Health Care Costs: Causes, Effects and Control [1 ed.] 9781617285165, 9781604569766

Citation preview

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved. Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved. Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Health Care Issues, Costs and Access

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

HEALTH CARE COSTS: CAUSES, EFFECTS AND CONTROL

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in rendering legal, medical or any other professional services.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

HEALTH CARE ISSUES, COSTS AND ACCESS The Health Care Financial Crisis: Strategies for Overcoming an “Unholy Trinity" Cal Clark and Rene McEldowney (Editors) 2001. ISBN 1-56072-924-4 Health Care Crisis in America James B. Prince (Editor) 2006. ISBN 1-59454-698-4 A New Epidemic: Harm in Health Care-How to make Rational Decisions about Medical and Surgical Treatment Aage R. Moller 2007. ISBN 1-60021-884-9 Decision Making in Medicine and Health Care Partricia C. Tolana (Editor) 2008. ISBN 1-60021-870-9 Social Sciences in Health Care and Medicine Janet B. Garner and Thelma C. Christiansen (Editor) 2008. ISBN 978-1-60456-286-6

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Health Care Policies Linda A. Bartlette and Ida F. Lawson (Editor) 2008. ISBN 978-1-60456-352-8 Handbook of Stress and Burnout in Health Care Jonathon R.B. Halbesleben (Editor) 2008. ISBN 978-1-60456-500-3 Health Care Costs: Outlook and Options Raymond W. Inhurst (Editor) 2009. ISBN 978-1-60692-151-7

Medicare Payment Policies to Physicians Katherine V. Bergen (Editor) 2009. ISBN 978-1-60692-131-4 Comparative Effectiveness of Medical Treatments Peter Villa and Sophia Brun 2009 ISBN: 978-1-60741-109-3 Health Care Costs: Causes, Effects and Control Bernice R. Hofmann (Editor) 2009. ISBN 978-1-60456-976-6

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Health Care Issues, Costs and Access

HEALTH CARE COSTS: CAUSES, EFFECTS AND CONTROL

BERNICE R. HOFMANN

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

EDITORS

Nova Science Publishers, Inc. New York

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2009 by Nova Science Publishers, Inc. 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: electronic, electrostatic, magnetic, tape, mechanical photocopying, recording or otherwise without the written permission of the Publisher. For permission to use material from this book please contact us: Telephone 631-231-7269; Fax 631-231-8175 Web Site: http://www.novapublishers.com NOTICE TO THE READER The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained in this book. The Publisher shall not be liable for any special, consequential, or exemplary damages resulting, in whole or in part, from the readers’ use of, or reliance upon, this material. Any parts of this book based on government reports are so indicated and copyright is claimed for those parts to the extent applicable to compilations of such works.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Independent verification should be sought for any data, advice or recommendations contained in this book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to persons or property arising from any methods, products, instructions, ideas or otherwise contained in this publication. This publication is designed to provide accurate and authoritative information with regard to the subject matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in rendering legal or any other professional services. If legal or any other expert assistance is required, the services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A COMMITTEE OF PUBLISHERS. LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA Health care costs : causes, effects & control / [edited by] Bernice R. Hofmann. p.; cm. Includes bibliographical references and index. ISBN  (H%RRN) 1. Medical economics. 2. Medical care, Cost of. I. Hofmann, Bernice R. [DNLM: 1. Health Care Costs. 2. Delivery of Health Care--economics. 3. Quality of Health Care--economics. W 74.1 H4338 2008] RA410.H38 2008 338.4'33621--dc22 2008035991

Published by Nova Science Publishers, Inc. Ô New York Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

CONTENTS Preface

ix

Expert Commentaries:

1

Expert Commentary A Health Technology: Moral Hazard, Inefficiencies, and Solutions Read G. Pierce and Kevin J. Bozic

3

Expert Commentary B The Future of Health-care Quality in the United States Brandon Roberts

9

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Short Communications: Short Communication A The Elasticity Of Physician Supply And Medical Malpractice Litigation in The United States Brandon Roberts and Euel Elliott Short Communication B How do the Drug Ads and other Factors Affect the Way Dentists Prescribe Medications and Has Prescription Medicine Increased the Overall Cost of Health Care? Kim L. Capehart Short Communication C The Affects of Contemporary and Alternative Medicine on Healthcare and Dentistry Kim L. Capehart Short Communication D America’s Weightiest Health Issue: Not Obesity, but Bloated Health Care Expenses George Everett

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

17

19

33

39

43

vi

Contents

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Research and Review Studies: Analysis of The Number of Days Required for Full-time Care Chapter 1 before Death and the Cost of Care in the Last Month of Life among the Chinese Elderly Danan Gu, Donghui Gu and Jie Zhou Chapter 2

Costing of Primary Care in Developing Countries: How much Health Can we Buy for a Few Dollars? Steffen Flessa and Paul Marschall

Chapter 3

True Cost of Transfusion and Measures to Tame it Aryeh Shander, Axel Hofmann and Mazyar Javidroozi

Chapter 4

Synthesis of "Statistical Innovations for Cost-Effectiveness Analysis" Translating Research into Policy and Practice (TRIPP) Melford J. Henderson

49 51

81 101

121

Chapter 5

Costs of Epidemiological Changes in Chronic Diseases in Mexico Armando Arredondo, Emanuel Orozco and Edson Servan

Chapter 6

Deficit and Health Reform – Two Different Phenomena? Z. Darmopilova and Z. Zigova

171

Chapter 7

Screening Colonoscopy in Germany. A Cost-saving Analysis Andreas Sieg and Hermann Brenner

195

Chapter 8

Costs of Alcohol Misuse in England Abiodun Olukoga, Graham Lister, Richard Fordham, Miranda Mugford, Edward Wilson and Dominic McVey

Chapter 9

A Fully Bayesian Cost--Effectiveness Analysis Using Conditionally Specified Prior Distributions M.Martel, M.A.Negrín andF.J. Vázquez–Polo

Chapter 10

Medication Management and the Older Person Maggi Banning

Chapter 11

Training Needs of Healthcare Professionals Working with Breastfeeding Mothers in England and in Poland Joanna Kosmala-Anderson, Louise Wallace, Susan Law and Orla Dunn

Chapter 12

The Design and Evaluation of a Radio Frequency Identification (RFID) Enabled Inpatient Safety Management System Chuan-Jun Su and Ta-Wei Chu

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

153

209 227 243

257

283

Contents Chapter 13

A Note on Estimating a Mean Cost of Hospital Stay with Incomplete Information Isabella Locatelli and Alfio Marazzi

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Index

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

vii

309 319

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved. Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

PREFACE It is difficult to find an area of such unyielding cost increases as health care except perhaps energy costs. These costs and the underlying health care systems continue to constitute major issues around the world. This new book tackles these tough issues head on from multiple perspectives around the world. Expert Commentary A - Rapidly rising health care costs in developed nations continue to be a major policy issue, and health technology (HT) adoption and utilization is frequently singled out as a primary driver of cost inflation. In the U.S., health systems and physicians enthusiastically acquire HT to further the scope and quality of patient care services, compete against other providers for patients, and enhance the financial viability of their practice. Patients pursue expensive HT in response to information asymmetry, which leads them to associate high-tech care with quality, and inefficient insurance coverage that shelters them from the true costs of their care. The resulting moral hazard threatens to reduce further technological innovation, as third party payers seek to control health care spending. To date, insurers and policymaker have yet to find an effective or popular strategy for encouraging more rational HT use. The authors propose four mechanisms for realigning incentives and risks to promote value-based technology adoption and utilization by health systems, physicians, and patients. Expert Commentary B - The issue of health-care quality in the United States has been of considerable concern for over a decade. An alarming report issued in November of 1999 by the Institute of Medicine (IOM) entitled To Err is Human: Building a Safer Health System catapulted this issue to the forefront during the Clinton administration. The IOM report contended that there were potentially 100,000 deaths annually in U.S. hospitals that were due to medical errors. This figure was staggering and suggested that deaths from “medical accidents” in hospitals exceeded the number of deaths from automobile accidents, breast cancer, and AIDS combined, and was over six times the number of murders committed nationwide. The report’s release prompted a number of ongoing public and private initiatives to address deficiencies in health care. Short Communication A - The issue of medical malpractice reform has occupied state legislatures for the last several years, due to the concerns expressed by physicians and others that the costs of malpractice litigation are harming the delivery of medical services in the United States. Using data from the 1997-2001 time period, the impact of malpractice litigation on physician supply in the U.S. is estimated. The results indicate that medical malpractice litigation is a significant factor in reducing physician supply in the U.S. More

