clinicians have the confidence to use it on a regular basis. Ongoing evaluation is intended to ensure the program retains relevance for clinical utility.

Acknowledgements The authors would like to acknowledge the generous funding from the John Burge Tuberculosis Trust for the development of this system and related projects and thank study participants for their generous time and insightful comments.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

References [1] [2] [3] [4] [5] [6] [7] [8]

[9]

Saleem JJ, Patterson ES, Militello L, Render ML, Orshansky G, Asch SM. Exploring barriers and facilitators to the use of clinical reminders. J Am Med Inform Assoc. 2005;12(4): 438-47. Sittig DF, Krall MA, Dykstra RH, Russell A, Chin HL. A survey of factors affecting clinician acceptance of clinical decision support. BMC Med Infor Decis Mak. 2006 1;6:6. Kawamoto K, Houlihan CA, Balas EA, Lobach DF. Improving clinical practice using clinical decision support systems: systematic review of trials to identify features critical to success. BMJ 2005;330:765 Surveillance of notifiable infectious diseases in Victoria 2008, Department of Health, Victoria, 2010. Available from: http://ideas.health.vic.gov.au/surveillance/annual-reports.asp. Accessed February, 2012 Global tuberculosis control: WHO report 2011. World Health Organization. Available from: http://www.who.int/tb/publications/global_report/en/index.html. Accessed February, 2012. Marcu A, Farley JD. A comprehensive infectious disease management system. Stud Health Technol Inform. 2009;143:364-7 e-TB Manager. Management Sciences for Health. Website: http://www.eTBmanager.org. Accessed February 2012 Darby JD, Black JF, Buising KL. Determining the requirements for a computerised information management system to improve the care of people with Tuberculosis (TB). Australasian Society for Infectious Diseases Annual Congress, 2011, Lorne, Victoria. Darby JD, Black JF, Buising KL. Using evidence to guide the development of an information management system for the care of patients with Tuberculosis (TB) in Australia. Australasian Society for Infectious Diseases Annual Congress, 21-25th March, 2012, Fremantle, Western Australia. (Upcoming conference, abstract accepted).

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

32

J. Darby et al. / An Information Management System for Patients with Tuberculosis

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

[10] Stop TB policy paper: contributing to health system strengthening: guiding principles for national tuberculosis programmes. World Health Organization. Available from: http://www.who.int/tb/health_systems/en/ Accessed February, 2012. [11] Kawamoto K, Del Fiol G, Lobach DF, Jenders RA. Standards for scalable clinical decision support: need, current and emerging standards, gaps, and proposal for progress. Open Med Inform J. 2010;4:235-44.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Health Informatics: Building a Healthcare Future Through Trusted Information A.J. Maeder and F.J. Martin-Sanchez (Eds.) IOS Press, 2012 © 2012 The authors and IOS Press. All rights reserved. doi:10.3233/978-1-61499-078-9-33

33

We are not educating the future clinical health professional workforce adequately for e-health competence: Findings of an Australian study

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Ambica DATTAKUMAR a, Kathleen GRAY a, Kerryn-Butler HENDERSON b c, Anthony MAEDER c and Helen CHENERY d a Health and Biomedical Informatics Research Unit, University of Melbourne, Australia b Department of Health Policy and Management, School of Public Health, Curtin University, Australia c School of Engineering, Computing and Mathematics, University of Western Sydney, Australia d Faculty of Health Sciences, University of Queensland, Australia

Abstract. This paper reports on a national study of the present approaches in Australian tertiary education, to preparing future clinical health professionals to work competently in an increasingly e-health enabled healthcare sector. The argument for increasing clinical health professionals’ knowledge about e-health and health informatics has been advanced repeatedly over past decades in Australia and elsewhere. However, peer-reviewed accounts of good practice in implementing and evaluating e-health education in health profession degrees anywhere are scarce. Our study reports on surveying approximately 100 degree coordinators in 40 clinical health professions in 30 universities across Australia. It finds that currently, teaching and assessment of future clinical health professionals does not ensure that Australia will have a clinical workforce that is adequately professionally empowered to work with e-health. This paper provides important baseline data for planning improvements to e-health education for Australia’s future clinical health professionals. Keywords. education, e-health, health professions

1. Background and Concise Literature Review Future clinical health professionals will need to be able to work competently with information and information technology in an increasingly e-health enabled healthcare sector. Clinical health professionals are those that are involved in observing and treating patients directly. The argument for increasing clinical health professionals’ knowledge about e-health and health informatics has been advanced repeatedly; see [1] for a recent Australian example and [2], [3] for contemporary international perspectives. The e-health education of entry-level clinicians in the health professions appears not to be keeping up with current trends and demands for the changing technological environment in which they work, especially e-health innovations. Previous qualitative

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

34 A. Dattakumar et al. / We Are Not Educating the Future Clinical Health Professional Workforce Adequately

accounts have documented cases where university education did not equip clinical health professionals to work in a professionally empowered way with information and communication technologies in their workplaces [4], [5]. Indeed, there appears to be a need for a comprehensive review of e-health education, especially to strengthen implementation and evaluation of curriculum [6]. This paper aims to provide such an overview of e-health education for Australia’s future clinical health professionals, across all professions, states and tertiary education institutions.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

