91323 SUSTAINING UNIVERSAL HEALTH COVERAGE IN FRANCE: A PERPETUAL CHALLENGE DISCUSSION PAPER JUNE 2014 Helene Barroy Zeynep Or Ankit Kumar David Bernstein SUSTAINING UNIVERSAL HEALTH COVERAGE IN FRANCE: A Perpetual Challenge Helene Barroy, Zeynep Or, Ankit Kumar, David Bernstein June 2014 i Health, Nutrition and Population (HNP) Discussion Paper This series is produced by the Health, Nutrition, and Population (HNP) Family of the World Bank's Human Development Network (HDN). The papers in this series aim to provide a vehicle for publishing preliminary results on HNP topics to encourage discussion and debate. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations or to members of its Board of Executive Directors or the countries they represent. Citation and the use of material presented in this series should take into account this provisional character. For information regarding the HNP Discussion Paper Series, please contact the Editor, Martin Lutalo at mlutalo@worldbank.org or Erika Yanick at eyanick@worldbank.org. © 2014 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 All rights reserved. ii Health, Nutrition and Population (HNP) Discussion Paper Sustaining Universal Health Coverage in France: A Perpetual Challenge Helene Barroy.a Zeynep Orb Ankit Kumarc David Bernstein a World Bank, Washington DC, USA b IRDES, Paris, France c Formerly OECD, Paris, France d Independent Consultant, Paris, France Abstract: While Universal Health Coverage (UHC) offers a powerful goal for a nation, all countries- irrespective of income- are struggling with achieving or sustaining UHC. France is a high-income country where health coverage is in effect universal. Health-related costs are covered by a mix of mandatory Social Health Insurance (SHI) and private complementary schemes, while benefit packages are comprehensive, uniform and of good quality. France provides some of the highest financial protection among countries in the OECD. Still, under pressure to sustain UHC without compromising equity of access, the system has been fine-tuned continually since inception. Much can be learned from France’s experience in its reforms toward better fiscal sustainability, equity and efficiency. The main purpose of the study is to assess major challenges that France has faced for sustaining UHC, and to share its experiences and lessons in addressing system bottlenecks to benefit less developed countries as they embark on the path to UHC. Keywords: Universal Health Coverage, Health Financing, France Disclaimer: The findings, interpretations and conclusions expressed in the paper are entirely those of the authors, and do not represent the views of the World Bank, its Executive Directors, or the countries they represent. Correspondence Details: Helene Barroy, World Bank, 1818 H Street, Washington DC, USA, hbarroy@worldbank.org; www.worldbank.org Table of Contents TABLE OF CONTENTS ......................................................................................................................IV ACKNOWLEDGMENTS .....................................................................................................................VI ACRONYMS......................................................................................................................................VII EXECUTIVE SUMMARY: LESSONS FROM FRANCE’S EXPERIENCE ................................................. 1 INTRODUCTION ..................................................................................................................................... 2 PURPOSE AND STRUCTURE ............................................................................................................ 2 SECTION 1: UHC IN FRANCE: A LONG-STANDING OBJECTIVE ........................................................ 4 1.1 UHC: THE ACHIEVEMENT OF A CENTURY ........................................................................... 4 1.2 COMPREHENSIVE BENEFIT PACKAGE WITH GOOD FINANCIAL PROTECTION ............... 6 1.3 FROM A PURE EMPLOYMENT-BASED SYSTEM TO A DIVERSIFIED FUNDING MODEL .... 7 1.4 A PRIVATE AND PUBLIC MIXED DELIVERY APPROACH ...................................................... 9 SECTION 2: THE FISCAL CHALLENGE .............................................................................................. 11 2.1 HIGH HEALTH SPENDING AND PERSISTENT DEFICITS .................................................... 11 2.2 INTRODUCING TAXES TO DIVERSIFY AND INCREASE REVENUES FOR HEALTH .......... 12 2.3 SETTING NATIONAL SPENDING TARGETS AS A WAY TO MANAGE EXPENDITURE ....... 15 2.4 REGAINING CONTROL OVER HEALTH EXPENDITURES .................................................. 17 SECTION 3: THE EQUITY GAP ............................................................................................................ 18 3.1 PERSISTENT HEALTH INEQUALITIES IN A HIGH-INCOME SETTING................................ 18 3.2 POLICY INSTRUMENTS USED TO REMEDY INEQUALITIES IN ACCESS TO CARE ......... 19 ALD ............................................................................................................................................... 22 CMU and CMUC ............................................................................................................................ 22 AME .............................................................................................................................................. 22 ACS ............................................................................................................................................... 22 Effectiveness of Different Strategies............................................................................................... 22 3.3 EQUITY CONCERNS OF RECENT STRATEGIES................................................................ 24 SECTION 4: EFFICIENCY AND PAYMENT IN SERVICE DELIVERY ................................................... 26 4.1 A DOMINANT FFS MODEL................................................................................................... 26 4.2 INVESTING IN PRIMARY CARE ........................................................................................... 27 Introducing Gatekeeping ................................................................................................................ 27 Performance-Based Approach ....................................................................................................... 27 iv Encouraging Multidisciplinary Group Practice ................................................................................. 28 4.3 REFORMING THE HOSPITAL SECTOR TO IMPROVE EFFICIENCY .................................. 29 Activity-Based Payment ................................................................................................................. 29 New Corporate Governance for Hospitals ...................................................................................... 31 SECTION 5: REFORMING HEALTH SYSTEM GOVERNANCE ............................................................ 32 5.1 THE CREATION OF RHAS ................................................................................................... 32 5.2 GREATER LOCAL REGULATION OF HEALTH CARE SUPPLY ........................................... 32 CONCLUSION ...................................................................................................................................... 35 REFERENCES .................................................................................................................................. 36 v ACKNOWLEDGMENTS This report was prepared by Helene Barroy (Health Economist, HDNHE, World Bank), Zeynep Or (Economist, Institute for Research and Information in Health Economics [IRDES], France), Ankit Kumar (Formerly, Economist, OECD), and David Bernstein (Consultant). It is a synthesized version of five separate studies conducted on France’s experience with universal health coverage (UHC), including the sequence of UHC reforms, fiscal sustainability, the equity gap, issues of efficiency and payment systems, and sectorial and hospital governance. The study is part of the Japan–World Bank Partnership Program for Universal Health Coverage, which documents lessons from UHC reforms across the globe. The Program was conceived as a joint effort by the government of Japan and the World Bank to support low-and middle-income countries in their aspirations for UHC, following the 50th anniversary of Japan’s own achievement of UHC in 1961. The authors would like to express their sincere appreciation to the Ministry of Finance of Japan, which provided financial support for the study. They would also like to thank the Ministries of Health and of Foreign Affairs of France, whose support and guidance were instrumental in the study. The authors are grateful to Akiko Maeda (Lead Health Specialist and Team Leader, HDNHE, World Bank), Michael Reich (Professor, Harvard School of Public Health), Cheryl Cashin (Consultant, World Bank and Joint Learning Network), Joseph Harris (Associate Professor, Boston School of Public Policy), and Edson Araujo (Economist, World Bank), for their strategic guidance and invaluable support under the Partnership Program, as well as to Mark Pearson (OECD) and Michael Borowitz (formerly OECD) without whom this study could not have happened. The authors are indebted to the study’s peer reviewers, Felix Faucon (Ministry of Health, France), Professor Naoki Ikegami (Keio University of Japan), Jerry La Forgia (World Bank), and Valerie Paris (OECD) for their excellent comments and feedback. The team is particularly grateful to Paul Dourgnon (IRDES) for his contribution to the equity section. The team would also like to thank many hospital practitioners who have supported, in many ways, the elaboration of the hospital section and shared their experiences of diagnosis-related groups. These include managers of the Necker and Foch Hospitals (respectively, Serge Morel and Sylvain Ducroz), Ines Gravey for synthesizing the effects of reforms at Necker, Florence Veber (APHP), and Catherine Tainturier (Foch). Shazia Amin is also to be thanked for her editing support, as well as Daniela Hoshino for her administrative support. The authors are grateful to the World Bank for publishing this report as an HNP Discussion