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

x

Bernice R. Hofmann

specifically, the authors find that laws in those jurisdictions that have direct influences on medical liability have more impact than those with an indirect effect. The authors conclude that the concerns over medical malpractice litigation are not misplaced or exaggerated, and that policymakers should strongly consider those policy changes that exert a more direct influence on litigation. Short Communication B - Dentistry is affected tremendously by drug ads, geriatric patients, and the costs of R&D of pharmaceutical products. In the overall picture, prescription drugs aren’t a major reason why health care spending is up. It is a significant cost in the overall cost of health care in the United States but not to the extent that most would believe. We all prescribe medications daily which are advertised and most see geriatric patients. Overall, I feel that dentists prescribe based on what is best for the patient and not what the patient requests, but dentistry is not immune to the fact that patients will ask daily for medications that they have seen in magazines, billboards, or television commercials. As consumers (patients) become more electronic savvy, there will be an increase in requests in the future as spending is increasing and more are internet cognizant as well as watching television. As dentists, we need to understand these factors and prepare as they come in to our offices. There are myriad of factors to cause the feeling that prescription medicines are increasing the cost of health care which include direct-to-consumer advertising, the aging population, and the cost of research and development accompanied with drug reps affect are all integral factors that have influenced this increase. There is no easy “fix” and the situation is very complicated causing a multidisciplinary effort. Having all these great minds is advantageous but it also causes different opinions and different ideals which can slow the process down. Short Communication C - Allopathic medicine has been the stable force in medical philosophy for many years. We have seen television programs with therapies from a plant in China to an herb in Brazil. It has been within the past few decades have osteopathic, acupuncture, chiropractic, and other forms of alternative therapies have become more prevalent in health care today. So how does this affect us as “oral physicians?” Well, visit the local GNC store or drug section of grocery stores and you will see the impact of contemporary and alternative medicine (CAM) therapies. Many products market towards those that utilize this therapy with labels stating herbal, or aromatherapy just to name a couple. The claims by these products range any where from skin rejuvenation to anything short of curing arthritis. So does CAM affect dentistry? It affects us daily. How many times do you hear a patient tell you that they had tempor-mandibular joint (TMJ) problems and their chiropractor fixed the problem or an herbal supplement alleviated the pain of their jaws or teeth or had a massage therapist fix their myofacial pain. It is estimated that Americans made more than 600 million visits to CAM practitioners in 1997, compared with approximately 400 million visits to all primary care physicians during the same year. CAM professional services exceeded $21 billion with approximately 60% paid out-of-pocket.. So how does this affect healthcare and dentistry? Whether you believe in alternative medicine treatment modalities or not, legislation is allocating significant sums of money to National Institute of Health for research. The government’s National Institute of Health (NIH) National Center for Contemporary and Alternative Medicine (CAM) budgeted $68.4 million for fiscal year 2000 with $72.4 million for 2001 and $110 million for 2002. This amount is compared to its original allocation of $2 million in 1992 when the center opened as Office of Alternative Medicine. You may say that is the government but they aren’t teaching it in medical schools.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Preface

xi

Well according to the Journal of American Medical Association (JAMA) in 1998, 75 out of 117 medical schools include some form of training in alternative medicine. Why is all this important for dentists? It is very important for us to recognize that many patients are looking for alternative therapies and medical schools and the government are also recognizing the trend. I want to make sure that we, as dentists, also recognize and not fall behind and know why CAM affects us as dentists. Short Communication D - The presidential campaign of 2008 has reawakened discussion of the rapidly rising costs of the United States of America (USA) health care system. Most of the political sparring has focused on how to cover the roughly 15% of Americans who are uninsured rather than how to deal with the far more troublesome and vexing problems of overwhelming total medical expenditures. After briefly stabilizing in the 1990’s, health care costs have soared to 16% of Gross Domestic Product (GDP) and are projected to approach 20% of GDP by 2017. If the general trends of the last four decades continue, health expenditures will grow by 2 – 3 % of GDP per decade. Because the baby boom generation is nearing Medicare eligibility, federal outlays for medical costs are expected to grow exponentially. The Medicare Trustees recently reported that $432 billion was spent by Medicare in 2007 and that Medicare expects to owe $36 trillion over the next 75 years. While most developed countries have also faced rising health care costs, the USA has, by far, the highest costs in the world, whether measured by percent of GDP or by per capita spending. Furthermore, nearly all other developed countries have universal or nearly universal health coverage. If the uninsured in the USA are covered in the near future, an additional spike in total health care costs is likely but with an improvement in subsequent health measures for the newly covered persons. Chapter 1 - A growing number of studies in western nations have found that an important component in improving the quality of end of life is to meet the care needs of individuals approaching death. Information on the cost of these care needs would be useful in determining how to reduce the burden on family and society, but such studies are near-absent in developing countries. This chapter provides an analysis of care needs and expenses before death in China, where the proportion of the elderly population will rapidly increase in the next few decades. Based on the fourth wave of the Chinese Longitudinal Healthy Longevity Survey (CLHLS) in 2005, the authors estimate the number of days needed for full-time care before death and the care costs in the last month of life among the deceased elderly aged 65 and over by using sequential models in a multi-level context. The CLHLS, a unique specially designed nationwide survey, included data on care needs before death and expenditure in 2005 for those respondents who died between the 2002 and 2005 waves in randomly selected half of counties and cities in 22 of 31 provinces in mainland China. Data on the number of days required for full-time care and care costs in the last month of life were obtained from 5,729 deceased persons in 774 counties by asking their next-of-kin. Both the number of days requiring full-time care and care expenses are re-classified into eight categories to best describe the data according to their distributions. A two-step estimation approach is applied to estimate the days and costs. Multi-level multi-nominal logit regressions are first applied to estimate the membership probability of each group for each study variable. These probabilities are multiplied by the mean of each corresponding group of days or costs to estimate the weighted average needs or costs. Three sequential models are employed to investigate how individual characteristics and community level factors are associated with these estimates. Model I controls for age, gender, ethnicity, urban/rural residence, education,

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

xii

Bernice R. Hofmann

family economic condition, personal economic independence, accessibility to healthcare plus community socioeconomic development factors measured by education, per capita GDP, and number of hospital beds per thousand population. Model II further controls for proximity to children, marital status, religious participation, health practices (smoking, alcohol use, and exercise), life satisfaction, and psychological disposition. Model III additionally adjusts for overall health condition measured by the frailty index. Our results show that the average number of days required for full-time care before death for a Chinese elder is approximately 80, and this number differs significantly by sex, education, access to healthcare, proximity to children, involvement in religious activities, and baseline overall health condition. However, it is not associated with community factors. Elders who are male, educated, have easy access to healthcare, live farther from children, are religious, and are healthier tend to require fewer days of full-time care before death. Our analyses further reveal that the average payment for care in the last month of life is 3,106 RMB (approximately $383). Those who are older, female, non-Han minorities, living in rural areas, non-educated, have a poor family economic condition, economically dependent, unable to access healthcare, not currently married, and from underdeveloped communities tend to spend a smaller amount on care in the last month of life. Observed differences by sex, urban/rural residence, and proximity to children are mediated by other confounders. The authors also find nonlinear relationships between costs and overall health conditions, and between costs and the community educational level. These findings are useful for the improvement of quality of end of life care and successful aging among the Chinese elderly in the context of largely reduced family size and greatly changing household structure due to population aging and migration. Chapter 2 - A growing number of studies in western nations have found that an important component in improving the quality of end of life is to meet the care needs of individuals approaching death. Information on the cost of these care needs would be useful in determining how to reduce the burden on family and society, but such studies are near-absent in developing countries. This chapter provides an analysis of care needs and expenses before death in China, where the proportion of the elderly population will rapidly increase in the next few decades. Based on the fourth wave of the Chinese Longitudinal Healthy Longevity Survey (CLHLS) in 2005, the authors estimate the number of days needed for full-time care before death and the care costs in the last month of life among the deceased elderly aged 65 and over by using sequential models in a multi-level context. The CLHLS, a unique specially designed nationwide survey, included data on care needs before death and expenditure in 2005 for those respondents who died between the 2002 and 2005 waves in randomly selected half of counties and cities in 22 of 31 provinces in mainland China. Data on the number of days required for full-time care and care costs in the last month of life were obtained from 5,729 deceased persons in 774 counties by asking their next-of-kin. Both the number of days requiring full-time care and care expenses are re-classified into eight categories to best describe the data according to their distributions. A two-step estimation approach is applied to estimate the days and costs. Multi-level multi-nominal logit regressions are first applied to estimate the membership probability of each group for each study variable. These probabilities are multiplied by the mean of each corresponding group of days or costs to estimate the weighted average needs or costs. Three sequential models are employed to investigate how individual characteristics and community level factors are associated with these estimates. Model I controls for age, gender, ethnicity, urban/rural residence, education,

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Preface

xiii

family economic condition, personal economic independence, accessibility to healthcare plus community socioeconomic development factors measured by education, per capita GDP, and number of hospital beds per thousand population. Model II further controls for proximity to children, marital status, religious participation, health practices (smoking, alcohol use, and exercise), life satisfaction, and psychological disposition. Model III additionally adjusts for overall health condition measured by the frailty index. Our results show that the average number of days required for full-time care before death for a Chinese elder is approximately 80, and this number differs significantly by sex, education, access to healthcare, proximity to children, involvement in religious activities, and baseline overall health condition. However, it is not associated with community factors. Elders who are male, educated, have easy access to healthcare, live farther from children, are religious, and are healthier tend to require fewer days of full-time care before death. Our analyses further reveal that the average payment for care in the last month of life is 3,106 RMB (approximately $383). Those who are older, female, non-Han minorities, living in rural areas, non-educated, have a poor family economic condition, economically dependent, unable to access healthcare, not currently married, and from underdeveloped communities tend to spend a smaller amount on care in the last month of life. Observed differences by sex, urban/rural residence, and proximity to children are mediated by other confounders. The authors also find nonlinear relationships between costs and overall health conditions, and between costs and the community educational level. These findings are useful for the improvement of quality of end of life care and successful aging among the Chinese elderly in the context of largely reduced family size and greatly changing household structure due to population aging and migration. Chapter 3 - Our knowledge of the cost of primary care in developing countries is limited. In particular, only few studies calculate the actual cost of preventive and curative services in first-line facilities in rural Africa. Consequently, health care planners have an inadequate understanding of the actual total cost, the cost per inhabitant or per service unit, so that health policies are frequently based on assumptions rather than on evidence. Since 2003 the authors have developed a cost-of-illness information system for the rural health district of Nouna, Burkina Faso. In this paper the authors present an analysis of the performance of 20 public health centres and pharmacies. Average and total costs are analysed using data on capital and recurrent costs. The step-down method is used to allocate costs. In addition, an estimate of standard costs is given for full coverage of the entire population. The analysis for the year reveals a great variation between the health centers concerning cost structures, average cost and efficiency. These variations result from differing quantity or quality of the services provided in each centre. The cost recovery rates were far from reaching break even, but there was a positive profit margin at each pharmacy. The authors conclude that the total costs of primary care would hardly rise if the coverage of the population were increased. As high fees prevent many diseased from seeking medical help, the cost recovery policy at the level of primary care has to be adjusted. These results are not fully representative for the cost of primary care in rural Africa as countries and locations differ strongly on this huge continent. However, their study demonstrates that costing of primary services in developing countries is possible and carries the opportunity of evidence-based health policies. Chapter 4 - Rapid increases in health care costs continue to be of significance to public, federal and state agencies, as well as private industry. Publicly funded insurance programs