2. Methods A survey, administered online, listed 10 statements, each with examples or prompts. A 7-point Likert scale measured the extent of agreement or disagreement with each statement (strongly disagree, disagree, neutral, agree, strongly agree, don’t know, not applicable). Each statement also invited free-text comments or examples. Based on the Australian and New Zealand Standard Classification of Occupations in medicine, nursing and other health professions [7], and on national university websites and technical college program handbooks, extensive efforts were made to identify the coordinator of every relevant degree, resulting in personal invitations to 400 prospective participants. They represented degrees in Medicine, Nursing, Allied Health, Complementary Therapies and numerous other clinical professions which prepared entry-level professionals at levels 6 through 10 of the Australian Qualifications Framework, i.e. Bachelor through to Doctorate degrees. 105 surveys were completed, during the second half of 2011 (incomplete surveys are not included in this total). The survey questions and information on the participating professions and educational institutions can be found in [8]. Descriptive statistical analysis of the quantitative survey data was automatically generated by the online survey tool. Thematic analysis of the qualitative data was based on four key questions that are founded on previous review of the literature [6]. As noted by Creswell on page 220 [9], ‘the quantitative and qualitative data collection may be presented in separate sections, but the analysis and interpretation combined the two forms of data to seek convergence or similarities amongst results’.

3. Findings and Discussion This section presents findings from the survey, grouped under the four key questions and accounting for a representative cross section of the data. Example quotes chosen from the survey responses are included to illustrate various viewpoints from a range of professions, with the actual participant number shown as PXX. 3.1. Why do health professionals need to be competent in e-health? Survey participants were asked whether the professional / industry body that accredits their degree program required that at least some aspects of e-health are included in learning, teaching and assessment. 31.5% of the participants agreed. 38.1% of the participants disagreed.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

A. Dattakumar et al. / We Are Not Educating the Future Clinical Health Professional Workforce Adequately 35

•

Accreditation body requires inclusion of e-health in learning, teaching and assessment: “It is required that students demonstrate an awareness of electronic record storage in the competency based outcome standards”. Speech Pathologist, P104 “The Bachelor of Nursing at SNM has been accredited against the draft Standards and Criteria for the Accreditation of Nursing and Midwifery Courses Leading to Registration, Enrolment, Endorsement and Authorisation in Australia for Registered Nurses (ANMC, 2009). Standard 4 - Curriculum Content requires us to report against the criteria of Information Technology for Health Care”. Nurse, P61 “Guidelines for curriculum content and accreditation requirements contain specific statements about the inclusion of e-health”. Behavioral and Social Sciences, P1

•

Accreditation body does not require inclusion of e-health in learning, teaching and assessment: “There is no explicit statement in the list of Standards and Elements on e-health in the Australian Physiotherapy Council accreditation requirements, however, e-health may be seen as imbedded within some of the standards eg. operate effectively across a range of settings; participate in quality improvement processes”. Physiotherapist, P96 “This is not stated specifically by the accrediting body”. Occupational Therapist, P8 “Not one of their criteria, but it is expressed in discussion”. Medical Practitioner, P24

This suggests that if the accrediting body does not include statements on e-health competencies, it may not be specifically addressed in curriculum and assessment. Thus, students may not receive formal education in this area. When participants were asked whether the major employers of their graduates expected some aspects of e-health to be included in their degree, 61.9% of the participants agreed and 21.9% of the participants disagreed.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

•

Employers expect graduates to be competent in e-health: “Graduates would be expected to be able to access all forms of electronic information regarding patient records and care”. Medical Practitioner, P89 “Employers expect our graduates to be able to use computed radiographic imaging systems including CT and the PACS interface that communicates with the hospital departments and wards remote from the radiology department”. Medical Imaging Professional/Radiographer, P88 “Our students would be required to know how to use patient records, how to access and undertake reviews of the literature”. Dietitian /Nutritionist, P29

•

Employers do not expect graduates to be competent in e-health: “Major employers require content & skill knowledge but not necessarily e-health capability. However, by the time our distance students graduate they have good etechnology skills”. Clinical Psychologist, P2 “Very little technology, apart from those who go into health placements and employment pathways, is required of social workers when compared with other disciplines. e-health notes and statistics would be the major requirement for hospitalbased students. Child protection requires knowledge and use of client information systems”. Social Worker, P95 “Ambulance service does not expect graduates to be able to conduct research to any great degree, and do not wish for us to teach them electronic patient care records as this is done once they are in their grad year”. Paramedic, P77

There is a divide between the accreditation requirements of degrees and the employer expectations of graduates.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

36 A. Dattakumar et al. / We Are Not Educating the Future Clinical Health Professional Workforce Adequately

3.2. What does it mean for a health professional to be competent in e-health? Participants were asked whether they use independent expertise to support decisionmaking about learning, teaching and assessing e-health in the degree. While 20.9% of participants agreed, 49.5% disagreed. •

Academics utilise external resources/expertise to support decision-making: “Yes we use info from NEHTA [National E-Health Transition Authority]”. Dietitian, P100 “E-technology consultant employed within the school”. Nurse, P79 “We are discussing content with other e-health academics”. Public Health professional, P94

•

Academics not utilising external resources/expertise to support decision making: “We have not yet included specific aspects in curriculum design”. Occupational Therapist, P84 “Curriculum design and implementation is always a balance between the needs of the program overall and the apparent needs of a discipline. It is generally not possible to access, assess and implement all external curricula in our curricula. We do however use them (at times) to bench mark our curriculum”. Medical Practitioner, P91 “No external expertise has been called upon at this point in time”. Social Worker, P95

This may indicate a lack of awareness of key resources on e-health competencies amongst those responsible for curriculum change. 3.3. How can we build e-health capability in students in the health professions? Participants were asked two questions on this issue, one about the curriculum content of the degree program and the other about teaching methods and whether they addressed e-health. While 83.8% of participants agreed, 9.6% disagreed that e-health is included in their curriculum content. Copyright © 2012. IOS Press, Incorporated. All rights reserved.

