The Impact of Health Insurance in Low- and Middle-Income Countries - Softcover

Softcover

ISBN 10: 0815705468 ISBN 13: 9780815705468

Publisher: Brookings Institution Press, 2011

Synopsis

Over the past twenty years, many low- and middle-income countries have experimented with health insurance options. While their plans have varied widely in scale and ambition, their goals are the same: to make health services more affordable through the use of public subsidies while also moving care providers partially or fully into competitive markets. Until now, however, we have known little about the actual effects of these dramatic policy changes. Understanding the impact of health insurance-based care is key to the public policy debate of whether to extend insurance to low-income populations - and if so, how to do it - or to serve them through other means. Using recent household data, this important new book reveals the impact of insurance programs in China, Colombia, Costa Rica, Ghana, Indonesia, Namibia, and Peru. These nations are betting that insurance-based health care financing can increase the accessibility of services, increase providers' productivity, and change the population's health care use patterns, mirroring the development of health systems in most OECD countries. The contributors also discuss potential design improvements that could increase impact. They provide innovative insights on improving the evaluation of health insurance reforms and on building a robust knowledge base to guide policy as other countries tackle the health insurance challenge.

About this title

About the Author

Maria-Luisa Escobar is lead health economist and health systems program leader at the World Bank Institute. Charles C. Griffin is senior advisor in the European and Central Asia regional office of the World Bank. R. Paul Shaw, a former World Bank lead economist, advises the Bill and Melinda Gates Foundation on health economics.

Excerpt. © Reprinted by permission. All rights reserved.

The Impact of Health Insurance in Low- and Middle-Income Countries

By Maria-Luisa Escobar Charles C. Griffin R. Paul Shaw

BROOKINGS INSTITUTION PRESS

Copyright © 2010 The Brookings Institution
All right reserved.
ISBN: 978-0-8157-0546-8

Contents

Preface..............................................................................................................................................xiAcknowledgments......................................................................................................................................xiiiChapter 1 Why and How Are We Studying Health Insurance in the Developing World?......................................................................1Chapter 2 A Review of the Evidence...................................................................................................................13Chapter 3 Low-Cost Health Insurance Schemes to Protect the Poor in Namibia...........................................................................33Chapter 4 Ghana's National Health Insurance Scheme...................................................................................................58Chapter 5 Impact of Health Insurance on Access, Use, and Health Status in Costa Rica.................................................................89Chapter 6 Health Insurance and Access to Health Services, Health Services Use, and Health Status in Peru.............................................106Chapter 7 The Impact of Health Insurance on Use, Spending, and Health in Indonesia...................................................................122Chapter 8 The Impact of Rural Mutual Health Care on Health Care Use, Financial Risk Protection, and Health Status in Rural China.....................137Chapter 9 Colombia's Big Bang Health Insurance Reform................................................................................................155Chapter 10 Main Findings, Research Issues, and Policy Implications...................................................................................178Editors and Authors..................................................................................................................................199Index................................................................................................................................................205

Chapter One

Why and How Are We Studying Health Insurance in the Developing World?

Maria-Luisa Escobar, Charles C. Griffin, and R. Paul Shaw

More than 2 billion people live in developing countries with health systems afflicted by inefficiency, inequitable access, inadequate funding, and poor quality services. These people account for 92% of global annual deaths from communicable diseases, 68% of deaths from non-communicable conditions, and 80% of deaths from injuries. The World Health Organization (WHO) estimates that more than 150 million of these people suffer financial catastrophe every year, having to make unexpected out-of-pocket expenditures for expensive emergency care (WHO various years).

Within countries, the burden of dysfunctional health systems is disproportionately felt by the poorest households. Their access and use of services, such as immunizations and attended deliveries, tend to be half those of richer households. They have limited recourse to purchase quality services from private providers. Their enrollment in health insurance tends to be marginal. And they are unable to shield themselves from catastrophic health expenditures by drawing on accumulated wealth.