Paper. vi ACRONYMS ABP Activity-Based Payment ACS Aide à l’Acquisition d’une Complémentaire Santé (Complementary Health Insurance Vouchers) ALD Exonération pour Affection Longue Durée (Long-term Illness Scheme) AME Aide Médicale d’État (State Medical Assistance) CAPI Contrats d’amélioration des pratiques individuelles (Contract for Improved Individual Practice) CMU Couverture Maladie Universelle (Universal Medical Coverage) CMUC Couverture Maladie Universelle-Complémentaire (Public Complementary Insurance Scheme) CNAMTS Caisse Nationale de l’Assurance Maladie des Travailleurs Salariés (National Health Insurance Fund for Salaried Workers) CSG Contribution Sociale Generalisée (General Social Contribution) DRG Diagnosis-related Group FFS Fee for Service GP General Practitioner HCFA-DRG Health Care Financing Administration–Diagnosis-Related Group IRDES Institut de Recherche et Documentation en Economie de la Santé (Institute for Research and Information in Health Economics) LMIC Low- or Middle-Income Country OECD Organisation for Economic Co-operation and Development ONDAM Objectif National des Dépenses de l’Assurance Maladie (National Objective for Health Care Spending) RHA Regional Health Agency (agence régionale de santé) SHI Social Health Insurance UHC Universal Health Coverage VHI Voluntary Health Insurance vii EXECUTIVE SUMMARY: LESSONS FROM FRANCE’S EXPERIENCE France has an employment-based social health insurance (SHI) system that attained universal health coverage (UHC) in 2000 by introducing a state-funded insurance scheme for the poorest segment of the population. Mandatory health insurance provides a broad benefit basket but demands high cost sharing for all essential services. Reliance on private complementary (top-up) insurance for covering these costs has raised concerns over solidarity, redistribution, and efficiency in the system. Under pressure to sustain UHC without compromising equity of access, the system has been fine-tuned continually since inception. Much can be learned from France’s experience in its reforms toward better fiscal sustainability, equity, and efficiency. Resources diversification and expenditure control To assure financial sustainability, sources of funding have been broadened in the past 10 years beyond just payroll contributions to include a broader range of income including that from financial assets and investments. The government has come to play a greater role in managing health expenditure through the introduction of spending targets and rigorous monitoring mechanisms for health insurance funds, deviating from its origins in independent management of social security. Necessity of complementary schemes to address equity concerns The original design, based on cost sharing for all services, required the launching of complementary schemes to address inequalities. The state-funded insurance program for the poorest (Universal Medical Coverage) has improved access to care for these groups, but the cost-effectiveness of other targeted measures, in the forms of vouchers and financial incentives improving access to private complementary insurance, is questionable. France’s experience suggests that a more comprehensive SHI design for essential services may be more cost-effective for tackling inequalities than incremental adjustments through a multiplicity of complementary schemes. More attention to primary care Improving the quality and efficiency of primary care has been recognized as a key strategy for ensuring equity of access and efficiency of the health system in France, as elsewhere. A soft gatekeeping instrument has been introduced to better manage care pathways, but has had limited effects on overall spending. France’s experience with pay-for-performance for general practitioners shows that this approach can improve quality and accountability of providers, but fee-for-service limits the ability to change clinicians’ behavior for ensuring care coordination and efficiency in the long run. Improving efficiency and governance of the hospital sector Payment associated with diagnosis-related groups has helped improve hospital productivity, but France’s experience shows that to deal with the adverse effects of such payment, a strong information system— monitoring both costs and quality of services—is essential. Empowering hospital leadership has proven necessary to drive public hospitals toward a more performance-oriented approach and to improve accountability. To reduce fragmentation in health care provision and improve allocative efficiency, France has also taken steps to move away from centralized governance, shifting responsibility to regional health agencies that oversee all health services for their populations. 1 INTRODUCTION To reach the highest attainable standard of health is an objective that has driven health policy nationally and globally for the last 50 years. In 2012, a resolution of the United Nations reaffirmed this goal by highlighting the importance of promoting universal health coverage (UHC) by improving financial protection and access to health services worldwide. In December 2013, the president of the World Bank, Jim Yong Kim, strongly committed the institution to helping countries advance UHC and proposed, with the World Health Organization, a set of core indicators to track progress toward UHC worldwide. While UHC offers a powerful goal for a nation, all countries—irrespective of income—are struggling with achieving or sustaining UHC. Nearly 30 low- and middle-income countries (LMICs) have made notable progress toward expanding health coverage and financial protection. Almost all high-income countries have achieved UHC but are also facing challenges in financing it efficiently and sustainably. There is growing interest among LMICs to incorporate UHC as one of the goals of their national development plans. Health coverage has several dimensions, including (i) breadth of coverage to ensure access to health services across different population groups (Who is covered?); (ii) depth and scope of coverage in the range of benefits and services offered (What is covered?); and (iii) financial protection against out-of- pocket and catastrophic health spending (World Health Organization, WHO). A UHC-related goal is to establish an inclusive and sustainable health system—one that ensures equitable and affordable access to health care for all members of society, that provides adequate financial protection, and that contributes ultimately to the well-being of individual citizens and of society as a whole. LMICs face multiple constraints in expanding health coverage, including limited financial resources from both public and private sources and low quality of care and productivity of services. Because of the complexity of interactions that influence health coverage, identifying the key constraints and designing feasible solutions to overcome these barriers is daunting. Still, lessons from high-income countries that have expanded coverage and financial protection in past decades could be useful in LMIC settings. France is a high-income country where health coverage is in effect universal. Health-related costs are covered by a mix of statutory insurance and private complementary schemes, while benefit packages are comprehensive, uniform, and of good quality. France provides some of the highest financial protection of health expenses among countries in the Organization for Economic Co-operation and Development (OECD) (OECD 2013). The state of health care in France is generally good but has some apparent contradictions. On the one hand, indicators such as life expectancy (81 years) and healthy life expectancy (72 years) show that the health of the population is good and above the OECD standard (OECD 2013). In particular, women live longer, and the elderly remain in better health than in many other European countries. France also compares well on cardiovascular diseases, while its relative position on mortality due to alcoholism, cirrhosis, and cancer of the cervix is improving. On the other hand, it suffers from a high rate of premature male deaths from accidents and unhealthy habits (smoking and alcoholism), and social and geographic inequalities in health remain substantial for such a high-income setting (Chevreul et al. 2010). France has one of the widest socioeconomic inequalities in health outcomes and in access to health services in Western Europe. Purpose and Structure The main purpose of this case study is to assess major challenges that France has faced—and still faces—for sustaining UHC and to share its experiences and lessons in addressing system bottlenecks to benefit less developed countries as they embark on the path to UHC. After reviewing the main features and the sequencing of France’s UHC strategy (section 1), the study assesses the fiscal challenges associated with UHC financing (section 2). The section characterizes health deficits and analyzes strategies used to diversify sources of revenue in recent years. It also profiles the policies used to curb spending. Section 3 looks at the equity gap. After characterizing the scope of health inequalities, it reviews existing measures and recent reforms to tackle inequalities in access to care. It closes by examining recent cost-control strategies and their potentially adverse effects on financial protection, especially among the worse off. 2 Section 4 looks at efficiency and payment in service delivery—analyzing the main sources of inefficiency in the service delivery system—and presents the contents and the early effects of recently introduced measures to improve system efficiency and quality in primary and acute care, where it pays particular attention to the effects of new payment systems. Section 5 rounds off the document with a review of health governance issues, in a period when—to move away from traditional fragmented and centralized management—France initiated structural reform for shifting decision-making power in the health sector to regional health agencies (agences régionales de santé). 