•

Curriculum content addresses e-health: “The curriculum has a strong emphasis on Research Enriched Learning and Teaching (RELT). The curriculum containing individual subjects on evidence based practice in addition to discipline specific EBP [Evidence-Based Practice]content dispersed throughout the curriculum”. Physiotherapist, P13 “Evidence-Based practice, Communication, Clinical Decision Support systems, Professional Experiences in actual health care agencies”. Nurse, P61 “Several lectures include content relating to electronic patient records, how they are accessed and how they relate to the hospital environment and private practice”. Exercise/ Sports Scientist/ Physiotherapist, P32

•

Curriculum content does not address e-health:

“This is not specifically taught as a topic within the current degree program but electronic record keeping is used throughout most clinics and students are expected to conform to current practices of reporting”. Audiologist, P22 “We cover evidence-based learning of psychological skills, knowledge & research, but nothing is specifically related to "e-health". Psychologist, P4 “Only limited specific discussion of evidence based practice however would not include this as part of e-health or informatics”. Nurse, P81

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

A. Dattakumar et al. / We Are Not Educating the Future Clinical Health Professional Workforce Adequately 37

Many participants stated that Evidence-Based Practice was a key part of the curriculum. This may reflect a misconception that this is a sufficient way to address ehealth in the curriculum. 51.4% of participants agreed that the teaching methods used in the degree program address at least some aspects of e-health. 28.6% of participants disagreed. •

Teaching methods address e-health: “Students have access to remotely located guest lecturers who present using teleconferencing”. Nurse, P15 “Guest speaker from the State Government Perinatal Data Collection Unit”. Midwife, P80 “The fact that the subject notes must be down loaded by students provides a basis for this type of learning environment. However, there is no designated aspect of the course in which this sort of information is presented”. Massage/ Musculoskeletal/ Myotherapist, P69

•

Teaching methods do not address e-health: “Some staff are more e-savvy and use technologies such as Skype & Adobe Connect but there is no requirement. We have guest lectures at residential schools”. Psychologist, P4 “The teaching methods to date do not use telehealth or other methods”. Dietitian/ Nutritionist, P90 “Not at this stage, again as above it’s a new program and we are looking towards doing this”. Dietitian, P55

There seems to be no clear view of what methods are effective for teaching ehealth to future health professionals. 3.4. How can we assess the e-health capability of students in health profession degrees?

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Participants were asked whether the assessment strategies in their degree program addressed some aspects of e-health. 30.5% of participants agreed and 43.8% disagreed. •

Assessment addresses aspects of e-health: “Our assessment strategies are always aligned with our learning objectives and so yes some of our strategies will address some aspects of e-health”. Medical Imaging Professional/Radiographer, P88 “Have introduced a lecture that will expand to incorporate International Dietetic Terminology which will link into using electronic case notes in the future”. Dietitian/ Nutritionist, P53 “Not a subject of an assessment item. Some components of public health essays may touch upon e-health as a component of a question or as part of an answer”. Medical Practitioner, P89

•

Assessment does not address aspects of e-health: “There are no assessment strategies in this degree program which include this area”. Massage/Musculoskeletal/Myotherapist, P6 “Our assessment strategies do not address anything related to e-health”. Psychologist, P4 “Our students use electronic records in our clinic however they are not assessed on them”. Chiropractor, P102

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

38 A. Dattakumar et al. / We Are Not Educating the Future Clinical Health Professional Workforce Adequately

The absence of assessment strategies related to e-health in these degree programs makes it hard to judge whether future health professionals have gained competence in this area.

4. Conclusion At present we cannot be confident that Australia is equipping its future clinical health professional workforce to be competent in e-health. This may come as a surprise to public and private sector investors in e-health, and indeed to the general public. Although there may be pockets of good practice within particular professions or institutions, this study has sought to take a broad view across universities, professions and levels of qualification. The current teaching and assessment of future health professionals is mostly inconsistent and informal, and as many as 80% of those surveyed lack confidence in the adequacy of e-health inclusion in their curricula. Based on these findings, more needs to be done to educate future clinicians. In order to make significant improvements to the e-health education of future clinical health professionals in Australia, considerable efforts are needed to develop ehealth expertise among academic teaching staff and to develop curriculum resources for teaching and assessing e-health competence. Accordingly, these findings are part of a larger project [10] to encourage and support Australian universities to include ehealth curriculum where it is not yet in place, and to engage in collaborative continuing improvement where it is.