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

xiv

Bernice R. Hofmann

such as Medicare and Medicaid are continually challenged with difficult decisions in allocating health care dollars. Private industries are similarly challenged in providing adequate health care benefits to their employees. The need to contain health care costs forces us to consider which interventions produce the greatest value. Cost-effectiveness analysis (CEA) offers a structured approach for determining economic evaluations of health care programs. It can be used for optimizing health benefits from a specified health care budget, or in finding the lowest cost strategy for a specific health benefit. CEA has also been promoted as a useful tool in the effort to prioritize expenditures on health care programs. By quantifying the trade-offs between resources that need to be deployed and health benefits that accrue from use of alternative interventions, CEA offers guidance in decision-making by structuring comparisons between these interventions. The Agency for Healthcare Research and Quality has funded investigator-initiated research projects for promoting developments related to Translating Research into Policy and Practice. This article summarizes the work led by Joseph Gardiner and colleagues. The goals of this research were to develop new statistical methods that fill methodological gaps, and resolve inconsistencies in CEA. Adopting a framework in which both costs and benefits are stochastic in nature, the research team describes summary measures used in CEA, such as the cost-effectiveness ratio as functions of parameters in an underlying stochastic model. In estimation of these summary measures, the inherent variability in the estimates can be quantified. Markov models provide a probabilistic description of the evolution of events in patients through different health states. In this longitudinal framework, Gardiner et al. use stochastic models that reflect the experience of patients in sustained and changing states of health. Costs are incurred in random amounts at random points in time during the course of an intervention. By compiling these expenditure streams at the individual level into costs per unit time of sojourn in a health state, and in transition between health states, Gardiner et al estimate the net present value of all expenditures. Health outcome measures such as life expectancy and quality-adjusted life years can also be estimated. In summary, several aspects and complexities in the analyses of health care costs and outcomes are incorporated into these models. The team’s work is continuing. Their methods promise useful applications in CEA. Chapter 5 – Objective: To determine, for insured and uninsured populations of health care public institutions, the costs of health services and their financial consequences of changes in the epidemiological profile of chronic diseases (diabetes and hypertension) for the Mexican Health System. Study Design: A longitudinal analysis of costs was conducted for diabetes and hypertension, in order to analyze required adjustments to improve health system’s performance for these health needs and for both types of population. Methods: 1) Costs were estimated applying the Box-Jenkins and consensus techniques, using confidence intervals of 95% and the Box-Pierce test for 2007-2009; 2) Using previous estimations, epidemiological changes and financial consequences are inferred. Findings: It was predicted a greater increase in costs for diabetes and hypertension. Comparing financial consequences for health services required by insured and uninsured populations, the greater increase (17%) will be for insured population. The financial

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Preface

xv

requirements for both diseases will amount to 9.5% of the total budget for the uninsured and 13.5% for the insured population. Conclusions: If the risk factors and the different health care models remain as they currently are, economic impact of expected epidemiological changes on the Mexican Health System will be particularly strong. These results suggest mayor challenges for the performance capability of health system to attend these needs and their financial consequences. Chapter 6 – Objective: Analyze eventual causal relationship among financial balance of public health insurance system, necessity of health reform and success of reform enforcement. Identify key player who, due to his strategy of supporting the reform, has an influence on successful enforcement and implementation of examined reform policy in the Czech Republic. Background: The paper is inspired by longstanding unavailing attempts to enforce health reform in the Czech Republic. Besides general reasons, determining the increase in a growth rate of healthcare costs and leading to undertaking reform measures in most developed countries, in the Czech Republic additional, specific factors play a role too (such as specifics of economy in transition in early nineties, discontinuity of transformation etc.). Authors deal with two mutually conditioning issues: cumulative debt of public health insurance system and success in reform enforcement. After more than ten years of persisting deficits of public health insurance system, the balance of health insurance funds ended in surplus in 2006 (abstracting away from eventual latent deficit) due to several parametrical changes. Simultaneously, after the same period of time, the first steps of a complex health reform proposal have started to be implemented in 2007. Nevertheless, it seems that these steps do not have substantial support of many key players. The question is, firstly, whether this phenomenon results from reduction of “crises awareness” although, on the other hand, the phase of stabilization of funds may be just an appropriate time for reform implementation (supposition provided that health sector itself tends to deficits). Secondly, what are the costs and the benefits of such a radical reform especially when the last deficit was completely reduced by relatively simple administrative interventions. In accordance with [6], the authors presume that the success of reform implementation is, i.e., determined exactly by the support of at least one key player, benefiting from improving his financial, power or political position. Design and Settings: Paper briefly defines the main pillars and specifics of Czech health system, with a focus placed on public health insurance system. In outlines, it introduces the principles of health reform and examines the reactions of key players. Chapter 7 – Background: Screening colonoscopy was introduced into the National Cancer Prevention Program in Germany in 2002. The authors explored costs and savings of screening and surveillance colonoscopy to investigate whether the induced savings may compensate for the costs of screening. Methods: Study design was a model calculation based on data of a large-scale documentation of screening colonoscopy. The costs and savings of screening colonoscopy were evaluated over a defined period of 10 years. Basic data about findings, adverse effects and costs of screening colonoscopy were obtained from a large-scale online registry of 109,989 procedures and from the actual payments of procedures in Germany. Plausible baseline parameter values of the characteristics of screening and surveillance colonoscopy, of adenoma progression and recurrence, and of costs for diagnosis and treatment of colorectal

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

xvi

Bernice R. Hofmann

cancer were based on available data. The impact of major model assumptions was evaluated by sensitivity analyses. Results: A program based on one-time screening colonoscopy could result in net-savings over a period of 10 years in Germany due to prevention of cancer treatment costs compensating for the costs of screening, surveillance and adverse effects. Average net savings from 121 € to 623 € per screen could be achieved according to our model assuming different progression and recurrence rates of adenomas and carcinoma costs from 21.820 € to 40.000 €. Limitations: For some major model parameters assumptions had to be derived from the literature. Conclusions: This analysis based on empirical data from the nationwide screening colonoscopy program in Germany suggests net savings resulting fro Chapter 8 - Alcoholic drinks are a valued consumer item and an aid to sociability—less than 10% of adults in England do not drink alcohol. It is therefore essential to distinguish the misuse of alcohol from general consumption. Alcohol misuse is taken as drinking above the recommended weekly limit of 21 units of alcohol per week for a man and 14 units for a woman. By this standard, 30% of adult men and 17% of women report that their weekly consumption is above the recommended level; this is over 9 million people in England. Expenditure on alcoholic drinks in England in 2005 is estimated at £32 billion, split among beer, wine, spirits and fortified wines, cider and other drinks. During the past 10 years, “alcopops” (a variety of ready-to-drink mixes designed to appeal to younger people) were the fastest-growing subsector. Total alcohol consumption has increased by more than 60% in the last 30 years, and latest figures suggest that consumption has risen to the equivalent of 9.5 litres of pure alcohol per capita per year, higher than most other EU countries. Consumption of alcohol is 60% higher for men than for women; however, the highest rates of growth in consumption are among young people 16 to 24 years old. Alcohol consumption in this group grew by about 65% from 1993–2003, with fastest growth among young women. Alcohol consumption in England accounts for about a third of household expenditure on food and drink. It generates some £11.75 b in tax from excise duty and VAT. A current analysis of the full social economic and social costs of alcohol misuse identifies a total of £49 billion costs, including intangible social costs to society of £16 billion, costs to individuals and families of £23 billion, costs to employers of £2 billion, costs to public services of £6 billion and incapacity and income support costs of £2 billion. There are also tax receipts of some £3 billion attributable to excess alcohol consumption. Current evidence of the cost effectiveness of alcohol harm prevention measures in England and Scotland suggests that action on the legal minimum drinking age, blood alcohol laws, lower limits for young drivers and selective breath testing are effective in reducing road traffic injuries and deaths. Interventions aimed at training bar staff also shows evidence of effectiveness in reducing drunkenness. Chapter 9 - The Bayesian approach to statistics has been growing rapidly in popularity as an alternative to the classical approach in the economic evaluation of health technologies, due to the significant benefits it affords. One of the most important advantages of Bayesian methods is their incorporation of prior information. Thus, use is made of a greater amount of information, and so stronger results are obtained than with frequentist methods. In a costeffectiveness analysis, the authors relate the costs and effectiveness of the two technologies being compared, the parameters of interest being the mean effectiveness and mean cost of each. The most common prior structure for these two parameters is the bivariate normal