3 SECTION 1: UHC IN FRANCE: A LONG-STANDING OBJECTIVE Summary: France provides health coverage to its entire population through social health insurance (SHI). Health coverage is provided through multiple schemes for the various employment-based groups. The benefit package, which became equal across schemes only recently, is quite broad. Financial protection, though it relies on cost sharing, is among the highest in OECD countries. Most of the population has complementary private insurance to cover copayments. Service delivery is dominated by a fee-for-service (FFS) payment system for a mix of public and private providers. 1.1 UHC: THE ACHIEVEMENT OF A CENTURY France’s current health insurance system was shaped over more than a century. Post–World War II reforms established a wage-based system for the formal sector; civil servants and remaining population groups were gradually covered in the following four decades. The objective of providing coverage for the whole population was finally achieved in 2000 when the state passed the Universal Medical Coverage 1 (CMU) Act for all legal residents, irrespective of work status. The system derives from the tradition of mutuelle-based organizations, in place since the French Revolution (1791). At that time, provident societies (sociétés de secours mutuel) provided their contributing members with financial support in the event of death or disease. The 19th century was marked by the launch of targeted schemes run by the state to cover, among other contingencies, health risks. However, population coverage and the provision of benefits remained low, as the schemes were based on voluntary contributions. After World War I, the first compulsory scheme was set up for industry workers in the Alsace-Lorraine region bordering Germany. This served as a catalyst for the development of a broader insurance system in the 1930s to cover the whole country (the Social Health Insurance Act). The system was based on contributions from employees and employers in industry and business and provided coverage for illness, motherhood, disability, old age, and death. This period in France (characterized as the “welfare state”) was marked by deeper state involvement in social protection. A major stride in establishing a nationwide compulsory health insurance system took place just after World War II. In 1945, a general insurance scheme (the Régime Général) was adopted under the Social Health Insurance Act to provide a compulsory system guaranteeing uniform rights, initially for salaried workers (travailleurs salariés) in five areas: illness, maternity, disability, old age, and death. Nevertheless, the objective of one uniform scheme was not achieved, as very quickly differences arose based on competing interests. Certain socio-professional groups (such as civil servants, miners, and railway workers) already benefiting from insurance with more favorable terms succeeded in maintaining their particular schemes, which still exist. Coverage for various population groups expanded incrementally over the next four decades. General statutory insurance was extended to farmers in 1961 and to the self-employed in 1966, but access to health insurance remained problematic for very low-income groups and people with variable incomes until 2000 when the CMU Act came into effect. This law opened the right to health insurance for all legal residents not previously covered. It also mandated means-tested subsidies for those who cannot pay for their insurance. Thus the goal of providing coverage for the whole population was finally achieved in 2000 (figure 1.1). 1. Couverture Médicale Universelle (CMU) is the name given to the state-funded scheme for the poor providing basic (CMU) and complementary Couverture Maladie Universelle–Complémentaire (Public Complementary Insurance Scheme, or CMUC) package of health services with no copayment and is, for the vast majority of enrollees, a noncontributory program. CMU is different from UHC, which relates to the goal of providing access to health services and improved financial protection to everyone in a country. 4 Figure 1.1 Milestones in France’s Health Insurance System Source: Authors. Today, the three major statutory insurance schemes cover more than 95 percent of the population. The largest one, the Régime Général, insures wage and salary earners and their dependents and covers 2 about 85 percent of the compulsorily insured. It is managed by the Caisse Nationale d’Assurance Maladie des Travailleurs Salariés (National Health Insurance Fund for Salaried Workers, or CNAMTS). The two other schemes, covering the self-employed (Régime Social des Indépendants) and farmers and agricultural employees (Mutualité Sociale Agricole), cover about 13 percent of the population. In addition, 16 small schemes cover certain professional categories, including miners, the clergy, employees of SNCF (the national rail company), and the central bank (1 percent) (figure 1.2). Enrollment in insurance schemes is compulsory and automatic for workers (covering their spouses and dependent children). Consumers cannot choose their scheme or insurer, and cannot opt out. With the exception of CMU, enrollment follows employment status, which means that an agricultural employee automatically falls under the agricultural scheme and is not allowed to switch to the general insurance scheme. There are thus no competing health insurance markets for core social health coverage. 2. CMU beneficiaries are also part of the Régime Général. 5 Figure 1.2 Population Coverage by Statutory Health Insurance Affiliation, 2013 1% 2% 5% 7% General fund (salaried) Agricultural fund Self employed fund Other funds (miners, railways…) Universal coverage fund 85% Source: Authors. 1.2 COMPREHENSIVE BENEFIT PACKAGE WITH GOOD FINANCIAL PROTECTION The French health system is known for providing extensive service coverage, with solid financial protection. While population groups are covered through separate schemes and funds, the schemes are tightly harmonized, which was previously a major issue and finally achieved only quite recently. The same benefit package now applies for all schemes and for the whole country, including CMU beneficiary groups. Each SHI member is legally entitled to the same package regardless of his or her scheme. The standard package under the statutory health insurance system includes a wide range of goods and services (box 1.1). Box 1.1 Service Coverage under SHI in France, 2013 • Hospital care and treatment in public or private institutions providing health care, rehabilitation, or physiotherapy. • Outpatient care provided by general practitioners (GPs), specialists, dentists, and midwives. • Diagnostic services and care prescribed by doctors and carried out by laboratories and paramedical professionals (including nurses, physiotherapists, and speech therapists). • Pharmaceutical products, medical appliances, and prostheses prescribed and included in the positive lists of products eligible for reimbursement. • Prescribed health care–related transport. Source: Adapted from Chevreul et al. 2010. The benefit package is comprehensive, with cost sharing for almost all its services but still high financial protection relative to other OECD countries. In the breakdown for total health expenditure for common goods and services for 2011, SHI covered 75.5 percent; complementary voluntary insurance 13.7 percent; and direct household spending 9.6 percent. SHI spending fell from 76.7 percent of total health expenditure in 2000 to 75.5 percent in 2011. Over the same period, the share covered by voluntary health insurance (VHI) increased from 12.4 percent to 13.7 percent (figure 1.3). 6 Figure 1.3 Total Health Expenditure by Source of Funding, 2011 Source: Authors, from Eco-santé data, 2011. In general, the insured are expected to advance full costs for ambulatory care and to claim reimbursement from insurance funds based on predefined rates (negotiated tariffs). Copayments are fixed rates defined by insurance funds and the Ministry of Health. The same rates apply regardless of the 3 scheme or the patient’s income. The share of costs left to patients varies by type of service and by type of medication: cost sharing for patients amounts to 30 percent of the cost of ambulatory visits, about 20 percent of the cost of hospital care, and 35–85 percent of the cost of (approved) treatment drugs. The rate of reimbursement by statutory insurance is determined according to the effectiveness of the prescription drug and the seriousness of the disease treated: 100 percent for rare, highly effective and expensive drugs; 65, 35, or 15 percent for those of diminishing therapeutic value; and no reimbursement for ineffective drugs. As the system relies heavily on cost sharing, the insured are encouraged to enroll in private complementary (top-up) insurance to cover costs not supported by statutory insurance: 94 percent of the population is now enrolled in VHI schemes through private firms, provident firms, and mutuelles, all of which generally cover full copay, though significant disparities persist across complementary contract provisions. A growing number of patient groups are eligible for a third-party payment system (tiers payant), which exempts them from paying the entire cost of services at the point of delivery. Additionally, some population groups are exempt from bearing the direct costs. Patients with long-term chronic diseases (under the Exonération pour Affection Longue Durée [ALD], or Long-Term Illness Scheme), pregnant women, and newborns are exempt from cost sharing for services related to their condition, while CMU beneficiaries do not carry any user charges as these are systematically covered by the CMUC scheme (see section 3.2). 1.3 FROM A PURE EMPLOYMENT-BASED SYSTEM TO A DIVERSIFIED FUNDING MODEL On expenditure relative to its economy, France is the third highest for health care among OECD countries (figures 1.4 and 1.5). In 2011, France spent 11.6 percent of gross domestic product (GDP) on health care, against the OECD average of 9.3 percent and behind the United States (17.7 percent) and the 3. With the exception of a few small regimes that may offer more generous reimbursement for some services. 