References

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

[1]

Smith, S.E., Drake, L.E., Harris, J.G., Watson, K., & Pohlner, P.G., Clinical Informatics: A Workforce Priority for 21st Century Healthcare, Australian Health Review, 35 (2011), 130-135. [2] Hersh, W., The Health Information Technology Workforce: Estimations of Demands and a Framework for Requirements, Applied Clinical Informatics, 1 (2010), 197-212. [3] Eldredge, J.D., Morley, S.K., Hendrix, I.C., Carr, R.D., & Bengston, J., Library and Informatics Skills Competencies Statements from Major Health Professional Associations, Medical Reference Services Quarterly, 31 (2012), 34-44. [4] Gray, K., & Sim, J., Factors in the Development of Clinical Informatics Competence in Early Career Health Sciences Professionals in Australia: A Qualitative Study, Advances in Health Sciences Education, 16 (2010), 31-46. [5] Nagle, L. Everything I know about Informatics, I didn’t Learn in Nursing School, Nursing Leadership, 20 (2007), 22-25. [6] Gray, K., Dattakumar, A., Maeder, A., Chenery, H., Educating Future Clinicians about Clinical Informatics: A Review of Implementation and Evaluation Cases, European Journal of Biomedical Informatics, 7 (2011), 48-57. [7] Australian Bureau of Statistics, Australian and New Zealand Standard Classification of Occupations. (2009). Available from http://www.abs.gov.au/ausstats/[email protected]/mf/1220.0 [8] Dattakumar, A., Gray, K., Chenery, H., Butler-Henderson, K., & Maeder, A., E-health education for Australia’s future health professionals: a discussion paper (2012). [9] Creswell, J.W., Research Design: Qualitative, Quantitative, and Mixed Methods Approaches, Sage Publications, USA, 2009 [10] Clinical Informatics Education, Coordinated Interprofessional Curriculum Renewal for e-health capability in clinical health professional degrees. (2012). Available from http://clinicalinformaticseducation.pbworks.com

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Health Informatics: Building a Healthcare Future Through Trusted Information A.J. Maeder and F.J. Martin-Sanchez (Eds.) IOS Press, 2012 © 2012 The authors and IOS Press. All rights reserved. doi:10.3233/978-1-61499-078-9-39

39

Influencing factors for adopting personal health record (PHR)

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Karen DAY and Yulong GU National Institute for Health Innovation The University of Auckland

Abstract. Background This paper reports on a small study in which we investigated factors influencing the adoption of Personal Health Records (PHRs), with specific reference to those linked to Electronic Health Records. Long term conditions are a growing concern in healthcare delivery, especially as the need for services outstrips resources. PHRs could be useful in supporting self-care of people with long term conditions, e.g. diabetes. Method We used a Grounded Theory approach to gather initial data to inform future research. This meant that we were guided and informed by data collected.[1] The aim was to find a baseline for PHR research using the Technology Acceptance Model [2], after which emerging topics will be pursued with constant comparison to this first study and subsequent studies. We interviewed ten patients from a Wellington primary care practice who had been using a PHR linked to their doctor’s Practice Management System, and observed them using and navigating the software at home. Results Three groups of factors appeared to influence the participants’ use of the PHR. (1) Their perception that convenience, time saving (for them, their GP’s practice personnel and their GP), efficiency and effectiveness of care is a result of using the PHR. (2) Computer and health literacy contribute to being able to effectively use the PHR. PHR usage impacts positively on the relationship with their doctor and vice versa, and improves their ability to navigate the health system. (3) It is not clear how PHR-related services are paid for, who pays and under what circumstances. Discussion These findings are in keeping with those of Santana et al[3] who indicate that patients are interested in viewing and using laboratory results, medicines and diagnosis lists, and ordering repeat prescriptions and corresponding with their clinicians. Literacy does matter, especially in terms of PHRs providing a transparent ‘source of truth’ and memory aid about medications, laboratory results and diagnoses. The combination of computer and health literacy requires further research. The cost of incorporating PHRs into healthcare services needs to be articulated so that it makes sense in the current capitation funding model. Limitations of this research include the small sample size, bias based on small context and self-selection into the interview cohort. In conclusion, three groups of influencing factors have been elicited from the interviews and observations to indicate potential acceptance and adoption of PHRs in self-care of long term conditions. The next step is to examine the social situatedness of long term conditions and how PHRs support self-care. Keywords. personal health record, self-care, chronic illness management

Introduction More and more people are developing long-term conditions. As they get older, they develop combinations of conditions that result in complex treatment plans, e.g. diabetes. As clinical resources diminish and the rate of long term conditions rises, self-care