In view of these shortcomings, policymakers in many low- and middle- income countries are debating the virtues of scaling up health insurance to improve health outcomes. Major international conferences have been convened in Berlin (2005) on social health insurance in developing countries and in Paris (2007) on social health protection in developing countries. Regional conferences have followed, as in Africa in 2009. Related to these initiatives, the World Health Assembly passed a policy resolution whereby the WHO would advocate formally mandated social health insurance to mobilize more resources for health in low-income countries, pool risk, provide more equitable access to health care for the poor, and deliver better quality care (WHO 2005a).

All rich countries have adjusted their health finance systems to reduce out-of-pocket expenditures for health, which plunge as per capita income rises across countries (table 1.1). In terms of purchasing power parity (PPP), our preferred measure, per capita gross national income (GNI) is 29 times higher in the richest group than in low-income countries, but health spending per capita is 63 times higher. The share of gross domestic product (GDP) devoted to health more than doubles, the governments' share in the total rises, and the burden on individuals plummets as out-of-pocket spending falls as a proportion of the total. The bottom of table 1.1 shows how much this result reflects the situation in South Asia because of its large share of the total low-income population. The situation is slightly less dire in Africa, but only a bit.

Rich countries achieve these results through general revenue tax financing in support of national health insurance or subsidies for specific groups (such as the poor or the elderly), payroll taxes to support social health insurance, or, most commonly, some combination of both. Rich countries provide prepaid entitlement to health care benefits, reduce vulnerability to the expenses of care at times of illness or injury (financial risk protection), and use copayments and deductibles chiefly to manage demand rather than to raise revenue. They seek to reduce the discontinuity of care so common when people are navigating the system on their own and paying out of pocket at each point of contact. For the most part, richer countries have also separated financing from the provision of care, depend on a mix of public and private providers that are reimbursed through the insurance system, and rely increasingly on primary care providers as gatekeepers to more expensive higher level services. In a nutshell, poor countries want to mimic these successful and desirable behaviors of rich countries sooner rather than later. Mysteriously, donors have historically financed the direct delivery of health services in poorer countries with almost no attention paid to helping them build sustainable financial and purchasing institutions that could emulate some of the core successes of richer countries.

Whatever policymakers and donors want to do or think they should do to emulate successful health financing reforms, there are knowledge gaps that create enormous risks of failure for any reformer. This book attempts to begin filling some of them, but much more work remains.

The widest knowledge gap concerns the impact of health insurance on health status. Do people with health insurance in low- and middle-income countries, or even rich countries, have better health status indicators than those without? Evidence from rich countries suggests yes (box 1.1). But what about low- and middle-income countries? An affirmative on this issue would surely seem essential to consider health insurance as a health policy intervention rather than simply as a financial protection intervention. The vast array of people involved in health care because they want to improve health—nutrition advocates, family planning advocates, tuberculosis and AIDS activists, vaccine supporters, Millennium Development Goal supporters, health systems improvers—would have to see health insurance as an intervention that would be more effective in improving health outcomes than other directly focused options. Obviously, carrying a health insurance card by itself does not make one healthier, but if that card increases the use of appropriate services, makes a person more likely to access new proven technologies, creates incentives for providers to deliver the right services, and equalizes use among the rich and the poor, most analysts would be satisfied that it can have a powerful impact on improving health. They then can devote themselves to making sure the services work.

A second knowledge gap concerns the impact of health insurance on out-of-pocket expenditures for health. Do people with health insurance have lower out-of-pocket spending than those who do not, especially when they are struck by health emergencies? Do the uninsured poor pay a higher proportion of their income for health care than the rich? When out-of-pocket spending is the principal means of securing health care, emergencies result in people borrowing, selling assets, not getting needed care, and engaging in other coping mechanisms. A high proportion of out-of-pocket spending also leads to poorer households spending more of their income on health care than richer households do, just as they spend a higher proportion on other necessities, like food and shelter. Health insurance should address this problem, yet the empirical evidence is slight in our focus countries. The more one explores this issue, the more it becomes apparent that success depends on the design of the program and who is covered; health insurance is not a homogeneous product. A yes on reducing out-of-pocket spending would be essential to argue that health insurance can help prevent people from sliding into health-related poverty.