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Preface

xvii

structure. Since Stevens and O'Hagan (2002) showed that the elicitation of a prior distribution on the parameters of interest plays a crucial role in a Bayesian cost-effectiveness analysis, relatively few papers have addressed this issue, although Leal et al. (2007) recently presented a computer-based model to elicit uncertainty on parameters. In this paper the authors study the use of a more general (and flexible) family of prior distributions for the parameters. In particular, the authors assume that the conditional densities of the parameters are all normal. This structure allows us to incorporate a large range of prior information. The bivariate normal distribution is included as a particular case of the conditional prior structure. Chapter 10 - As the number of older people rise globally there is also a concomitant increase in the use of medicines to manage long-term illness. In England, older people are prescribed over 56% of the total prescriptions administered. Of these 78% are repeat prescriptions. Increasingly, older people are concomitantly prescribed more than four medicines. As a result of this, medication review was implemented as a measure to assess the efficacy of prescribed medicines, recognize adverse drug reactions, monitor adherence to medicines being administered for long-term conditions and to reduce unplanned admission to hospital. Medication review is predominantly undertaken by pharmacists in primary care/community settings or in hospitals. Although medication review is compulsory, current practice in Acute NHS and Mental Health Trusts is variable and the recent report suggests that further investment is needed to ensure that older people receive optimal medication and related information in Acute NHS Trusts. Of equal importance is the need for additional robust and extensive economic evaluations to assess the impact of medication review on measurable outcomes such as reduced unplanned admission to hospital and benefit to older people and the management of long-term conditions. Chapter 11 - The authors compared the training needs of English and Polish healthcare professionals in breastfeeding support skills. Self assessed competence in 23 skills, and knowledge of policies and guidance on breastfeeding were compared. The sample comprised non-medical (midwives, nurses and health visitors) and medical healthcare professionals (doctors). The results for both sub-samples were analyzed separately. The self-report questionnaire was developed in consultation with experts in breastfeeding practice, pedagogy and psychology, and potential participants. Four hundred and twenty seven English and 131 Polish non-medical healthcare professionals completed our questionnaire. Results showed that Polish respondents were more likely to see themselves as competent in managing specific breastfeeding problems and in clinical practice, while English practitioners more often perceived themselves as competent in regard to socio–cultural aspects of breastfeeding. In both countries breastfeeding knowledge was low. Unsurprisingly, staff who perceived themselves as more competent tended to have better knowledge. One hundred and seventy seven English doctors and 54 of their Polish colleagues took part in the survey. There were no differences in perceived clinical practice competence levels between the two samples although Polish doctors assessed themselves as more competent in regard to educative skills. Unlike the English sample, doctors within the Polish sample who perceived themselves as less competent expressed a greater need for updating. Polish doctors had wider access to policies and guidance on breastfeeding as well as greater breastfeeding knowledge. The authors conclude that training needs are very similar in both countries. Core training and regular knowledge and practice-based updates are recommended to all practitioners. Even

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

xviii

Bernice R. Hofmann

in UNICEF Baby Friendly Initiative-accredited units, more attention should be given to ensuring consistent and evidence-based practice in both countries, by updating breastfeeding policies and guidance on breastfeeding and ensuring that they are used by all healthcare practitioners. The authors conclude with recommendations for breastfeeding policy and for practice that are relevant to healthcare professionals in all developed countries. Chapter 12 - Clinical work in modern hospitals is characterized by a high degree of mobility, frequent interruptions, and much ad hoc collaboration between colleagues with different expertise. Today’s hospitals are under an increased pressure to become more efficient. There is an increasing number of patients and less budget for treatment. Hospital information systems promise such a necessary increase in productivity. However the medical point of care is different from any traditional office workplace. Usability is of utmost importance because the primary concern of doctors and nurses is the patient and not the information system. In this research the authors describe the design of a Radio Frequency Identification (RFID)-enabled inpatient safety management system (RISMS) and how it can be utilized to achieve a usable solution for the point of care. A user-centered design process is followed to achieve the necessary usability. The performance evaluations as well as the expected benefits of the system are also discussed. Chapter 13 - The estimation of a mean hospital cost has a great economic impact in hospital budgeting, funding and reimbursement. A typical feature of hospital data is that not all patients are followed up until the regular endpoint (home discharge). For example, a patient may die or be transferred to a different hospital before the end of his stay. In this case, his complete length of stay (LOS) and cost of stay are unobserved. In such situations cost and LOS are said to be “right censored.” Survival analysis techniques can be used in order to estimate the mean LOS in the presence of right censoring. These techniques are usually based on the assumption of independence between the unobserved complete LOS and the unobserved censoring time. Under this assumption, the censoring mechanism is said to be “non-informative.” Unfortunately, on the cost scale, the censoring mechanism is always informative, i.e., the unobserved complete cost and the unobserved censoring cost are generally correlated, due to the inherent patient heterogeneity with respect to cost accumulation. When standard survival analysis techniques - such as Kaplan Meier or Proportional Hazard models - are directly applied to cost data, the mean cost estimates are biased. In this note the authors discuss informative censoring and its effects on mean cost estimates. Moving from a very general and flexible model for the relationship between unit cost (e.g., daily cost) and LOS, the authors show that both a positive or negative correlation can arise between the unobserved complete cost and the unobserved censoring cost. This result contradicts the general believe that the correlation between cost and censoring cost is always positive and, therefore, that the mean cost is generally overestimated. The authors explain why direct application of survival analysis techniques to the cost distribution can also lead to an underestimation of the mean cost.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

EXPERT COMMENTARIES

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved. Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

In: Health Care Costs: Causes, Effects and Control Editor: Bernice R. Hofmann

ISBN 978-1-60456-976-6 © 2009 Nova Science Publishers, Inc.

Expert Commentary A

HEALTH TECHNOLOGY: MORAL HAZARD, INEFFICIENCIES, AND SOLUTIONS Read G. Pierce Resident Physician, Department of Internal Medicine, University of California, San Francisco, USA

Kevin J. Bozic Associate Professor in Residence, Department of Orthopaedic Surgery and Philip R. Lee Institute for Health Policy Studies, University of California, San Francisco, USA

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Abstract Rapidly rising health care costs in developed nations continue to be a major policy issue, and health technology (HT) adoption and utilization is frequently singled out as a primary driver of cost inflation. In the U.S., health systems and physicians enthusiastically acquire HT to further the scope and quality of patient care services, compete against other providers for patients, and enhance the financial viability of their practice. Patients pursue expensive HT in response to information asymmetry, which leads them to associate high-tech care with quality, and inefficient insurance coverage that shelters them from the true costs of their care. The resulting moral hazard threatens to reduce further technological innovation, as third party payers seek to control health care spending. To date, insurers and policymaker have yet to find an effective or popular strategy for encouraging more rational HT use. We propose four mechanisms for realigning incentives and risks to promote value-based technology adoption and utilization by health systems, physicians, and patients.

Commentary Rapidly rising health care costs in developed nations continue to be a primary concern for governments, policymakers, payers, and providers [2,9,13,25]. While the economics of care delivery is complex, technological innovation and costs associated with adoption and use of health technologies (HT) is considered the primary driver of health care cost inflation, particularly in the United States [1,13,26,28,34,35], which is the world leader in health care

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

4

Read G. Pierce and Kevin J. Bozic

technology innovation. Many HTs, including pharmaceuticals, devices, diagnostic tools, and systems of care (such as intensive care units), have improved the capacity of medicine to treat disease, thereby substantially improving patient care [1,35]. However, evidence abounds of inefficient and inappropriate HT use, which contributes to cost inflation and variable healthcare quality. This, in turn, threatens further innovation, as payers struggling to contain costs become more reluctant to pay for expensive HT without clear evidence of both clinical benefit and economic value [1,5]. Inefficiencies in HT adoption and use arise largely from misaligned risks and incentives perceived by users (health care systems, providers, and patients), and purchasers/payers (state and federal governments, private insurers). The future of HT, and further benefits for patient care, will require policies that alter the alignment of these incentives and risks. Providers of healthcare services, defined here as health systems (hospitals, integrated delivery networks, ambulatory surgery centers, primary care clinics, etc.) and physicians, are a key driver of HT adoption and use [11,34]. Three primary factors influence health systems’ HT decision-making. First is competition within regional markets for physician services and affiliation, which generate patient referrals and related reimbursement dollars. Both physicians and patients frequently seek high-tech centers of care, assuming HT availability is a reasonable proxy for quality, so health systems strive to adopt HT whenever possible despite potential redundancy within a given region. Second is fulfillment of institutional mission, i.e. providing up-to-date, high quality health services that will satisfy patients within a given community. Third is pursuit of profits, which allow institutions to more aggressively pursue growth strategies and insulate themselves against changes in reimbursements and regulatory requirements that may disrupt cash flow [1,3,15,26,29,35]. HT utilization patterns among individual physicians are more complex, but a number of similar factors tend to drive adoption and use. First is a desire to provide high quality care to patients. In pursuit of quality physicians rely on decision-making heuristics developed during their training, which generally value efficacy, safety, and patient satisfaction above costeffectiveness. Second is advanced training, which is often associated with specialization and is more common in the U.S than other developed nations. Third is competition with other physicians for revenues, with HT serving as a quality signal to potential patients. Fourth is profit, both for the sake of personal benefit and as a hedge against future reimbursement uncertainties [8,11,19,20,29,34]. A second important driver of HT adoption and use is patient demand. With reason, patients desire good health and willingly pursue HT that promises improved quality of life, greater longevity, or decreased risk of a catastrophic illness [3,11,29]. However, two critical factors—information asymmetry and moral hazard—encourage inefficient utilization. Information asymmetry arises when multiple parties engage in exchange of good or services, but one party has substantially greater information than the other(s). Patients seeking health care in the U.S. suffer information asymmetry on two counts. First, despite growing consumerism and unprecedented access to information through news media and the internet [15], patients’ expertise at the point of care is vastly inferior to physician knowledge and experience [4,11,13]. Faced with resulting uncertainty, patients either must trust in physicians’ recommendations [32,36] (which may tend toward over-utilization of HT) or rely on simplifying principles to guide decision-making. In the U.S., one such principle is the belief that technological innovation is generally good, and that newer HT tends to be superior to older HT regardless of cost [7]. Second, health insurance providers create complex, often