7 Netherlands (11.9 percent). France also ranks above the OECD average on health spending per capita at US$4,118 (in 2011 parity purchasing power dollars), compared with the OECD average of US$3,339 (figure 1.5), although French health spending per capita was less than half that of the United States (US$8,508). Figures 1.4 and 1.5 Health Expenditures as a Share of GDP and Per Capita Source: OECD 2013. Note: France is highlighted in grey. Until the end of the 1990s, SHI financing depended almost exclusively on payroll contributions from 4 employers (63 percent) and employees (32 percent). Since 1998, most employee payroll contributions have come through earmarked taxes, known as the General Social Contribution (Contribution Sociale Generalisée, or CSG). The CSG, earmarked for health, is levied at 5.25 percent of income and income from capital and 3.95 percent of unemployment and disability benefits, and is now one of the main sources of statutory health insurance funding (36 percent). In addition, specific taxes (on tobacco and 4. Commission of Social Security Accounts, 1990 figures. 8 alcohol consumption and the pharmaceutical industry) now account for 13 percent of funding for health insurance (figure 1.6). Figure 1.6 Sources of Funding for Statutory Health Insurance 5.9 7.9 3.8 1.7 4.1 12.6 35.4 34.6 35.7 3.5 98.4 57 49.9 47.9 1968 1995 2000 2011 Wages contribution from employers from workers CSG Other Taxes Other Source: CNAMTS 2011. 1.4 A PRIVATE AND PUBLIC MIXED DELIVERY APPROACH Health care delivery is shared between the private and public sectors. Primary care is quasi-exclusively delivered through self-employed physicians on an FFS basis. GPs working in the ambulatory sector contract with health insurance funds and are paid according to a negotiated FFS schedule. GPs who agree to charge on the basis of the nationally negotiated fee (called “sector 1” contractors) get, in return, their social contributions, including pension, paid by CNAMTS. Others are authorized to charge higher fees (“sector 2” contractors) but must purchase their own pension. About 65 percent of specialists and 85 5 percent of generalists work in sector 1. 5. The creation of sector 2 in 1980 did not reduce total expenditure of the statutory health insurance fund, as was the intention. Consequently, access to that sector has been limited to 1,000 new doctors a year. 9 Table 1.1 Tariffs and Reimbursement Mechanisms for Self-Employed Doctors in Ambulatory Care (€) 1 Tariff Standard Flat Extra Total Covered by 2 3 copayment deductible billing copayment statutory health insurance Sector 1 doctors, fixed tariff GP 23 6.9 1 0 7.9 15.1 Specialist 25 7.5 1 0 8.5 16.5 Sector 2 doctors,4 no-limit billing (example) GP 40 6.9 1 17 24.9 15.1 Specialist 50 7.5 1 25 33.5 16.5 Source: Authors. Note: 1. December 2013 tariff for a basic consultation. 2. In principle this cannot be covered by complementary insurance. 3. Copayments are typically covered by private complementary insurance, but within a limit for sector 2 doctors (depending on the contract). 4. Sector 2 physicians (15% of GPs and 35% of specialists) can set their tariffs freely (there is no limit). The patient will be reimbursed by public insurance on the basis of the negotiated/fixed tariff. Hospital care is split among the public sector (with two-thirds of hospital beds) and private for-profit and not-for-profit facilities. The hospital payment system has evolved over the last two decades from global budgets in public hospitals and an inflationary itemized billing system in private clinics to a more performance-oriented and public–private alignment model through the introduction of an activity-based payment system in all hospitals. Traditionally, the French health care system has been characterized by strong central governance. Regulation and management are predominantly divided between the state and the statutory health insurance funds. The institutions at regional and département level (metropolitan France is divided into 22 regions and 96 departments) also play an executive role, but despite several reforms to reinforce the role of the regions, decision making remains quite centralized. The state sets out sector-level expenditure targets, determines the level of health care provision and training, regulates care quality, and defines priority areas for national programs. The salaries and working conditions of public hospital staff as well as prices of diagnosis-related groups (DRGs) are regulated by the government. The health insurance funds play the main role in defining the benefit baskets; regulating the prices of procedures, drugs, and devices, which will be reimbursed to patients; and defining the levels of copayment. The sickness funds are also in charge of setting tariffs for health professionals in private practice via collective negotiations with the professionals’ unions. Doctors working in private hospitals contract with health insurance funds and are paid according to the negotiated FFS schedule, while those in public hospitals are salaried. 10 SECTION 2: THE FISCAL CHALLENGE Summary: Revenues matter for the financing of health, particularly in the context of SHI. France’s experience is of particular interest to other countries as it has shifted from a purely wage-based model at the time the system was introduced to a more diversified financing model in recent years, aiming to generate resources through different taxes. On the expenditure side, France has also taken some steps to curb the size of health care deficits over the past few decades. The combination of these measures has improved financial sustainability, though deficits remain a feature of the system. 2.1 HIGH HEALTH SPENDING AND PERSISTENT DEFICITS Health is the second-largest area of public spending in France, after social protection (OECD 2013). In 2011, health accounted for 14.7 percent of total government spending (close to the OECD average). France also has the second-highest level of government spending among OECD countries, together giving the country one of the largest shares of public spending on health relative to GDP among OECD countries (figure 2.1). Figure 2.1 Government Health Spending in OECD Relative to Total Government Spending as a Share of GDP, 2011 (or latest available) 10.0 DNK FRA USA DEU NZL Government health spending (% of GDP) 9.0 AUT JPN NOR CAN GBR SWE BEL 8.0 CHE ISL ITA ESP OECD PRT FIN 7.0 CZE SVN AUS IRL GRC SVK LUX 6.0 HUN EST POLISR 5.0 TUR KOR 4.0 MEX 3.0 2.0 1.0 0.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 55.0 60.0 General government spending (% of GDP) Source: Authors, from OECD data. Health deficits have been substantial in recent years and are a major part of social security deficits, about half of which stem from health insurance costs. Health care and other social security deficits (net of health care) became persistent and onerous over the 2000s (above € 10 billion) (figure 2.2). 11 Figure 2.2 Social Security and Health Deficits since 1998 (€ billion) Social Security and Health financial equilibriums since 1998 (General Scheme) 0.5 0.7 1.2 -2.4 -0.7 -1.6 -2.1 -6.1 -11.1 -11.6 -8 -5.9 -4.6 -4.4 -10.6 -11.6 -8.6 -2.5 -3.5 -10.2 -8.7 -9.5 -10.2 -11.9 -11.6 -17.4 -20.3 -23.9 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Social Security (not including health) Health Source: Adapted from Social Security Accounts reports. 2.2 INTRODUCING TAXES TO DIVERSIFY AND INCREASE REVENUES FOR HEALTH Although initially the system was almost entirely funded from wage-based contributions, today these account for less than half of France’s health care revenues. The government has sought to diversify sources of financing for health over the last two decades, expanding the base of health financing to lessen reliance on only one key form of revenue—wage-based contributions. The most profound change was the introduction of the CSG during the 1990s. The CSG brought in a basket of taxes applied to a broader range of incomes than just wages. The CSG includes wage income but also extends to incomes from financial assets and investments, pensions, and unemployment and disability benefits. As well as broadening the sources of revenue, one of the motivations of this policy was to reduce labor costs. At the time of its creation, the CSG rate was low, at 1.1 percent of each of these sources of revenue, and was allocated to the family branch of the social security system. The appropriation of the CSG to fund SHI effectively began in 1997, as part of Plan Juppé (box 2.1). Alongside measures to control expenditure (discussed later), this plan sought to reduce wage-based contributions for health care and replace the forgone revenues with an increase in the CSG—that is, a reduction of 1.3 percentage points on employee wage-based contributions was replaced by an increase of 1.0 percentage point on CSG rates across the various sources of income. The CSG was also widened to include gambling revenues at this time. Following this, a major push toward diversifying revenues for health care came in 1998. The employee wage-based contribution was reduced by 4.75 percentage points, and the CSG on wage income and capital income was increased by 4.1 percentage points. At the same time the CSG on pensions and unemployment benefits was raised by 2.0 percentage points, and contributions from these social benefits for health were almost entirely removed. Since then, the remainder of the wage-based contribution toward health care revenue has been almost entirely financed by contributions levied on employers. The CSG now accounts for 38 percent of health care revenues. 