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

40

K. Day and Y. Gu / Influencing Factors for Adopting Personal Health Record (PHR)

becomes essential for improving health outcomes. It is no longer enough for people to comply with their medication prescriptions – they need to co-create with their clinicians care plans that match their needs and abilities to care for themselves. This self-care relies on patients being “informed and activated” and clinicians being “prepared and proactive.”[4] PHRs offer an opportunity to leverage self-care in terms of being “informed and activated” by providing access to clinical data held in their Electronic Health Records (EHRs) by their clinicians, and being able to correspond with their clinicians about aspects of their health. The potential of PHRs lies in strengthening a patient’s ability to manage their own health and possibly achieve cost savings in healthcare delivery.[5] PHRs have been used to support patient-clinician communication and provide convenient record access. PHRs for management of long term conditions such as diabetes are designed with an objective to support self-care.[6] Adoption of PHRs, especially in the context of the New Zealand health system, is seldom reported. Some people use computers more readily and effectively than others. Regardless of how willing people are to adopt technology, certain factors and beliefs influence that adoption. The Technology Acceptance Model (TAM) is a useful way to identify people’s propensity to use technology.[2] This model examines perceived usefulness, perceived ease of use, belief indicators that support use, and teases out factors that predispose people to using technology. Increasingly New Zealand GPs will be able to offer their patients use of PHRs with the move towards shared care functionality outlined in the National Health IT Plan.[7] In New Zealand, almost all primary care clinicians use electronic Practice Management Systems (PMS) for clinical, financial and administrative purposes.[8] The software in some PMS products contains additional functionality that supports a linked PHR, giving patients access to aspects of the EHR kept by their doctor, and some functionality that supports interaction with their doctor. General Practitioners (GPs, also known as primary care physicians), invite patients to use the PHR, activating the software in their PMS. The patient signs on at home via the Internet, accepts the electronic invitation, and is able to view laboratory results, diagnoses, immunisations, and medicines lists amongst other things. Although this functionality is available to most GPs via their PMS systems, adoption is not wide-spread in New Zealand. We conducted a small exploratory study to see how TAM could be used to help doctors and nurses of people with long term conditions identify patients most likely to use their PHR when invited to do so. Our research questions were “What factors influence PHR use?” and “Do perceptions of ease of use influence patients’ engagement with the software?” and “What is it about the available software that is considered useful by patients?” The focus of this report is the first question: What factors influence PHR use?

1. Method We used a Grounded Theory approach to gather initial data to inform future research. This meant that we were guided and informed by the data collected.[1] The aim was to find a baseline for PHR research using TAM, after which emerging topics would be pursued with constant comparison to this first study and subsequent studies. We approached a GP in Wellington who was known to use a PMS-linked PHR in his every day practice, to participate in our research. The GP and colleagues in that

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

K. Day and Y. Gu / Influencing Factors for Adopting Personal Health Record (PHR)

41

practice had extended a PHR invitation to patients two years before. Ethics approval was granted by the Central Ethics Committee (reference CEN/11/EXP/019). The GP sent an email via the PHR to all 118 patients who were using the PHR, inviting them to participate in our research. To be included in the study, the patients had to have one or more long term conditions, be 18 years or older and able to consent, and have used their PHR during the last two years. Twelve patients consented to participate although one later withdrew due to health issues, and another was subsequently unavailable. We interviewed ten participants in their homes or workplace (whichever was convenient for the participant) using the interview questions in Table 1. We also observed them using their PHR. Table 1. Interview questions and prompts Interview question

Prompts

The person’s experience with the healthcare system

Complexity, navigating the health system, good and bad experiences, first experience, memories of diagnosis of chronic condition, stories about using healthcare services (planned and unplanned)

Computer competencies

Access to computers, using the Internet, purposeful use of computers/mobile phones/Internet services. This included observing the person using the PHR and discussing their experiences with it.

Stories and insights indicating perceptions of usefulness as per the TAM [2]

Intention to use, actual use, quality of the PHR, content, perceived usefulness, and ease of use (e.g. functionality complexity, literacy, software glitches).

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

The interviews were recorded, transcribed and analysed by the authors. The data were analysed using the Grounded Theory data analysis process: line by line coding forming categories, followed by axial coding (finding relationships between the categories) resulting in themes[1]. Field notes on the observation of participants using their PHRs and memos about the researcher’s assessment of the interview and observations were also included in the analysis.

2. Findings 2.1. Why use (or not use) certain PHR functions? The participants’ perceived usefulness of the PHR was strong. Participants repeatedly commented on their PHR being “convenient,” “open,” and “helpful.” A theme that emerged from the interview data is a belief that using the PHR saves time for the GP’s practice staff and clinicians. They talked about how the simplicity of ordering repeat prescriptions and having laboratory results available online saves time for both clinician and patient. One participant described what happens in the absence of the online facility as complex, time consuming and tiresome. “… if I have to ring up the nurse to get a prescription, she’s got to write it down. And that’s got to be transferred to [GP]. … Whereas now if I want a repeat prescription I can just go on [PHR] and order it, and then it’s ready for me. [GP] will send me a quick reply saying done. So I know it’s ready to collect. … It just makes it easy for [GP] as well.”

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

42

K. Day and Y. Gu / Influencing Factors for Adopting Personal Health Record (PHR)

Participants added that the PHR contributes to efficiency and effectiveness, saying, “…it’s more effective for them and it’s certainly efficient for them.” Participants described their GP as someone who is busy, who likes technology and easily incorporates it into his everyday work style. This means that they may be inclined to use the PHR to communicate, as indicated by a participant who said, ‘I don’t want to be crashing into his day and ringing him directly on his phone when he’s a busy guy.’ Some PHR functions were not used, e.g. ‘Goal Tracking,’ ‘Health Indicator,’ ‘MedicAlert,’ ‘Calendar’ and ‘Journal.’ They did not perceive these functions to be useful. This is exacerbated by some aspects of the interface being difficult to use, e.g. copying and pasting information from MS Word to the Journal. Most participants found it easier to continue recording their monitoring data manually. Most participants commented that their PHR in general is easy to learn and use, but that navigation in general was not intuitive, e.g., saying that the ‘Inbox’ interface was clumsy. They recommended potential improvements for the tool. x x x