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Health Technology

5

confusing legal contracts that govern access to health care services. In response, patients select more comprehensive coverage, which allows greater access to HT [13,14,32,33]. Third party payers, which divest patients of financial risk by providing health insurance, support this inclination toward greater technology utilization [32]. The health economics literature chronicles several decades of debate regarding optimal levels of insurance [6,1012,13,18,21-24,27,31,33], but in developed nations insurance levels have remained high, and in many cases have increased, throughout the past half-century. The result is moral hazard [3,4,6,11,13,17,19,23,27,32,36]—patients consume more health services because they are shielded from the full costs of their decisions. In addition to driving up expenditures by making expensive HT available to many who could not otherwise afford it, moral hazard also discourages patients from seeking value (i.e. choosing an equally effective but less expensive technology) [11,13,22,23,31,32]. As a result, more expensive diagnostic and therapeutic approaches, such as MRI for headache evaluation or minimally invasive surgery for knee replacement, become commonplace, even if data supporting superior outcomes or quality are equivocal. Policymakers in the U.S. have employed a variety of cost-control and supply-side mechanisms, including prospective payment systems, managed care, capitation, and pay-forperformance (P4P), in an effort to control health system and physician HT utilization. Similarly, efforts to decrease patient demand for HT have involved increasing co-payments, HMO gatekeepers, and health savings accounts with high deductible insurance plans. With the exception of P4P, which is relatively new and largely untested, all of these methods have failed to alleviate perceived overuse of HT [22,30]. As Herzlinger recently argued, these failures arise in large part because consumers have little incentive to demand value when spending other people’s dollars [16]. Encouraging physicians and patients to adopt a value-based perspective will likely insure more effective and judicious adoption and utilization of expensive resources [13,17]. Such a value-based approach can be accomplished through four mechanisms. First, provide HT adopters better information about costs by incorporating cost-effectiveness analysis in all clinical trials [6]. Second, help patients become more savvy consumers by creating a standardized, federally regulated series of carefully vetted quality metrics that they can use to more objectively compare physicians and hospitals. This will eliminate tendencies for patients and physicians to use HT as a quality proxy. Third, make costs of care more transparent. While this will not aid patients who suffer acute illnesses, it will encourage those seeking more elective therapies to purchase based on value. Fourth, give patients greater control over the healthcare dollars they spend, either through more robust health savings accounts or other innovative tools that allow consumers to share a greater portion of the financial risk borne by third party payers [13,14,16,17]. Because technology diffusion increases with economic power (i.e. increasing wealth per capita) [26], setting an arbitrary proportion of GDP aside for healthcare will remain unrealistic, though this remains a popular strategy currently [14]. Until policy tools adequately address the persistent misalignment of incentives and risks surrounding HT utilization, healthcare cost inflation will continue to rise rapidly.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Read G. Pierce and Kevin J. Bozic

6

References [1] [2]

[3] [4] [5] [6] [7] [8] [9] [10] [11]

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

[12] [13] [14] [15]

[16] [17] [18] [19] [20]

Baker LC. Managed care and technology adoption in health care: evidence from magnetic resonance imaging. J Health Econ. May 2001;20(3):395-421. Baker LC, Birnbaum H, Geppert J, Mishol D, Moyneur E. The relationship between technology availability and health care spending. Health Aff. Nov 5, 2003; Web Exclusive. www.healthaffairs.org/WebExclusives.php. Beringer DC, Patel JJ, Bozic KJ. An overview of economic issues in computer-assisted total joint arthroplasty. Clin Orthop Relat Res. Oct 2007;463:26-30. Bessho S, Ohkusa Y. When do people visit a doctor? Health Care Manag Sci. Feb 2006;9(1):5-18. Bozic KJ, Pierce RG, Herndon JH. Health care technology assessment. Basic principles and clinical applications. J Bone Joint Surg Am. Jun 2004;86-A(6):1305-14. Braithwaite RS, Rosen AB. Linking cost sharing to value: an unrivaled yet unrealized public health opportunity. Ann Intern Med. Apr 17, 2007;146(8):602-5. Campion EW. Medical research and the news media. N Engl J Med. Dec 2 2004;351(23):2436-2437. Chakravorti B. The new rules for bringing innovations to market. Harv Bus Rev. Mar 2004;82(3):58-67. Coyte P, Holmes D. Health care technology adoption and diffusion in a social context. Policy Polit Nurs Pract. Feb 2007;8(1):47-54. de Meza D. Health insurance and the demand for medical care. J Health Econ. Mar 1983;2(1):47-54. Doran E, Robertson J, Henry D. Moral hazard and prescription medicine use in Australia—the patient perspective. Soc Sci Med. Apr 2005;60(7):1437-43. Ellis RP, Manning WG. Optimal health insurance for prevention and treatment. J Health Econ. Dec 2007 1;26(6):1128-50. Farnsworth D. Moral hazard in health insurance: are consumer-directed plans the answer? Ann Health Law. Summer 2006;15(2):251-73. Feldman R. The ability of managed care to control health care costs: how much is enough? J Health Care Finance. Spring 2000;26(3):15-25. Greenberg D, Peiser JG, Peterburg Y, Pliskin JS. Reimbursement policies, incentives and disincentives to perform laparoscopic surgery in Israel. Health Policy. Apr 2001;56(1):49-63. Herzlinger R. Who Killed Health Care: America’s $2 Trillion Medical Problem—And the Consumer-Driven Cure. McGraw-Hill: Boston, 2007. Huston CJ. Quality health care in an era of limited resources. Challenges and opportunities. J Nurs Care Qual. Oct-Dec 2003;18(4):295-301. Kim J, Ko S, Yang B. The effects of patient cost sharing on ambulatory utilization in South Korea. Health Policy. Jun 2005;72(3):293-300. Lundin D. Moral hazard in physician prescription behavior. J Health Econ. Sep 2000;19(5):639-62. Millenson ML. Moral hazard vs. real hazard: quality of care post-Arrow. J Health Polit Policy Law. Oct 2001;26(5):1069-79.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Health Technology

7

[21] Newhouse JP. Reconsidering the moral hazard-risk avoidance tradeoff. J Health Econ. Sep 2006;25(5):1005-14. [22] Nyman JA. American health policy: cracks in the foundation. J Health Polit Policy Law. Oct 2007;32(5):759-83. [23] Nyman JA. Is 'moral hazard' inefficient? The policy implications of a new theory. Health Aff. Sep-Oct 2004;23(5):194-9. [24] Nyman JA. The economics of moral hazard revisited. J Health Econ. Dec 1999;18(6):811-24. [25] Oberlander J. The US health care system: on a road to nowhere? CMAJ. Jul 23, 2002;167(2):163-168. [26] Oh EH, Imanaka Y, Evans E. Determinants of the diffusion of computed tomography and magnetic resonance imaging. Int J Technol Assess Health Care. Winter 2005;21(1):73-80. [27] Pauly MV. Medicare drug coverage and moral hazard. Health Aff. Jan-Feb 2004;23(1):113-22. [28] Pierce RG, Bozic KJ, Hall BL, Breivis J. Health care technology assessment: implications for modern medical practice. Part I. Understanding technology adoption and analyses. Am J Orthop. Jan 2007;36(1):71-76. [29] Pierce RG, Bozic KJ, Hall BL, Breivis J. Health care technology assessment: implications for modern medical practice. Part II. Decision making on technology adoption. Am J Orthop. Feb 2007;36(2):71-6. [30] Pierce RG, Bozic KJ, Bradford DS. Pay for performance in orthopaedic surgery. Clin Orthop Relat Res. Apr 2007;457:87-95. [31] Richardson J, McKie J. Economic evaluation of services for a National Health scheme: the case for a fairness-based framework. J Health Econ. Jul 1, 2007;26(4):785-99. [32] Rosenthal MB, Newhouse JP. Managed care and efficient rationing. J Health Care Finance. Summer 2002;28(4):1-10. [33] Savage E, Wright DJ. Moral hazard and adverse selection in Australian private hospitals: 1989-1990. J Health Econ. May 2003;22(3):331-59. [34] Selder A. Physician reimbursement and technology adoption. J Health Econ. Sep 2005;24(5):907-30. [35] Teplensky JD, Pauly MV, Kimberly JR, Hillman AL, Schwartz JS. Hospital adoption of medical technology: an empirical test of alternative models. Health Serv Res. Aug 1995;30(3):437-65. [36] Vera-Hernández M. Structural estimation of a principal-agent model: moral hazard in medical insurance. Rand J Econ. Winter 2003;34(4):670-93.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved. Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

In: Health Care Costs: Causes, Effects and Control Editor: Bernice R. Hofmann

ISBN 978-1-60456-976-6 © 2009 Nova Science Publishers, Inc.

Expert Commentary B

THE FUTURE OF HEALTH-CARE QUALITY IN THE UNITED STATES Brandon Roberts Premier Insights, Inc., Canton, MS, USA

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Overview The issue of health-care quality in the United States has been of considerable concern for over a decade. An alarming report issued in November of 1999 by the Institute of Medicine (IOM) entitled To Err is Human: Building a Safer Health System catapulted this issue to the forefront during the Clinton administration. The IOM report contended that there were potentially 100,000 deaths annually in U.S. hospitals that were due to medical errors. This figure was staggering and suggested that deaths from “medical accidents” in hospitals exceeded the number of deaths from automobile accidents, breast cancer, and AIDS combined, and was over six times the number of murders committed nationwide. The report’s release prompted a number of ongoing public and private initiatives to address deficiencies in health care.

Extent of the Problem Although the IOM report focused significant attention on this issue, the diminution of healthcare quality was not a new phenomenon. Research had been noting for several years that the frequency of medical errors was a serious matter that needed to be addressed, but prior to the IOM report, it unfortunately received little attention. Today, however, the issue continues to receive attention in the media, largely through anecdotal evidence and reports by private agencies and other public interest groups chiding poor hospitals, physicians, and medical organizations for poor quality care or medical mistakes. The 1999 IOM report estimated that medical errors generated $37.6 billion in costs each year. This figure is sizeable, as annual

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

10

Brandon Roberts

expenditures for durable medical goods, an important component of medical costs, totaled only $23.7 billion in 2006. The Agency for Healthcare Research and Quality (AHRQ), which operates under the auspices of the U.S. Department of Health and Human Services, admits that there is little data on medical errors outside of hospitals. It is clear, however, that many medical mistakes occur outside of hospital settings. For example, many involve medication errors. The Massachusetts State Board of Registration in Pharmacy estimates that 2.4 million prescriptions are fulfilled in error each year in that state alone.1 The AHRQ indicates that medical errors occur in virtually all medical settings, including physicians’ offices, nursing homes, urgent care centers, and home health environments.