12 Box 2.1 Plan Juppé First announced in 1995, Plan Juppé sought to reform the social security entitlements of public sector workers (particularly the pensions of transport and energy workers), with a view to bringing them closer to those in the private sector. At the same time, it also aimed to establish spending targets in health care while reforming the governance of social security institutions and of the hospital sector. Led by unions but with a coalition of other bodies, opposition to these reforms (particularly among public sector workers) was significant, and prompted major strikes across France in 1995. While the controversial measures relating to social security and retirement of public sector workers did not succeed, the plan did result in the introduction of policies aimed at bringing social security deficits under control, including the diversification of revenue sources. Though initially introduced under the administration of a prime minister from the major center-right party (Rassemblement pour la République [RPR]), these measures have been maintained through multiple changes of government. Source: Authors. Earmarked taxes have come to contribute a greater share of health financing needs. They have been the fastest growing source of financing, now accounting for some 13 percent of revenues for health care, against just 4 percent a decade ago. These more than 20 types of taxes are known as impôts et taxes affectés. In 2011, 55 percent of these earmarked taxes financed health insurance funds for a total of some €27.8 billion (10 percent of SHI revenues). These taxes are taxes on enterprises, such as five different taxes on pharmaceutical companies, taxes on company cars, or a share of a global tax above a certain level of net sales for all companies. There are also taxes on consumption or behavior, such as a share of value-added tax, a tax on private complementary health insurance, and taxes on tobacco and alcohol (figure 2.3 and table 2.1). Figure 2.3 Allocation of Earmarked Taxes, 2012 (€ billion and %) Source: High Council for Social Security Financing report, October 2012. 13 Table 2.1 List of Earmarked Taxes Related to Health, 2012 (€ billion) Portion of VAT 10.6 Tax on tobacco 8.0 Taxes on complementary insurance 2.9 Tax on companies with net sales above 0,76 M€ 1.4 Pharmaceutical taxes 1.1 Tax on motor insurance 1.1 Tax on alcohol 1.0 Tax on company cars 1.0 Gaming taxes 0.5 Tax on oils and flours 0.2 Total of earmarked taxes affected to Health 27.8 Source: High Council for Social Security Financing report, October 2012. The introduction of the CSG has also marked a worthwhile shift toward better equity in financing health care. The share of labor income (principally wages) in household incomes fell from 80 to 71 percent between 1970 and 2011 (figure 2.4). At the same time, income from capital and social benefits came to form a larger share of household incomes—29 percent in 2011. The shift from taxing only wage-based incomes to a broader definition of incomes has thus improved the equity of revenue raising for health, as wealthier individuals generally have higher incomes from capital and other financial assets and are more likely to have accumulated substantial social benefits during their working lives. Nonetheless, calls remain for more ambitious changes to financing social security. The CSG collects the equivalent of around 5 percent of GDP, against the 2.9 percent (in 2011) of income tax. While exemptions have been set for the unemployed and retirees, the tax is applied to a fixed proportion of incomes and does not increase for those on higher incomes (that is, it is not progressive). A reform to consolidate and harmonize the tax regime is under Government discussion in 2014. 14 Figure 2.4 Household Income Distribution, 1970 and 2011 100% 15% 22% 80% 4% 7% 60% Socia l Benefits Ca pita l income 40% 80% Work income 71% 20% 0% 1970 2011 Source: Data from the Institut national de la statistique et des études économiques (INSEE) with OECD analysis. 2.3 SETTING NATIONAL SPENDING TARGETS AS A WAY TO MANAGE EXPENDITURE Setting an overall expenditure target for health care, known as the National Objective for Health Care Spending (Objectif National de Dépenses de l’Assurance Maladie, or ONDAM), has been a key aspect of the strategy to control health spending. This involves setting an a priori global budget for health each year. For most of its history, the government has not played a proactive role in influencing overall health care spending, with independently operated insurance funds responsible for managing spending in line with their social entitlements to citizens. ONDAM marked a significant break from this tradition, and represents the reassertion of the government’s control of health care spending. ONDAM is now at the heart of the Social Security Financing Law (Loi de Financement de la Sécurité Sociale), specified in monetary terms as the total amount of health spending for the forthcoming calendar year. Once published, it gives all stakeholders a precise spending objective. In 2012, ONDAM was €171 billion (or 38 percent of the total €454 billion in social security spending). The monetary ONDAM target is used to signal the proportion of health spending growth that the government is willing to accept in any given year. For example, the 2013 objective was €174.5 billion, or 2.7 percent growth from 2012. ONDAM’s overall target is split into three subtargets for the main health service providers: ambulatory care, hospitals, and medico-social facilities. Ambulatory care and hospitals absorb the vast majority of ONDAM, with 2013 targets of 45.9 percent and 43.6 percent of total ONDAM, respectively. When setting ONDAM, the government draws up a precise list of savings necessary to meet budget targets (figure 2.5). 15 Figure 2.5 Main Savings Measures Included in ONDAM, 2010 to 2013 (€ million) 2413 2236 50 155 2034 145 1843 271 133 222 Other measures 170 192 259 Ambulatory care providers' tariffs decrease 166 567 200 Measures on drugs excl. Price decrease 510 121 Hospital performance 638 Prescription control 438 675 470 Drugs prices decrease 502 492 697 695 333 425 2010 2011 2012 2013 Source: Social Security Law 2011, 2012, and 2013. Although the creation of ONDAM was seen as a step toward better monitoring of health spending, it has taken over a decade for the policy to begin demonstrating consistent results. In its first five years, ONDAM targets were regularly overrun. As its nonbinding nature meant no stakeholder was obligated to uphold it, the policy did not suffice to constrain SHI spending within budgetary targets (figure 2.6). Therefore, additional measures were put in place (section 2.4) Figure 2.6 Voted ONDAM vs. Realized Health Expenditures (% growth) 8.0% 7.1% 6.4% 6.0% 5.6% 5.6% 4.9% 4.2% 4.0% 4.0% 4.0% Voted 3.5% 3.5% 3.1% Achieved 2.6% 2.7% 2.7% 2.6% 5.3% 4.0% 4.0% 2.0% 3.5% 3.3% 1.5% 3.2% 3.0% 2.9% 2.8% 2.7% 2.6% 2.6% 2.6% 2.5% 2.5% 2.3% 1.7% 0.0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 P 2013 Source: Social Security Accounts reports. 16 2.4 REGAINING CONTROL OVER HEALTH EXPENDITURES For much of their history, statutory health insurance funds have been the key operational entities that pay health care providers for services delivered to patients. While the most important fund (CNAMTS for the formal sector) continues to have considerable influence in the administration of the health care system, reforms over the last few years have seen the state expand its role in determining health care spending. Somewhat unusually for a social insurance–based system, the French Parliament ultimately sets the fiscal parameters within which CNAMTS (and other health insurance funds) are required to maintain spending. A separate Social Security Financing Law has been in place since the late 1990s.; it subjects the expenditure and financing sources of all aspects of French social security (including health, pensions, and family benefits) to annual ratification. This process is now enshrined in the Constitution, making Parliament a major player in health sector developments and reasserting the government’s influence on social security spending. The annual process behind the Social Security Financing Act requires several ministries to work together to control health expenditure. Every year, the Ministry for the Economy produces a draft act, with the Ministry of Health, the Ministry of Labor, and other social security entities. The draft is prepared between June and October by the Social Security Direction (Direction de la Sécurité Sociale), an administrative body under the joint supervision of the Ministry of Budget and the Ministry of Health. Following its publication (as the Projet de loi de financement de la sécurité sociale), the document is debated in Parliament in accordance with the legislative calendar (figure 2.7). Figure 2.7 Timeline of the Social Security Act If the two versions are not similar, June to October : National Assembly preparation of the Within 20 days : and Senate have PLFSS within the National 50 days from the Health and Assembly has to registration date Finance vote a first version to find an admnistrations of the Act agreement October 15th : Within 15 days Mid december : Constitutionnal after receiving the the final Act is due date for draft from the either voted by registring the Assmembly, the the Parliament or PLFSS in National Senate has to vote legislated by an Assemly its own version of emergency decree the Act Source: Article LO111-6 and following, Code of Social Security. The inability to reach ONDAM targets in early years led the administration to conclude that there should be closer monitoring of health expenditures and of ONDAM’s prospects for success during the year, and indeed ONDAM targets have evolved progressively from vague objectives to binding targets. An Alert Committee was set up in 2004 to closely monitor growth of health expenditure. Its mission is to flag to governing bodies, including Parliament, when spending is growing faster than the ONDAM target. Since 2010, the Alert Committee has gained additional powers and can undertake an ex ante evaluation of ONDAM before the draft (projet de loi) of the Social Security Financing Act is submitted to Parliament. If hospital-ONDAM targets are expected to be overrun, DRG tariffs can be adjusted (decreased). In recent years, the budgetary processes ushered in by ONDAM have begun to show signs of improved monitoring of health expenditures as well as better working relations between stakeholders, though it is 17 too early to say that it has permanently enhanced control over health spending. While growth of health expenditures has been decreasing for a decade, the ONDAM target has been met only since 2010. Such growth has decreased to 3 percent since 2010, from a high of 7 percent in 2002 (see figure 2.6). However, it is expected that France will need to achieve a 2.5 percent ONDAM growth rate in forthcoming years to end health deficits. This would, according to the Ministry of Health, represent an average extra annual saving of €2.8 billion over 2013–17—a not inconsiderable challenge. SECTION 3: THE EQUITY GAP Summary: France has worked hard to harmonize different insurance schemes and cover the whole population over the past couple of decades. Despite universal entitlement since 2000, significant inequalities in health and use of health services have persisted. The social insurance system—heavily based on cost sharing—relies increasingly on private complementary insurance for covering costs, raising concerns over solidarity and redistribution. While extension of insurance coverage through state-funded mechanisms (CMU and CMUC) was instrumental in expanding effective access for the worst off, the cost- effectiveness of other targeted measures, such as Complementary Health Insurance Vouchers (Aide à l’Acquisition d’une Complémentaire Santé, or ACS), is questionable. Another important lesson is that cost-control strategies must be continually monitored for their impact on equity and financial protection. 