The correspondence functionality could allow an automatic “out of office” reply from GP so that patients would know that their messages have not been picked up if and when their clinician is away for a period of time. The PHR could allow patients to “remember” their “preferred GP” as a default setting to improve convenience and ease of use. A “search” function could be added across a patient’s health information to facilitate information seeking and use in self-care within the PHR.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

2.2. Does user literacy matter in PHR adoption? Although it was not our intention to measure literacy (health or computer), the participants gave us some insight into their perceived literacy requirements for PHR use. All except one participant had access to computers at home. They all had phones (landlines) and access to broadband, with the exception of one person who uses dial-up, saying that she couldn’t afford broadband. Three women and seven men were interviewed. Their ages ranged from 35 to 79: one person was in their 30s, one in their 40s, two in their 50s, one in their 60s and five were in their 70s. Six people were retired, one was self-employed (although over retirement age), two worked in jobs that relied on computers and one worked in social services. Participants referred to phone calls to and from their GP to ask and answer questions, refine a prescription, and discuss follow up from a laboratory result or other medical activity. Patient initiated correspondence within the PHR was used for getting advice or signalling that there may be a problem needing further attention. The GP also encourages information seeking behaviour, as one participant put it: “He [GP] does quite often give me, you know, if I've got something, he’ll say ‘oh go and look that up’, so I do.” The PHR functionality was considered to be an adjunct to the doctor-patient relationship, e.g. using laboratory results and correspondence via the PHR, the GP supervised three participants using Warfarin in conjunction with the patient attending routine appointments with the doctor. The PHR also became a memory aid that contains information that can be referred to repeatedly as a person gains understanding and extends their capacity to act. At times it is used as a ‘source of truth’ for drug names, diagnosis names, and laboratory results. It also provided participants with an

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

K. Day and Y. Gu / Influencing Factors for Adopting Personal Health Record (PHR)

43

audit trail for prescription orders and correspondence with their GP. From this ‘source of truth’ the participants are able to order repeat prescriptions, and prepare for face to face appointments with their GP and/or a specialist. 2.3. What about cost? When asked if they were prepared to pay for their PHR two said they had paid for some services, e.g. email consultation. Most participants showed a positive attitude towards paying for IT-enabled healthcare services in the future, as long as the fee was comparable with conventional doctor visits. One participant suggested the model used in Internet banking, saying patients could get a fixed number of free electronic consultations and pay for any extra.

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

3. Discussion These findings are consistent with evidence about benefits of patients’ access to their EHRs, such as improving care and safety, encouraging open and honest clinicianpatient relationships, assisting memory and treatment concordance, and supporting positive health behaviour.[6] The preference for manual recording of monitoring data when PHR use becomes difficult is disappointing since chronic conditions such as diabetes are often complex and require strict monitoring.[9] Failure to support patients with a monitor log appears to limit the usefulness of PHRs. The mixture of computer and health literacy raises the need for a new understanding of literacy to be useful in the PHR context. Digital (or computer) literacy is the ability to critically appraise information on the Internet and use it for decision making.[10] Familiarity with computers was considered necessary but not sufficient for effective decision making in self-care. Health literacy is the ability to find, process and use health information in order to make health-related decisions appropriately[11]. Health literacy is an important factor in the doctor-patient relationship [11], and this was evident in the interview discussions. Older people use computers if the design meets their needs and they are able to afford the technology [12]. It is possible that the bias toward people in their 70s in our study represents people who can afford and use computers instead of being representative of the general population. Healthcare in New Zealand is free at the point of care except for co-payments for primary care appointments and some services, e.g. repeat prescription orders. Most primary care funding is modelled on capitation [13] making it cheaper to care for people remotely with fewer clinic visits. Future research could examine ways to articulate this ‘invisible’ incentive wrapped inside the capitation model to encourage GPs to use PHRs as a clinical tool. There were some limitations to this study. This study relies on a small number of self-selected people who chose to use a PHR, and responded to our invitation to participate in research. Due to this self-selection, and the small number of people who had an opportunity to try out the PHR at all, we did not at this early stage recruit patients who had tried and abandoned PHRs, nor those who elected to not use it at all. These groups should be included in future research on larger cohorts of PHR users. All participants were using one PHR system and were patients of one PHR-supportive GP.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

44

K. Day and Y. Gu / Influencing Factors for Adopting Personal Health Record (PHR)

4. Conclusion PHRs have the potential to support self-care of long term conditions, particularly through supporting communication with clinicians and providing access to their clinical information. Factors that may influence PHR adoption include patient perceptions about time savings for GP practice personnel and clinicians, personal literacy capabilities, and the cost of incorporating PHR services in healthcare services. The participants suggested improvements to the existing PHR functions available to them. This is the first of a series of grounded theory studies. Future research themes emerging from this study include the social situatedness of long-term conditions in which PHRs play a supporting role in self-care; computer and health literacy; the role of PHRs in supporting the clinician-patient relationship; the effect of PHR functionality on self-care, and health outcomes. Future research should also include patients who have never used the PHR, and those who have or would like to use it for purposes other than for self-care of a long term condition.