Current State of Health-care Quality

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

The AHRQ began monitoring health-care quality shortly after the release of the 1999 IOM report. The most current data available at the time of this writing was for 2001 and 2003, contained in the 2006 report. Although this report from the AHRQ pointed out improvement in the quality of health care, review of specific data elements was less encouraging. For example, the percentage of emergency room visits in which the patient left without being seen increased by 48% between 1997 and 2004, and the rate of suicide deaths in hospitals increased by nearly 4% between 2000 and 2003. More importantly, deaths per 1,000 hospital admissions for low-mortality diagnoses reflected increases between 2001 and 2003 (Figure 1). Overall, the rate per 1,000 low-risk admissions increased from .75 in 2001 to 1.18 in 2003. The rate for Medicare patients increased by over three times from 1.04 to 3.75, and the rate for the 85 and over age group jumped from 7.00 to 18.28.2 Deaths Per 1,000 Admissions in Low Mortality Diagnosis Related Groups 18.28 20.00 15.00 7.00

10.00

2003

3.75 5.00

0.75

1.18

2001

1.04

Overall

Medicare Patients

Age 85+

Source: Agency for Healthcare Research and Quality.

Figure 1. Deaths per 1,000 Hospital Admissions in Low Risk Groups.

1

See for sources and discussion: Medical Errors: The Scope of the Problem. Fact sheet, Publication No. AHRQ 00P037. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/qual/errback.htm. 2 Health-care quality reports are available from AHRQ as noted in the previous footnote. Some data are also assessable via online query at: http://nhdrnet.ahrq.gov/nhdr/jsp/nhdr.jsp. Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

The Future of Health-care Quality in the United States

11

Accidental Puncture or Laceration During Procedures per 1,000 3.76 3.8 3.7 3.49 3.6 3.5 3.4 3.3 2001

2003

Source: Agency for Healthcare Research and Quality.

Figure 2. Accidental Punctures or Lacerations During Procedures.

Accidental Puncture or Laceration During Procedures per 1,000 by Type of Hospital

3.75

Overall

3.49 4.26

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

500 Beds+

3.67

2003 4.22

Teaching

3.76

Public 0

2001

3.79

3.49

1

2

3

4

5

Source: Agency for Healthcare Research and Quality.

Figure 3. Accidental Punctures or Lacerations by Type of Hospital.

Between 2001 and 2003, the rate of accidental punctures during procedures also increased (Figure 2). Overall, the rate per 1,000 increased from 3.49 in 2001 to 3.75 in 2003. The rates were higher and reflected greater increase in public hospitals, teaching hospitals, and hospitals with 500 beds or more (Figure 3). In addition to the public sources, a number of independent firms monitor health care quality. One such concern is Healthgrades, a private firm which provides ratings information to consumers, businesses, and the health-care community. A report released in 2007, HealthGrades Patient Safety in American Hospitals Study, suggested a significant increase in

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

12

Brandon Roberts

medical errors in hospitals between 2003 and 2005. The study was limited to Medicare patients treated in hospitals and found a total of 1.16 million medical safety incidents nationwide between these years. The report further stated that during the three years examined, there were 247,662 occurrences of preventable deaths, and that Medicare patients involved in safety incidents had a one out of four chance of dying. The study also notes considerable variation in health-care quality among hospitals and considerable costs associated with medical errors. Again, these figures are disturbing and if the figures prove accurate, signify a serious problem and warrant considerable concern.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Public Perception Public opinion research suggests a growing awareness of the prevalence of medical errors and accident occurrences in the U.S. A study by the National Patient Safety Foundation found that 42% of respondents had been affected by a medical error, either personally or through a friend or relative, with 32% indicating that the error had had a permanent and negative effect on the patient’s health. Surveys conducted by the Kaiser Family Foundation indicated a growing awareness of and familiarity with the term “medical errors” among the public.3 In 2002, 40% of respondents indicated that they had never heard the term before, compared to only 17% in 2006. In 2006, over 50% indicated that they understood what the term meant, compared to 31% in 2002. Examination of polling results also suggests a general decline in public satisfaction with health-care quality since 1993. In May of 1993, 51% of respondents in a national sample indicated that they were generally satisfied with the quality of health-care in the U.S., compared to 44% in August of 2006. Likewise, in 1993 46% indicated that they were generally dissatisfied, compared to 51% in 2006. Public opinion polls also point towards a perception of growing frequency of medical errors. In 2004, 36% of a national sample indicated that preventable medical errors occur “somewhat” or “very” often, compared to 43% in a 2006 poll.

The Market for Health Care and Quality In a previous volume, I noted the shift in health-care expenditures by source of funds since 1960 and the inverse relationship with medical costs as a percentage of GDP (Figure 4).4 As depicted by the graph, medical costs as a percentage of GDP have been generally increasing since 1960, while the out-of-pocket share—the portion of medical expenditures paid directly by the consumer—has been decreasing. It is interesting to note how these trends correspond. In particular, in the period between 1994–1999, out-of-pocket expenditures remained relatively constant and medical expenditures, as a percentage of GDP was also constant, remaining between 13.6%–13.7% of GDP. As the share of out-of-pocket expenditures began to fall again in 2000, medical expenditures as a percent of GDP began climbing. 3 4

Available at: http://kff.org. Sophie J. Evans, “The Political Economy of Healthcare,” Public Policy Issue Research Trends, Nova Publishers, 2008: pp. 11–15.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

The Future of Health-care Quality in the United States

13

With a greater share of expenditures being paid by private and public sources and less directly to the provider by the consumer, economic theory suggests a resulting demand shift and a subsequent increase in price. A significant body of literature studying the role of medical insurance supports this notion as well. Demand for medical care is subject to increase as a greater share of medical costs is borne through comprehensive medical insurance plans. Under this scenario, consumers have an incentive to maximize their usage of care, as all but a small portion of their costs is essentially fixed. In addition to the role of insurance, health-care demand has increased due to the aging of the population and longevity. In particular, growth in the age 80 and over cohorts exerts considerable pressure on health-care services. This is further compounded by very limited increases in physicians. The number of new physicians entering the workforce has been relatively constant since the 1980’s. If we assume that the market for medical care is essentially competitive; this is easily illustrated in Figure 5. Further assuming relatively inelasticities of both the supply and demand sides of the market, we can model the impact of a demand shift. Because physician supply has remained virtually constant, increased supply is limited to movement on the supply curve rather than a subsequent shift in supply. As depicted, the increase in quantity of medical services from movement along the supply curve is marginal under an inelastic scenario, and the impact on price substantial. 50.0 45.0 40.0

% of GDP

% Out-Of-Pocket

35.0

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

30.0 25.0 20.0 15.0 10.0 5.0 0.0 1960 1961 1962 1964 1965 1966 1967 1969 1970 1971 1972 1974 1975 1976 1977 1979 1980 1981 1982 1984 1985 1986 1987 1988 1990 1991 1992 1993 1995 1996 1997 1998 2000 2001 2002 2003 2005 2006 %of GDP

5.2 5.4 5.5

%Out-Of-Pocket

46.7 45.15 44.14 43.5 42.9 39.8 35.5 34.01 33.2 31.53 30.7 29.3 27.91 26.6 25.9 23.8 23.0 22.3 21.85 21.5521.7521.9721.30 20.7 19.06 17.9216.8815.91 14.3914.23 14.44 14.74 14.26 13.61 13.18 12.95 12.55 12.18

5.8 5.9 5.9 6.3 6.7 7.2

7.4 7.5 7.8

8.1 8.4

8.5 8.6

9.1 9.4 10.2 10.2 10.4 10.6 10.8 11.2 12.3

13 13.4 13.7 13.7 13.7 13.6 13.6 13.8 14.5 15.3 15.8

16 16.1

Source: National Healthcare Accounts, CMS.

Figure 4. Medical Expenditures as a Percent of GDP & Out-of-Pocket Expenditures 1960–2006.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

14

P

Brandon Roberts

D

D’

p’

Increase in price due to demand shift

p

Increase in quantity due to movement along supply curve

S

q

q’

Q

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Figure 5. Impact of Demand Shift—Case of Demand & Supply Inelasticities.