We can stop with those two questions. Both must be positive to even consider health insurance as a sensible health policy tool in low- and middle-income countries. There are many other practical questions of implementation, but they reside in the realm of sufficiency, not necessity, for considering insurance as a health policy option rather than just a financial protection option.

Objectives of this study and how it was conducted

This study aims to contribute to current policy debates on scaling up health insurance in low- and middle-income countries by shedding light on these two issues: its impact on measures of health status and reducing out-of-pocket spending. Four objectives guide the research and analysis.

Objective 1. Rigorously review and synthesize published and unpublished studies to determine what we know about the impact of health insurance on access and use of health services, the impacts on financial risk protection, and the methodological and data issues in ascertaining causality.

Objective 2. Undertake new country case studies to assess the impact of health insurance on access and use of health services as well as financial risk protection using the latest data sources and statistical methodologies.

Objective 3. Cast more light on the inclusion of the poorest quintile of the population in health insurance in low- and middle-income countries, as well as the benefits they experience compared with the uninsured poor.

Objective 4. Identify the challenges, risks, and opportunities of undertaking retrospective evaluation of health insurance in developing countries using household data.

Shedding light on these objectives requires more than applying good econometrics. Researchers require a fundamental understanding of how health systems work to know what questions to ask and what models to use to find answers. This requires familiarity with design elements that might affect the measured impacts of health insurance on health outcomes (such as enrollment criteria, benefits entitlements, and copayments). Beyond this, however, the study does not assess whether the organizational design of health insurance in different countries is the most efficient or most cost-effective arrangement in satisfying clients, providing quality care, paying providers, or being financially sustainable over the long term. These issues, while important, are complex and demanding enough to require another volume. In short, this study focuses on impacts of health insurance schemes as presently designed and implemented, not what such schemes might accomplish if implemented differently.

Reading this book may raise more questions than it answers, which is good, as we want to present the evidence available today on the topic. We began by identifying low- and middle-income countries that had experienced insurance reforms of interest. We narrowed the list according to whether data existed that could be used to measure what happened at the household level in response to these insurance reforms. We sought researchers who knew the country well and were qualified to do the work. We paired them with advisers and peer reviewers who would commit to read and advise as drafts of the chapters took shape. We tried to keep all of the individual projects advancing along the same timeline and hoped that the ensuing chaos would result in a good collection of work. We did not have the luxury of perfection in any part of this process.

There are many technical impediments to showing an impact of health insurance on anything. These are discussed in the literature review in chapter 2. For some of the chapters readers may conclude that the evidence provided does not get far past associations; in other chapters the evidence may look conclusive that causation has been established. The consistent theme that there is an impact—despite the variety of situations, data, methods, and policies examined—becomes inescapable after reading all the chapters.

Country case selection

Four considerations guided our selection of country case studies. First, we sought countries with sufficiently diverse backgrounds to shed light on the extent findings could be generalized across different contexts. Second, to gain insight into impacts of scaling up health insurance for relatively disadvantaged or poor households, we sought countries with a pro-poor orientation in the design and implementation of health insurance. Third, we sought countries with sufficiently well developed surveys or data systems that would facilitate rigorous statistical analysis of impacts of health insurance on measures of health status and financial risk. Fourth, we sought researchers with a solid knowledge of health insurance who were capable of performing complex statistical modeling to tease out causal impacts. Table 1.2 provides summary data on the seven countries in this volume. We have two giant countries in the mix, China and Indonesia; however, except for Namibia and Costa Rica, all are sizeable. The data used in the chapters are nationally representative except in Namibia, Ghana, and China. There are some important differences across countries in the state of health, income, and health spending, but perhaps the widest range lies in out-of-pocket spending, ranging from 3%–8% of total health spending in Namibia and Colombia to over 50% in China. It is low in Namibia because of good penetration of private insurance; it is low in Colombia because of high government spending, primarily through its insurance system.