References [1] [2] [3] [4] [5] [6] [7]

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

[8] [9] [10] [11] [12]

[13]

Charmaz, K., Constructing grounded theory. 2006, London: Sage Publications. Davis, F.D., Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 1989. 13(3): p. 319-340. Santana, S., et al., Online communication between doctors and patients in Europe: status and perspectives. J Med Internet Res, 2010. 12(2): p. e20. Wagner, E.H., et al., A survey of leading chronic disease management programs: Are they consistent with the literature? Managed Care Quarterly, 1999. 7(3): p. 56-66. Detmer, D., et al., Integrated personal health records: transformative tools for consumer-centric care. BMC Med Inform Decis Mak, 2008. 8: p. 45. Mulvaney, S.A., et al., An Internet-Based Program to Improve Self-Management in Adolescents with Type 1 Diabetes. Diabetes Care, 2010. 33(3): p. 602-604. National Health IT Board, National Health IT Plan. Enabling an integrated healthcare model. 2010, National Health IT Board: Wellington. Schoen, C., et al., A survey of primary care physicians in eleven countries, 2009: Perspectives on care, costs and experiences. Health Affairs - Web Exclusive, 2009: p. w1171 - w1183. Glasgow, R.E., M. Peeples, and S.E. Skovlund, Where is the patient in diabetes performance measures? Diabetes Care, 2008. 31(5): p. 1046 - 1051. Brown, C.A. and R. Dickson, Healthcare students' e-literacy skills. Journal of Allied Health, 2010. 39(3): p. 179-184. McCray, A.T., Promoting health literacy. Journal of the American Informatics Association, 2005. 12: p. 152 - 163. Festervand, T.A., D.B. Meinert, and S.J. Vitell, Older adults' attitudes toward and adoption of personal computers and computer-based lifestyle assistance. Journal of Applied Business Research, 2011. 10(2): p. 13 - 22. Cumming, J., N. Mays, and B. Gribben (2008) Reforming pimary health care: is New Zealand's primary health care strategy achieving its early goals? Australia and New Zealand Health Policy 5, 11 DOI: 10.1186/1743-8462-5-24.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

Health Informatics: Building a Healthcare Future Through Trusted Information A.J. Maeder and F.J. Martin-Sanchez (Eds.) IOS Press, 2012 © 2012 The authors and IOS Press. All rights reserved. doi:10.3233/978-1-61499-078-9-45

45

Exploring the role of pathology test results in the prediction of remaining days of hospitalisation Blanca GALLEGO, Oscar PEREZ-CONCHA, Frank LIN and Enrico COIERA Centre for Health Informatics Australian Institute of Health Innovation

Abstract. Accurate prediction of discharge time and identification of patients at risk of extended length of stay (LOS) can facilitate discharge planning and positively impact both the patient and the hospital in a variety of ways. To date, however, most studies only focus on the prediction of the overall LOS, which is generally estimated at admission time to hospital, emergency department or intensive care unit. This paper explores whether individual laboratory results can improve predictions of time of discharge as the tests become available. This study suggests that there is a statistically significant relationship between individual test results and remaining days in hospital and that there is a trend towards better estimates as more consecutive tests are taken into consideration. Their effect on the estimate of discharge time is generally weak. Further work integrating groups of test results into a more sophisticated dynamical model is required. Keywords. length of stay, laboratory results, pathology results, prediction

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

Introduction Hospital length of stay (LOS) is a useful indicator of healthcare activity and management, and a major determinant of cost. Monitoring LOS is also important for personalised health care delivery and for the efficient management of healthcare resources [1-2]. Previous studies showed that the inclusion of laboratory and comorbidity data collected during routine practice, in addition to administrative data improved the predictive ability of regression models of LOS [3]. Nevertheless, prediction was usually carried out at one single point during the hospital stay, generally at time of admission to hospital or at admission to specific units such as intensive care units or emergency departments [4]. The aim of this study is to provide a preliminary exploration of the role of 10 commonly performed pathology test results in the estimation of remaining days in hospital (hereafter RDH) during a hospital admission. For this purpose a regression model of RDH was built for each of the test types and its performance was analysed as up to 3 consecutive results were added to the model.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

46

B. Gallego et al. / Exploring the Role of Pathology Test Results in the Prediction of RDH

1. Methods 1.1. Study Population The data used in this study consisted of 16,171 consecutive admissions to a tertiary hospital in Sydney, Australia, between February and June 2011. It contained the timestamps of when the tests were performed, the type of test, the test result, and a flag indicating whether the test was outside the reference range. Data linkage was performed to match laboratory data to administrative data, where the dates of admission and discharge, the Australian-refined diagnosis-related group (AR-DRG) code, and the age of the patient were extracted. The patient’s vital status at the time of discharge was also collected. Patients who died during hospitalisation were excluded from the analysis. This study focuses on the following 10 commonly performed pathology tests:

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

Activated Partial Thromboplastin time (APTT) Serum Creatinine Concentration (CREAT) Haemoglobin Concentration (HAEM) Random Glucose Concentration (GLUC) International Normalized Ratio (INR) Oxygen Saturation (O2SAT) Platelet Count (PLAT) Serum Sodium levels (SODI) Troponin T (TROP) White Blood Cell Count (WBC)