3.1 PERSISTENT HEALTH INEQUALITIES IN A HIGH-INCOME SETTING France has relatively good health outcomes and enjoys one of the highest life expectancies in the world. Nevertheless, the significant improvements in health status over the last decades have not been equally distributed across socioeconomic groups. Compared with other European countries, France had wide socioeconomic inequalities in premature mortality, especially for men with cancer and cardiovascular diseases (Kunst, Groenhof, and Mackenbach 1998), and although reducing health inequalities has been a public health priority since the late 1990s, France’s low ranking among Western European countries remained unchanged a few years ago (Mackenbach et al. 2008). Disparities in health across socioeconomic groups could partly be linked to socioeconomic inequalities in utilization of health care services (Or et al. 2009). France shows relatively equitable use of generalists, while it ranks last among 13 OECD countries for visits to specialists and to dentists (figure 3.1). Recent studies suggest that utilization of preventive services, such as breast cancer and cervical cancer screening, are also largely conditioned by income, and more than in many other European countries (Devaux and de Looper 2012). 18 Figure 3.1 Horizontal Inequity Indexes for Probability of a Specialist Visit in 13 OECD Countries, 2009 (or nearest year) (95 percent confidence interval) 0.10 0.08 0.06 0.04 0.02 0.00 -0.02 -0.04 Source: Adapted from OECD data 2011. Note: The probability of a doctor, GP, or specialist visit is inequitable if the horizontal inequity index is significantly different from zero. It favors low-income groups when it is below zero, and high-income groups above zero. The index is adjusted for need. In 2008, 16.5 percent of the population age 18 to 64 years reported they had forgone health care in the last 12 months for financial reasons (Allonier et al. 2008). Dental health care is of greatest concern (10.7 percent of the population age 18 to 64 years had forgone dental health care in the last 12 months), followed by optical care (4 percent) (Allonier et al. 2008). Forgoing health care is inversely related to income: people in the poorest quintile (under €820 a month) forgo three times more care than people in the richest quintile (more than €2,000 a month) (Allonier et al. 2008). Moreover, the more degraded the health status is, the likelier that care will be forgone. Despite the high emphasis on equity in the system, a number of structural weaknesses may explain why inequity in access to care remains a problem. First, copayments can be quite significant both for ambulatory and hospital care (where specialists often extra-bill), for those who do not have complementary insurance (about 7 percent of the population), and for those with low-cost complementary insurance. The fact that fees charged by almost half of physicians (sector 2) are not regulated means that access to quality care depends partly on ability to pay. Second, there may also be nonfinancial barriers in accessing care. The system is very fragmented and complicated from the point of view of users. Education (beyond income) is an important determinant of care use. Higher education is significantly associated with an increased probability of consulting specialists in France, as elsewhere (Or et al. 2009). Furthermore, there are geographic inequities. The principle of allowing doctors and nurses in the ambulatory sector to practice where they like results in wide variations in supply between and within regions. Yet policies to improve supply in rural and low- income areas have been rather weak because of strong resistance from medical unions. 3.2 POLICY INSTRUMENTS USED TO REMEDY INEQUALITIES IN ACCESS TO CARE As seen, Social Health Insurance does not cover all health care expenses: there are considerable copayments for almost all health services and the chances of overbilling by some practitioners. Given the importance of cost sharing, from its inception the French system introduced protective mechanisms to 19 reduce the financial burden of care for patients suffering from chronic illnesses (ALD). In the past decade, new policy instruments have been added to tackle inequalities in access to basic public and private complementary insurance: CMU, CMUC, and ACS. In parallel, one scheme, AME (Aide Medicale d’Etat, or State Medical Assistance) targeting immigrants in illegal situations was introduced (table 3.1). 20 Table 3.1 List of Equity-Oriented Schemes and Corrective Measures Number of beneficaries (% Scheme Objective Inclusion criteria Income threshold Description Examples of beneficiaries of population) Covers all copayments Medical : diagnosis Cancer patients, asthma, Chronic diseases scheme (Affections de Reduce the financial (sector 1) related to of chronic or serious None CVD, diabetes, mental 9, 500 ,000 (15%) Longue Durée ) burden of disease treatment of the disease problems… specific disease Free for all income People who have not Legal residency and Provide basic statutory below 9534 Provides statutory contributed to Social State-funded Basic Insurance Scheme not being covered insurance to all euros/annum; 8% of health insurance Security either they never 2, 159, 253 (3.3%) (Couverture Maladie Universelle de base ) by another residents income contribution entitlements worked or were working statutory scheme above abroad Reduce financial State-funded Complementary Insurance Below 8593 access problems linked Legal residency and Unemployed, working scheme (Couverture Maladie Universelle euros/annum/capit Covers all copayments 4, 203, 711 (6.5%) to cost sharing under very low income poor… Complémentaire ) a basic scheme Reduce financial Between 8593 and Cash support in the Vouchers for private complementary access problems linked Legal residency and 11600 form of vouchers People just above the low insurance scheme (Aide à l'Acquisition d'une 935, 067 (1.4%) to cost sharing under low income euros/annum/capit purchasing voluntary income scheme threshold Complémentaire Santé ) basic scheme a insurance Enable illegal Below 8593 Provides statutory Illegal immigrants' scheme (Aide Médicale immigrants and their Residency for over 3 Refugees waiting for euros/annum/capit health insurance and 263, 962 (0.4%) d'Etat ) family (children) to months asilum status a covers all copayments have healthcare access Source: Authors. 21 ALD Introduced in 1945, ALD aims to reduce the financial burden of medical care for beneficiaries suffering from chronic diseases. Irrespective of their income status, patients are exempted from copayments for treatments associated with a chronic disease. Initially introduced to cover four groups of diseases (cancer, tuberculosis, poliomyelitis, and mental illness), the scheme was extended over time and now covers 32 groups of diseases. In 2012, over 9 million people were covered by ALD, or 15 percent of SHI beneficiaries. ALD accounts for nearly 60 percent of health expenditures reimbursed by the health insurance fund. Expenditures linked to ALD recorded an average annual growth of 4.9 percent over 2005–10, against 1.8 percent for other health expenditures (Dourgnon et al. 2013). CMU and CMUC While the majority of residents in France are entitled to social security via their professional activity, some categories remained uncovered until 2000. Introduced that year, CMU established the right to statutory health insurance coverage on the basis of residence in France. It allowed those groups who were not covered by any insurance scheme to benefit from insurance coverage with means-tested subsidies. Only 3 percent of CMU beneficiaries pay social contributions; the rest is funded by the state using earmarked taxes (Rapport d’évaluation de la loi CMU 2011). CMUC was introduced to reduce the financial burden of cost sharing for the poorest part of the population generally not covered by VHI. The income threshold varies by household composition (about €8,500 a year for a single person, €13,000 a year for a couple). CMUC covers 100 percent of costs of all drugs and services included in the benefit package (no cost sharing). It also covers dental care and optics, which were poorly reimbursed by SHI. Moreover, patients are exempted from upfront payments, and professionals are not allowed to overbill CMUC patients. CMUC covered some 4.3 million people in 2012 (Rapport d’évaluation de la loi CMU 2012). AME AME was enacted in 1999 and implemented in 2000. It is fully state funded and provides access to a standard benefit package for illegal immigrants. It is means tested, and applicants must be resident for more than three months. Income eligibility is the same as for CMUC. As of 2010, 227,705 people benefited from the scheme. ACS ACS was introduced in 2004 to subsidize private complementary health insurance among low-income people ineligible for CMUC. The target population included those with incomes up to 15 percent above the CMUC eligibility line. The limit was pushed to 25 percent in 2011, then to 35 percent in 2013 (annual income below €11,600 for a single individual, €17,400 for a couple). ACS provides cash support in the form of vouchers that can be used to buy complementary insurance. Financial support accounts for €200 for an adult and between €350 and €500 for seniors, covering up to 60 to 70 percent of insurance premiums annually. The target population was about 2 million, but by end-2008 the number of enrollees had reached only half that (Rapport d’évaluation de la loi CMU 2011). The lack of information on the availability