I use this elementary illustration simply to point out that strong demand, coupled with a large increase in price and restricted supply, would result in a less competitive market. Research from both heavily regulated and unregulated industries suggests strong association with service quality and competition.5 If we further consider as well that the service provider is receiving the majority of revenues from a source other than the patient, it seems that the environment created is not conducive to quality care. Researchers have noted a shift as far back as 1993 toward more price- and payer-driven competition, rather than patient-driven competition, with the result being patient care dictated by third-party payers rather than health outcomes or quality of care (Footnote 5). Even with the abundance of information available today to consumers in regard to health and the target-marketing of consumers directly for medical related services (particularly by pharmaceutical companies), consumers may be in less control of their health care now than ever before. The key point is that the structure of the market for health care lacks many of the critical incentives that would encourage quality care, and thus are not in favor of the patient. These conditions are likely to persist because there are projected shortfalls in the number of needed physicians over the next several years. One possible offset to this trend, however, is the 5

See: D. Dranove, M. Shanley, and W.D. White. “Price and Concentration in Hospital Markets: The Switch from Patient Driven to Payer-Driven Competition.” Journal of Law and Economics. 1993; 36:179-204; also Gary J. Young, James F. Burgess, Jr., and Danielle Valley, “Competition among Hospitals for HMO Business: Effect of Price and Nonprice Attributes,” Health Serv Res. 2002 October; 37(5): 1267-1289.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

The Future of Health-care Quality in the United States

15

increase in health-care services provided by non-physicians. The results from care administered by non-physicians—including professionals such as chiropractors, nurse practitioners, and psychologists—have been positive and the trend is expected to continue.5 If Figure 5 is an accurate portrayal of the health-care market, one could argue that it is in a state of “excessive” demand. The large increase in price from p to p’ and the marginal change in supply from q to q’ suggests an unfavorable environment exists for those in need of medical care, and this may be reflected not only in price but also in quality. If we further combine this with the fact that roughly 90% of the increase from p to p’ is paid from a source other than the patient, the possibility of an adverse impact on quality is even more plausiblemedical service providers must simply satisfy the third-party payers to collect for their services, and they typically do so independent of patient outcomes and without consideration of quality measures.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Policy Options It seems clear that there are indicators of a serious and growing problem with health-care quality in the United States. Using the limited data that is available, the trend in number of medical errors is alarming, and the results of policy initiatives do not appear promising. Projected shortfalls in numbers of physicians, growing medical costs, and increasing demand suggest little change in the short term. That being said, it should be noted that data from both public and private sources cite considerable variation in the quality of care and in the incidence of medical errors among differing hospitals. This seems to suggest that there are solutions to minimize these adverse conditions even without major structural changes. This issue is fertile ground for research of policy alternatives. One potential solution posed in November of 2007 by CMS is to reduce Medicare hospital reimbursements by between 2% and 5% to fund an incentive pool for hospitals that meet quality standards. The idea is to pay hospitals based on quality performance; this would be based on a number of factors, including patient outcomes.6 Some type of performance measures appears badly needed for Medicare patients, as reflected in the data presented earlier. This initiative, however, is being introduced along with a 10% reduction in payments to physicians, scheduled to take effect January 1, 2008. The question is, would policies that potentially impose further restrictions on the supply-side of the market be truly beneficial for patients? Again, returning to our elementary diagram from Figure 5, we can model the effect of a cap on physician fees as a price ceiling. Assuming the same elasticities as before and assuming a competitive model, a ceiling on physician fees would result in a change from the q1 to q2, while demand would move to q’ and thus create a shortage equal to q’ – q2. Although the government would benefit under this scenario by reduced costs, Medicare recipients would likely suffer, given that these are the only facets of the change. Another interesting initiative that has been proposed by CMS is the implementation of a testing program to monitor service quality via “secret shoppers.” This approach is frequently used in the private sector to test, monitor, and motivate customer service. The typical method 5

For a discussion see: “Trends in Care by Nonphysician Clinicians in the United States,” Druss BG, Marcus SC, Olfston M, Tanielian T, Pincus HA. New England Journal of Medicine, 348 (2), Jan 9 2003, pp. 130–137. 6 See: Francis, Wall Street Journal, 11/27/07. Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

16

Brandon Roberts

is to recruit “shoppers” to test the level and quality of service provided, as well as the knowledge of staff and the accuracy of information provided. These programs are highly effective in competitive, service-driven industries, and it will be interesting to observe the effectiveness of such an application in the public sector. Competitive Model Model A

Model B

Before Medicare Cap

After Medicare Cap

D

P

D

P

Consumer Surplus

Deadweight Loss

p

p1

Consumer Surplus Producer Surplus

p2 Producer Surplus

S

q

Q

Shortage = q2-q’

S

q2 q1 q’

Q

Figure 6. Potential Effect of Medicare Physician Fee Reduction on Recipients.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Conclusion The disturbing trend of diminishing healthcare quality does not appear to be going away anytime soon. Despite the alarm sounded a decade ago and subsequent intervention efforts the available data suggests continued deterioration in the quality of medical care. This issue has immense social, economic, and political implications. Potent solutions must be developed and implemented to halt the downward spiral that the industry appears to be in. Changing behavior requires rearranging of incentives, and matters with social and economic policy implications vacillate on a continuum between regulation and market based approaches. Because this issue is of profound importance, decisive action is necessary but with great care - policy changes usually produce both intended and unintended consequences. This requires empirically driven, erudite answers. Appropriate solutions must be formulated in the full contexts of potential impacts while centered on the interests of the patient. The scientific community has the intellectual resources to develop properly balanced solutions to counter the negative trend of medical quality. This area is in need of immediate focus. I am hopeful that scholars from diverse fields applying their toolsets will provide viable policy alternatives.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

SHORT COMMUNICATIONS

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved. Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

In: Health Care Costs: Causes, Effects and Control Editor: Bernice R. Hofmann

ISBN 978-1-60456-976-6 © 2009 Nova Science Publishers, Inc.

Short Communication A

THE ELASTICITY OF PHYSICIAN SUPPLY AND MEDICAL MALPRACTICE LITIGATION IN THE UNITED STATES Brandon Roberts Premier Insights, Inc., Canton, MS, USA

Euel Elliott University of Texas at Dallas, USA

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Abstract The issue of medical malpractice reform has occupied state legislatures for the last several years, due to the concerns expressed by physicians and others that the costs of malpractice litigation are harming the delivery of medical services in the United States. Using data from the 1997-2001 time period, the impact of malpractice litigation on physician supply in the U.S. is estimated. The results indicate that medical malpractice litigation is a significant factor in reducing physician supply in the U.S. More specifically, we find that laws in those jurisdictions that have direct influences on medical liability have more impact than those with an indirect effect. We conclude that the concerns over medical malpractice litigation are not misplaced or exaggerated, and that policymakers should strongly consider those policy changes that exert a more direct influence on litigation.

1. Introduction The issue of medical liability has consumed a significant portion of the public policy debate over the last few years. Emotionally charged and at times polarizing, policy makers and politicians have experienced considerable pressure from activists and groups on both sides of the issue arguing for and against the need for liability reform. The issue seemed to reach a pinnacle in 2003 when the American Medical Association declared a large number of states in “crisis” because of the potential shortage of physicians willing to practice in jurisdictions deemed as higher risk and the reluctance of insurers to provide malpractice liability coverage

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Brandon Roberts

20

in these jurisdictions.1 The degree of attention called to the issue sparked numerous efforts at medical liability and tort reforms in a number of states over the past few years.2 There has been considerable activity across the states in terms of efforts to rein in medical malpractice litigation with varying degrees of decisiveness. Our Table 5 shows reforms in place by state for the time periods in our study categorized as per Table 1. According to the National Conference of State Legislatures, all states currently have at least some type of laws in place that govern medical liability. In addition to action at the state level, national legislation has been proposed (with White House support) and discussed for several years, although no legislation has been passed. Clearly, sufficient concern exists to mobilize political support for legal remedies.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Table 1. State Law Differences

Reform

Description

Impact

Caps on awards

Damages are capped at a specific dollar amount

Direct

Abolition of punitive damages

Malpractice defendants are not liable for punitive damages

Direct

No mandatory prejudgment interest

Accrued interest from date of injury or filing not required

Direct

Collateral-source rule

Damages reduced by dollar value of collateral source payments

Direct

Caps on contingency fees

Portion of award a plaintiff can agree to pay an attorney is capped at a statutory level

Indirect

Mandatory periodic payments

Part or all of damages have to be paid out in the form of annuity over time

Indirect

Joint and several liability reform

Joint and several liability abolished for claims in which defendants did not act in concert

Indirect

Patient compensation fund

Doctors receive government-administered excess malpractice liability insurance, generally through a tax on malpractice premiums

Indirect

1

American Medical Association, Press Release, 3/3/2003, “18 States Now in Full-Blown Medical Liability Crisis. ” See also “Liability for Medical Malpractice: Issues and Evidence,” Joint Economic Committee Study, house.gov/jec; also “Confronting the new health care crisis: Improving health care quality and lowering costs by fixing our medical liability system,” U.S. Department of Health and Human Services, July 24, 2002: available at http://aspe.hhs.gov/daltcp/reports/litrefm.htm. 2 A summary of recent activity by state legislatures is available from the National Conference of State Legislatures available at: http://www.ncsl.org/standcomm/sclaw/medmaloverview.htm.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

The Elasticity of Physician Supply …

21

This paper explores the extent to which the state and national political mobilization is justified. To what extent, in other words, is the litigation involving medical malpractice actually having a negative impact and the quality of medical care in the United States? More specifically, we address the question of whether and to what extent such litigation influenced the supply of physicians for a five-year time period leading up to the debate’s climax in 2003. Previous research has examined the effects of liability reforms on the supply of physicians by jurisdiction. Some of these studies have indicated that areas in reforming states have either a greater supply of physicians or a faster rate of growth in the number of area physicians (Hellinger and Encinosa, 2003; Encinosa and Hellinger, 2005; Kessler, Sage and Becker, 2005), while others suggest that the relationship between physician supply and liability reform is tenuous (Mello, Studdert, Schumi, Brennan, Sage; 2007). Prior research has also modeled physician supply as a function of the local liability environment. In a prior work (authors, 2008 forthcoming), we modeled physician supply as a function of the level of malpractice litigation within jurisdictions within a single state and found negative and significant association between medical malpractice litigation and the supply of area physicians. We carry out a similar investigation in this study. Using a number of control variables identified by prior research, including our own, along with area randomeffects to control for unobserved heterogeneity, we measure the potential influence of varying levels of medical malpractice litigation on the supply of area physicians in the United States.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

2. Research Design and Data Our study utilizes a panel dataset covering time periods 1997 - 2001. We observe our variables over time and estimate the unknown parameters as a pooled, time fixed, and spatial random-effects regression. Our unit of analysis is state metropolitan and non-metropolitan areas. The parameter estimation was done with GLS regression, and we use log transformations of the key variables to estimate the elasticity of the supply of physicians with respect to litigation. There were a total of 920 observations in our dataset, with a total of 184 geographic areas over a five-year time period. The data and methodology are described in greater detail below.