The results of the tests were categorised by classifying them into three levels against the reference range: “L” for low, “N” for normal and “H” for high. A special group “Z” indicates that no test of the type under consideration was performed during the admission. For example if a patient was tested four times for platelet counts during hospitalisation, with high counts during the first three times and a normal level in the last, the output would be “H,H,H,N”. In addition, “Z” would indicate that no platelet count was performed. 1.2. Regression Methods The regression models of RDH consisted of linear mixed models with dummy variables of the categorical test results and patient age as fixed-effect variables. Groups of combinations of DRGs and days already in hospital were included as random intercepts. The model parameters were estimated using proc mixed in SAS (Statistical Analysis System) 9.2 [5]. To examine the effect of adding new test results to the model, three regression experiments were performed. In the first experiment only the first test available was included (or the category “Z” was assigned if no test was performed). Here, RDH represented the days remaining from the day the test was performed (or LOS if no test was performed). In the second and third experiments up to 2 and 3 consecutive test results were included. Here, RDH corresponded to the days remaining from the day the last test was performed.

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

B. Gallego et al. / Exploring the Role of Pathology Test Results in the Prediction of RDH

47

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

2. Results There was a statistically significant relationship between RDH and all of the 10 types of pathology results included in this study. In all cases the fact that a test was performed (regardless of the outcome) was associated with higher RDH values. Similarly but to a lesser extent, it was often the case that higher number of consecutive tests was associated with higher RDH values. Table 1 shows the regression coefficients for APTT and WBC. The coefficients of the regression model have plausible interpretations clinically. For example, the positive association between RDH and the high values of APTT correspond to patients treated with intravenous heparin who require frequent inpatient therapeutic monitoring during hospitalisation (and hence more RDH). This is in contrast with patients with low or normal APTT values, which may include patients for whom the test was ordered as part of baseline screening, or at the weaning stages of heparin therapy (and hence fewer RDH). For WBC count, having a high WBC value corresponded to a greater slope (H vs. L and N groups, ș=2.24 vs. 1.72 and 1.76 respectively). This finding is consistent with clinical practice as leucocytosis is a marker generally used to indicate the severity of inflammation and infection. In general, the separate predictive power of these pathology tests was weak, and a significant heterogeneity was observed within groups of admissions with the same test result. This is a expected result since there are a variety of missing factors, not included in the current model, with the potential to control LOS. As an illustrative example, Figure 1 shows the observed and modelled RDH values considering the first test for APTT, for patients in their first day of hospitalisation with a DRG of disorders of pancreas except for malignancy. For each of the 10 selected pathology tests, this study compared the Root Mean Square Error (RMSE) associated with: a prediction given by the mean over all the admissions with the same DRG and the same number of days already in hospital; and a prediction given by the regression model taken into account up to 1, 2 or 3 consecutive tests. The results are displayed in Table 2. Including information about the test results represents a small but consistent improvement on the estimation of RDH. Adding an extra consecutive test in the regression models is always associated with a smaller RMSE. When analysing the performance of the model as a function of the values of RDH, accuracy was found (not unexpectedly) to decrease as RDH increases (see Figure 2 for an example with APTT and WBC).

3. Discussion With the growing adoption of electronic health records, an opportunity exists to enhance the quality of healthcare by learning from data. Pathology tests are routinely collected in electronic format in most major hospitals in Australia and are amenable to quantitative analysis. This study explored the power of sequences of laboratory test results to predict remaining days in hospital using data from 16,171 admissions to a tertiary hospital in Sydney, Australia, between February and June 2011. Understanding the value of electronic and routinely collected clinical data It was found that the impact of each test in predicting remaining days in hospital (RDH) is statistically significant but small and that the trend is towards improvement

Health Informatics: Building a Healthcare Future Through Trusted Information : Selected Papers from the 20th Australian National

48

B. Gallego et al. / Exploring the Role of Pathology Test Results in the Prediction of RDH

when more tests are added to the sequence (at least for up to 3). It was also noted that the predictability of RDH decreases the further away the prediction date is from the discharge date. As illustrated in the examples in the results section, many patterns of laboratory results can be explained by the underlying processes of healthcare, the test ordering behaviour of clinician, and the distinctly stages of the journey of admission. Thus, it is plausible to utilise these patterns of laboratory tests to implicitly predict useful health services parameters such as RDH. In addition, this method is also applicable to the identification of unrecognised patterns which may contribute to unexpected RDH ranges. This study represents a preliminary exploration on the role of pathology test results in the prediction of remaining days of hospitalisation, and has several limitations. For example, the linear character of the regression should be relaxed to include more general functions; other covariates responsible for LOS should be included; and more than 3 consecutive tests as well as numeric tests results instead of categorical ranges could be considered. Ultimately, a more sophisticated model that integrates various selected tests results into one prediction of RDH, and a proper out of sample model evaluation is required and left for future work.

4. Funding and Ethics

Copyright © 2012. IOS Press, Incorporated. All rights reserved.

This work was funded by National Health and Medical Research Council (NHMRC) Program Grant 568612 and approved by the Hospital Ethics Committee and the UNSW Human Research Ethics Committee. Its contents are the responsibility of the authors and their institutions and do not reflect the views of NHMRC. The authors have no conflicts of interest to disclose. Table 1. Regression coefficients for RDH vs. APTT/WBC have been included). APTT TESTS Effect β S.E. P value UP TO 2 Intercept 5.00 0.32