of the scheme, conditions of access, and complexity of identifying potential beneficiaries were observed to be the main barriers to use (Guthmuller et al. 2012). Effectiveness of Different Strategies The measures to improve financial access have contributed to reducing inequalities in access to care to a certain extent. But the juxtaposition of measures over time has also led to a very complex regulatory framework for patients and funders without entirely resolving problems of access. The ALD scheme has been effective in reducing inequalities in pharmaceutical, GP, and specialist expenditures among beneficiaries (Dourgnon et al. 2013). However, its contribution to reducing inequalities in optical and dental care expenditures remains marginal (Dourgnon et al. 2013). Despite the 22 exemption from copayments for their main health problem, ALD beneficiaries often suffer from an accumulation of additional “unlisted” disorders and are faced with higher out-of-pocket payments, on average double of those non-ALD beneficiaries (Dourgnon et al. 2013). CMU and CMUC have significantly improved access to GP care for the poorest (Grignon and Perronnin 2003; Guthmuller et Wittwer 2012; Dourgnon et al. 2012; Rapport d’évaluation de la loi CMU 2011), while use of GPs has become more equitable over time (figure 3.2). Inequalities in the use of specialist care also decreased slightly in the decade to 2008 (figure 3.3), due partly to the introduction of CMUC. Nevertheless, these two schemes are not panaceas. Denials and delays in obtaining medical appointments have been observed for their enrollees. Desprès et al. (2009) showed that 30 to 40 percent of sector 2 doctors, in particular dentists, ophthalmologists, and gynecologists, did not accept CMUC patients. There is also evidence of access problems for those with incomes just above the eligibility threshold (not covered by the CMUC but without private complementary insurance). Controlling for health needs, these near-poor groups appear to experience significantly lower levels of health care utilization than individuals with complementary coverage, whether private (VHI) or public (CMUC) (Dourgnon et al. 2013). While ACS may have played a role in closing the gap in complementary health insurance coverage, the cost-efficiency of this type of instrument is uncertain. Ultimately, ACS subsidizes private insurance contracts, but their contents are not regulated and may encourage overutilization of services that are cost inefficient. Figure 3.2 Income Inequalities in GP Utilization, 1998–2008 Source: Dourgnon et al. 2013. Figure 3.3 Income Inequalities in Specialist Utilization, 1998–2008 Source: Dourgnon et al. 2013. 23 3.3 EQUITY CONCERNS OF RECENT STRATEGIES The French health system relies heavily on VHI to cover part of health care costs. VHI coverage rose from 69 percent of the population in 1980 to 95 percent in 2012. However, enrollment remains voluntary, and access to VHI is still associated with income and employment status. About 12 percent of the poorest do not have VHI compared with only 3 percent among the better-off. Recent attempts to subsidize VHI increase further the reliance on VHI and may have some adverse effects on equity since the contents of VHI contracts or their quality are not regulated. Moreover, the launch of flat deductibles that restrict demand for care can potentially affect access to care for the worst-off. Service coverage and financial protection differ widely across VHI contracts, and affect income groups in different ways. In terms of service coverage, only a minority of the insured (9 to 12 percent) benefited from broad protection for dental care and optics in their current VHI contract (Couffinhal and Perronnin 2004). Service coverage is unequal across contracts. Large discrepancies exist between the three types of providers (mutuelles, provident firms, and private firms) and between collective and individual contracts. Collective contracts purchased by firms for their employees tend to be more generous on service coverage (Arnould and Vidal 2008). In 2008, about half (42 percent) of those with VHI benefited from a collective contract (Garnero and Rattier 2011). It is estimated that two-thirds of collective contracts provide broad protection, versus only one-third of individual contracts. Financial protection, too, varies widely across income groups. Good financial protection is generally associated with higher incomes: of the richest, 53 percent benefit from broad protection, compared with only 13 percent of the lower-income groups. As for coverage of extra- billing, VHI contracts can cover some (or all) costs linked to it (by doctors working in sector 2, whose prices are not regulated). Only 27 percent of individual contracts are reported to offer coverage for extra billing, while 64 percent of collective contracts do (Kambia-Chopin and Perronnin 2010). The net share left to patients depends on contract provisions. For those contracts offering coverage for extra billings, the overall rate of coverage is 120 percent of the SHI reimbursement (Garnero and Rattier 2011). In 2013, in an attempt to improve access to collective contracts, the government introduced new measures. From 2016, all employers (irrespective of the size of their business) must offer private VHI to their employees. While this may reduce inequalities between employees, the new policy can potentially perpetuate inequalities between salaried employees and other groups (students, retirees, unemployed, and self-employed). Increasing reliance on VHI for funding basic health services raises concerns for solidarity—a key principle of social protection—and redistribution in the system (Askenazy et al. 2013; Or 2010a). As a result of efforts to contain costs, the share of VHI in total health expenditure increased from 12.4 percent in 2000 to 13.7 percent in 2011 (figure 3.4). While services with higher risk of catastrophic spending (for example, hospital care) are better covered by national SHI (near 90 percent of total hospital expenditure), patients’ contributions, especially in areas where overbilling is common (surgery), can be quite high. Moreover, dental care and optical treatment, which can be quite costly for low-income budgets, are not well covered by SHI. Pursuing a strategy to provide VHI for the entire population without regulating items covered in the benefit basket may be problematic in ensuring equity of access to care and for cost efficiency. 24 Figure 3.4 Trends in Shares of VHI, SHI, and Out-of-Pocket Payments in Total Health Expenditure, 2000–11 2000 76.70% 12.40% 9.70% SHI 2005 76.80% 13% 9% VHI OOP 2011 75.50% 13.70% 9.60% Source: Adapted from Eco-santé data, 2011. Flat-rate deductibles are often used in health care to deter unnecessary utilization of services and lead patients to consider costs and benefits of treatments. To contain demand for targeted services, non- reimbursable flat-rate deductibles were introduced in recent years in France (table 3.2). Those deductibles are not supposed to be covered by VHI; in practice policies vary between different insurers. A complex capping system has been introduced in parallel with flat deductibles: per service, per day, and per year caps are used to limit financial consequences for consumers. Nevertheless, by their nature (insensitive to income), deductibles risk causing inequities in access to care. Moreover, they have limited effects on the behavior of higher-income groups. Table 3.2 List of Flat-Rate Deductibles by Type of Goods and Services Type of Goods and Services Flat Rate Each GP visit €1 Lab or radiography tests €1 per service Each hospital day €18 Each service above €120 €18 Each drug package €0.5 Each ancillary service €0.5 Each transportation €2 Source: Adapted from CNAMTS. A study by Kambia-Chopin and Perronnin (2010) estimated that flat-rate deductibles particularly disadvantage low-income groups, especially those without complementary coverage. The lowest-income quintiles reduced drug consumption far more than the better-off after the introduction of drug deductibles. 25 They report buying only a part of their prescription or delaying purchase. Relatively high deductibles for hospital care are also problematic for the most disadvantaged groups. Since the economic downturn that began in the late 2000s, there has also been a tendency to curb public spending by reducing SHI coverage for certain high-cost services. Measures include the reduction of reimbursement rates from 35 to 30 percent for drugs for common “non-vital” diseases, omitting high blood pressure from the list of diseases (ALD) for which costs are fully covered, and reducing coverage for transport expenditure for chronically ill patients. Although the measures were contained, they reflect a move to shift costs to the sick—rather than using general contributions—thereby endangering the principle of solidarity. SECTION 4: EFFICIENCY AND PAYMENT IN SERVICE DELIVERY Summary: The health sector is largely dominated by Fee-For-Service (FFS). To curb escalating expenditures in a UHC setting and improve performance, new payment mechanisms have been introduced, showing some promising results. Giving a more prominent role to primary care was another strategy to improve efficiency and better manage care pathways, but the FFS setting appears to be a barrier to changing clinician behavior. To improve overall efficiency and management within the hospital sector, the government launched an ambitious reform plan in 2003: measures modified the funding model for public and private hospitals, the rules of hospital sector planning, and the governance of public hospitals. 