A. Dependent Variable Our dependent variable is the total number of active, non-federal physicians in a given year within each geographic area from the Area Resource File (ARF). The total number of physicians was expressed as the natural log of physicians in the model. Using the logged version of the variable avoids having the dependent as discrete data in the model while retaining the desirable properties of the GLS estimator.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Brandon Roberts

22

B. Independent Variable The independent variable used was the total number of medical malpractice settlements within each unit of analysis in each year obtained from the National Practitioner Data Bank. The logged value of this variable was used in the parameter estimation. Maintained by Health and Human Services, this data is publicly available at the state and national level, but a special extract was obtained at the Metropolitan Statistical Area (MSA) level and also for non-metropolitan areas of each state.3 The records within each MSA and non-MSA portions of each state were selected by zip code, and the data that was provided aggregated to the MSA or non-metro unit of analysis. The data contained the number of total malpractice settlements and the total dollar amounts of settlements. Because of confidentiality, the NPDB required a minimum of eleven cases in each given year for the data to be released. Each MSA that reached the eleven case minimums in all fivetime periods was used in the analysis along with the non-metropolitan portion of each state as long as it met the eleven case rule. For MSA’s that spanned more than one state, the area was split into two separate units of analysis, again providing that the eleven case minimum was met. Although a subset of the United States, the geographic areas covered included 80% of the population and mirrored the national demographic composition of the nation.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Table 2. Regional Composition

3

Region

States

Region 1

New England, Mid-Atlantic and Great Lakes states: Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, Vermont, New Jersey, New York, Pennsylvania, Illinois, Indiana, Michigan, Ohio and Wisconsin

Region 2

Delaware, District of Columbia, Maryland, Virginia and West Virginia

Region 3

Alabama, Georgia, Kentucky, Mississippi, North Carolina, South Carolina and Tennessee

Region 4

Florida

Region 5

Arkansas, Louisiana, Missouri, Montana, Oklahoma and Texas

Region 6

Colorado, Iowa, Kansas, Minnesota, Nebraska and Wyoming

Region 7

Arizona, California, Hawaii, Nevada, New Mexico, Oregon, Utah and Washington

We are especially grateful for Dr. Robert Oshel for his assistance and expertise in providing this data.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

The Elasticity of Physician Supply …

23

Table 3. Variables Employed and Sources Variable

Source

Mean

Std. Dev.

Area Resource File

7.59

0.92

NPDB

3.76

0.86

NPDB

9.11

0.99

DEPENDENT Ln(Physicians) INDEPENDENTS Ln(Number of Malpractice Settlements) Ln(Dollars of Malpractice Settlements in 1,000's) DEMOGRAPHICS Median Households Income 2000 in 1,000's

U.S. Census Bureau

42.48

8.97

Ln(Population 2000)

U.S. Census Bureau

13.66

0.79

Annual Population Growth as a Percentage

U.S. Census Bureau

1.08

1.05

Percent of Population that is White

U.S. Census Bureau

74.86

13.34

Percent = or > Poverty

U.S. Census Bureau

11.34

3.98

Percent with a B.S. Degree or >

Area Resource File

23.93

7.69

Area Resource File

81.23

5.91

CMS

6236.64

6452.34

Area Resource File

4272.03

4572.65

CMS

2367.31

506.01

CMS

112.73

99.77

Percent with a H.S. or > MEDICAL SERVICES Total Nursing Home Beds Number of Hospital beds UTILIZATION Medicare Part B Exp Per Enrollee

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Part B Enrollment in 1,000's STATE LAW DIFFERENCES State Law Dummy - Caps on Damages

McCullough, Campbell and Lane

NA

NA

State Law Dummy - Direct Impact Legislation

McCullough, Campbell and Lane

NA

NA

State Law Dummy - Indirect Impact Legislation AGE

McCullough, Campbell and Lane

NA

NA

Population Age 64-69

U.S. Census Bureau

42153.79

39402.79

Population Age 70-74

U.S. Census Bureau

39283.81

36536.86

Population Age 75-79

U.S. Census Bureau

32980.91

30632.92

Population Age 80-84

U.S. Census Bureau

21995.52

20469.19

Population Age 85+ REGIONAL CONTROLS

U.S. Census Bureau

18857.29

18121.97

Regional Dummies (7) TIME & AREA FIXED-EFFECTS

Extant Literature

NA

NA

Time Dummies (5)

Generated from Data

NA

NA

Random-Effects Area Dummies (184)

Generated from Data

NA

NA

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

24

Brandon Roberts

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

C. Control Variables We relied on the extant literature as well as physician supply forecasting models to determine factors to control for in our models.4 The demographic attributes used in the models included income, population size, educational levels, race (Shi and Starfield 2000; Kindig and Yan 1993; Hellinger and Encinosa 2003, 2005), and a dummy variable to distinguish between non-metro and metro areas.5 Because the demographic characteristics were based on 2000 Census data and constant across years, changes were controlled for by including the annual percentage population growth into the model. The elderly population represents a greater demand for medical services. We include total Medicare enrollment for Part B for the aged population as well as the age stratification of the 65 and over population compiled from the Medicare county fee-for-service files. To account for utilization and the availability of medical services, we include Medicare Part B perenrollee-expenditures, the total number of nursing home beds, and the total number of hospital beds. There is considerable variation across states in laws governing medical liability as well as recent changes and diversity of legislative and policy actions. We account for state law differences by dummy variables based on Table 1. Legislative differences are divided into three categories: (a) laws that are believed to have a direct effect on medical liability, (b) laws that include a provision for caps on damage awards, and (c) laws that are believed to have an indirect effect on medical liability. (Kessler and McClelland, 1996, 1997; Kessler, Sage and Becker, 2005).6 Demand for medical services and utilization may also vary regionally and independent of demographic attributes (Fuchs, McClelland and Skinner, 2001) which could subsequently affect demand. We use the regional classifications (Fuchs et. al) as controls in our models. Our Table 2 shows the composition of these areas. Finally, we control for trend with time dummies and for unobserved variation by the inclusion of spatial random-effects, using tests for fixed and random effects models.8

3. Model Specification Selecting the appropriate econometric model posed a number of challenges. First, using the number of physicians as a discrete dependent variable was undesirable (Long, 1997). Second, as would be expected, the number of medical practice settlements was largely dependent upon the number of area physicians – more practicing physicians creates an opportunity for more litigation. Third, plots of the key variables suggested relationships were non-linear. For example, physician density (physicians per capita) was initially negatively correlated with 4

5

See aso “Physician Supply and Demand: Projections to 2020,”U.S. Department of Health and Human Services, Health Resources and Services Administration, Bureau of Health Professionals, October, 2006. Available online: ftp://ftp.hrsa.gov/bhpr/workforce/PhysicianForecastingPaperfinal.pdf. For importance of metro and non-metro differences see “PHYSICIAN WORKFORCE: Physician Supply Increased in Metropolitan and Nonmetropolitan Areas but Geographic Disparities Persisted,” GAO report available at: http://www.gao.gov/new.items/d04124.pdf.

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

The Elasticity of Physician Supply …

25

population size, but reached an asymptote and was thus evaluated as non-linear.7 Similar patterns were noted with physicians and the litigation measure. Using log transformations of the data helped resolve these issues. Visual examination of the data, the significance of quadratic terms, and tests of the functional form (Gujarati, 2003) suggested the log linear model as the best fit and also allowed ease of interpretation with by modeling elasticity.8 Our model is given as:

Ln(yit ) = α + φt + λi + βLn( Χit ) + CΖit + εit where, yit = ln(number of area physicians)

φi = Time fixed - effects dummy λi = Random - effects dummy β = Medical malpractice litigation coefficient Χit = Ln(Medical malpractice litigation variable) C = Control variables coefficients Ζit = Vector of control variables

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

εit = Uncorrelated disturbance term

4. Parameter Results A. Control Variables Our complete regression estimates are presented in Table 4. The model was first executed without the litigation variable to assess the performance of the control variables and then further elaborated with the litigation variable. The control variables generally produced the expected parameter signs and significance (Model 1) and are discussed below. Time is controlled for in the model by dummies for the time periods with 1997 used as the omitted case. All four of the year dummies were positive and significant reflecting consistent growth in physicians relative to 1997 over the five-year period. Median household

6

The complete compendium used for state law controls is available online at: http://www.mcandl. com/introduction.html. Table I reflects the direct and indirect categorization of each reform, with reforms including caps treated as a separate variable in the regressions. 7 Plots available upon request email author:

Health Care Costs: Causes, Effects and Control, edited by Bernice R. Hofmann, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

26

Brandon Roberts

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

income was significant but negative as was annual population growth. Log of population was positive and significant, and the non-metro dummy was negative and significant. Percent of the population that was at or below poverty, percent white, and percent of the population with a high school education or greater were all insignificant, but percent of the population with a B.S. degree or greater was positive and significant. The number of persons aged 64 – 69 was negative and significant, and the number of persons aged 70 – 74 was positive and significant, and the remainder of the age variables was insignificant. The total number of hospital beds was positive and significant as were per enrollee Part B expenditures and total Part B enrollment. The number of nursing home beds was negative but insignificant. With respect to legislative differences, the variable for caps on damage awards was negative but insignificant while the variable for reforms believed to have a direct impact on medical liability was positive and significant, while the indirect variable was negative and significant. The coefficients indicated that jurisdictions with laws that are believed to directly affect medical liability have 2.3% more physicians relative to those that do not. Those with indirect reforms have 1.7% fewer physicians relative to those that do not. Many jurisdictions that have implemented direct reforms have also implemented caps (although treated separately in our model) which may explain the insignificance of the caps variable alone. Additionally, Kessler and McClelland (1996, 1997) found that indirect reforms have less impact and also that legislative changes require time to take effect.9 These results are somewhat consistent with other research (Kessler et. al, 2005). Finally, only region three (all southern states) was significant and only at P