Namibia. The Namibian health insurance industry is better developed than that of most Sub-Saharan African countries. It is organized primarily into nonprofit medical aid funds—about one-third mandatory social health insurance funds and about two-thirds voluntary, private plans. Many of the funds are closed, with membership limited to employees in a particular firm or industry or to government civil servants. This has resulted in large disparities in enrollment across socioeconomic categories; only 5% of individuals in the poorest quintile are enrolled, compared with 70% of individuals in the richest quintile. While some private insurance plans aim to broaden the insured population through low cost plans, the challenge is huge because of the country's high prevalence of HIV/AIDS, estimated at 20% for people ages 15–49, concentrated largely among the poor (Feeley, Preker, and Ly 2007).

The case study assesses differences in the consequences of health shocks between the insured and uninsured—stemming from death, hospitalization, weight loss, and HIV/AIDS—specific to households in different income quintiles. The impact of health insurance has been assessed using multiple regression analysis, using 2006 survey data that include both socioeconomic and biomedical information.

Ghana. In 2003 the government passed the National Health Insurance Act, with a vision of insuring 40% of the population by 2010 and 60% by 2015. About 60% of the population was enrolled by 2008, exceeding expectations, with the success attributable to the generous benefit package and prior familiarity with enrolling households in district level mutual health organizations. Other African countries are closely watching Ghana's attempt to scale up health insurance, given the far reaching implications for raising funds, purchasing, and providing care to a largely poor population.

The case study applies a pre-post evaluation design in two districts, one classified as deprived, the other as less deprived. The impact of the health insurance reform is assessed using pre-post bivariate comparisons of key indicators, multivariate regression analysis, and a tentative application of propensity score matching analysis (tentative because of the small sample sizes), using data from a baseline household survey in 2004 and an endline survey in 2007.

Costa Rica. This country has become a benchmark of health insurance attaining wide coverage with no copayments, based on a direct delivery model. Social health insurance was introduced in 1950, and the Universal Coverage Act passed in 1961. Since then, health insurance coverage grew from about 18% in 1961 to 45% in 1971, 60% in 1975, and a high of 92% in 1990. In 2009 about 88% were covered, although the surveys used by the authors in this book put coverage closer to 81% in 2006.

The case study looks at the 19% of the population without health insurance in the 2006 surveys to establish differences in their health status and other characteristics and to investigate whether their health-seeking behavior and results are different. The impact of health insurance on health and related behaviors has been assessed using instrumental variables and data from the 2006 National Health Survey; expenditure results are based on the 2004 Income and Expenditure Survey; and a database of hospital discharges from 2006 provides a unique perspective on how the insured and uninsured use the system differently when they are sick.

Why study a country where everyone is either insured or, if they are not, have equal access to hospital care if they need it? One would not expect to see differences in financial protection in such a system for sure, but because we are interested also in health outcomes, it might be a unique opportunity to see whether not being covered by the formal insurance program has any impact even with Costa Rica's equal access provision.

(Continues...)

Excerpted from The Impact of Health Insurance in Low- and Middle-Income Countriesby Maria-Luisa Escobar Charles C. Griffin R. Paul Shaw Copyright © 2010 by The Brookings Institution. Excerpted by permission of BROOKINGS INSTITUTION PRESS. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Edition details

• Publisher: Brookings Institution Press
• Publication date: 2011
• Language: English
• ISBN 10: 0815705468
• ISBN 13: 9780815705468
• Binding: Paperback
• Number of pages: 239
• Editor: Escobar Maria-Luisa