4.1 A DOMINANT FFS MODEL Historically, health care in France has been organized around four principles delineated by law: confidentiality of medical information; freedom of prescription and of practice for physicians; patients’ free choice of provider; and office-based FFS practice in the ambulatory sector. Thus doctors are free to choose their place of practice as well as the procedures and drugs to prescribe, and patients have free access to any physician or any facility—public or private—with no limit on the number of doctors seen or the frequency of visits. And until recently, there was no control of access to secondary or specialist care. As seen, doctors in the ambulatory sector and in private hospitals contract with health insurance funds and are paid according to a negotiated FFS schedule. A formal national negotiation process between the government, the insurance funds, and the medical profession sets official tariffs for reimbursement. Physicians and other health care providers, all paid by FFS, have neither the interest nor incentive to control the volume and the cost of their prescriptions or to invest in prevention, health promotion, and coordinated care. Fewer than 10 percent of consultations end without a prescription, and France ranks quite low among OECD countries for health promotion and prevention (OECD 2013). In addition, variations in medical practice among providers raise questions about efficiency and quality of care provision. In the past two decades, rising cost of health care has been a major concern. While France had visible success in controlling prices of health care services and pharmaceuticals through formal negotiations with health care providers and value-based pricing of drugs, low prices seem to have only a limited impact on health expenditure growth. Health care providers tend to compensate for these reduced revenues by increasing the volume of services. In the mid-2000s most parties and stakeholders recognized the need for further structural reforms to improve overall efficiency (Larcher 2007). The lack of articulation between ambulatory, hospital, and social care as well as between different levels of administration (central, regional, and departmental) has been recognized as a major drawback both in terms of cost control and quality of care. Moreover, uncoordinated care coupled with the high degree of independence and choice for both providers and patients have been identified as key drivers of health care cost. Increasingly, health care providers are asked to account for the cost and quality of services they provide. 26 4.2 INVESTING IN PRIMARY CARE Introducing Gatekeeping A noncompulsory gatekeeping system was introduced in 2004 as a means to better coordinate patient care and ultimately improve health sector efficiency. The system intended to prevent the unnecessary specialist consultations encouraged by freedom of choice and low rates of cost sharing (for the majority of the population who benefit from complementary health insurance). It also aimed at reinforcing the role of GPs as primary care providers to improve care coordination, quality, and efficiency. Under the new system, patients must declare a primary care provider or “preferred doctor” (médecin traitant), with whom they sign a contract, but whom they can change at any time and as often as they wish. Preferred doctors are expected to provide appropriate primary care and act as gatekeepers to guide their patients in the health system, coordinating treatment with other health care professionals. Patients are incentivized (with lower reimbursement rates) to get a referral for specialist care, but they remain free to choose any specialist without constraint. Direct access to gynecologists, ophthalmologists, psychiatrists, neuro-psychiatrists, and neurologists is permitted without penalty in certain circumstances (for instance, for contraception advice, screening, and prescription of eye glasses). Participation in the gatekeeping scheme is voluntary for patients and providers, but there are financial incentives to participate directed mainly at patients. If an individual does not register with a preferred doctor, the rate of reimbursement he or she is entitled to from the statutory health insurance fund is reduced from 70 percent to 30 percent. The same applies if he or she consults a specialist without referral. The payment mechanism (FFS) for physicians has remained the same. Preferred doctors did not receive 6 any extra payment for following up registered patients until 2013; however, for those suffering from a listed “chronic disease” such as cancer, diabetes, severe high-blood pressure, or long-term psychiatric disease, the physician receives an annual payment of €40 for drafting a care protocol. About 90 percent of the population has declared a preferred doctor. Nevertheless, despite generalists’ commitment to improve care coordination, promote prevention, and improve prescription habits by respecting guidelines, there has been little visible change in practice patterns. The impact thus far of this reform on overall functioning and efficiency has been marginal: an early evaluation showed that the new system did not achieve the expected financial impact, and that the rise in physicians’ fees largely offset the very modest reduction in visits to specialists. Eight years after the reform was launched, the National Auditing Office (Cour des Comptes) concluded that for the insured—the vast majority of whom had declared a preferred doctor—this reform remained essentially financial and somewhat incomprehensible (Cour des Comptes 2013). It criticized the little attention given to medical care coordination in the reform. Lack of coordination between GPs, specialists, and other professionals working in health care networks and hospitals thus remains a major issue for efficiency. Performance-Based Approach In 2009, as a new attempt to improve quality and efficiency of primary care, the SHI introduced a pay-for- performance scheme—Contracts for Improved Individual Practice, or CAPI (Contrats d'amélioration des pratiques individuelles). Offered to GPs and initially signed by them individually on a voluntary basis, the contracts sought to improve clinical quality of care by encouraging prevention and generic prescribing (box 4.1). The first contracts provided up to €7,000 annually if 100 percent of targets were achieved 6. Since July 2013, they have received €5 per registered patient. 27 Box 4.1 Initial Targets for CAPI Improving prevention. Target: to achieve a 75 percent flu vaccination rate for those over 65, and a breast cancer screening rate of at least 80 percent for women 50 to 74 years old. There was the further objective of lowering adverse reactions to drugs by reducing prescription of vasodilators (which are overprescribed despite being proven ineffective) and benzodiazepine (potentially dangerous and addictive) for elderly patients 65 and over. Increasing quality of care for patients with chronic diseases. Target: in particular, to improve the management of diabetes according to clinical guidelines (eye exams, HbA1c prescription) and management of high blood pressure (the target is to normalize blood pressure for 50 percent of patients over a three-year period. Rationalizing prescriptions. Target: specifically, to increase generic prescribing for seven groups of medications including antibiotics and antidepressants. Source: Or 2010b. The introduction of CAPI was highly controversial in the medical profession. Physician unions were concerned about the individual nature of the contracts and possibly a weak negotiating position in collective bargaining. The first contracts were signed in July 2009 by about 5,500 GPs; this number increased to 14,800 generalists (one-third of eligible GPs) by September 2010. Two-thirds of the GPs who signed the contract in July 2009 received remuneration in 2010. The analysis of results of initial participants by fund after one year of CAPI suggests that there was some progress in almost all targeted domains apart from prevention. The results for prevention and diabetes improved for all generalists, but the difference between CAPI signatories and others was not significant. Still, CAPI induced a change in French medical culture, demanding accountability from primary care providers for their results. The scheme was generalized to all GPs in 2011, with the stipulation that the payment of primary care providers could be related to their performance. Further, the objectives of CAPI were extended to include improved computer use in prescribing and in electronic data management. In 2012, with the “payment for public health objectives” scheme (Rémunération sur Objectifs de Santé Publique), all physicians were covered by the pay-for-performance contracts. Some objectives, such as organization of office practice and electronic records, are common to all physicians; others concern only GPs. In April 2013, CNAMTS reported that about 75,000 physicians (of whom two-thirds were generalists) had received some performance payment of about €3,700 on average (about €5,000 on average for GPs). Encouraging Multidisciplinary Group Practice Compared with traditional (solo) general practice, group practice is preferred in ensuring high-quality and efficient care provision. Finding an effective way of funding group practice that will emphasize prevention and care coordination in primary care has long been a policy objective. Despite several initiatives for encouraging group practice, take-up has been very slow; however, fewer than 50 percent of generalists work in group practice, and the size of practices and their distribution vary widely across regions. Different primary care structures have been created over recent years through various legal frameworks and payment schemes. Traditionally, health centers were group practices, mainly oriented to primary care, but they can include specialist services. More recently, multidisciplinary health houses have come to refer to group practice structures bringing together self-employed medical and paramedical professionals in a single practice. In health centers, professionals are salaried, but the main funding comes from FFS for their medical activities. In multidisciplinary group practices, doctors (and other health professionals) are paid by FFS. Thus collaborative work in groups—for example, for better disease management—is not particularly rewarded, and FFS is seen as a barrier for developing coordination among health professionals. To find new solutions, the 2007 Social Security Financing Act scheduled a period of five years starting in January 2008 to experiment with supplementary or alternative remuneration schemes in primary care structures (health centers and multidisciplinary health houses). The pilots, Expérimentation de Nouveaux Modes de Rémunération (Experimentation on New Modes of Payment, or ENMR), began in January 28 2010 in six regions. In all, 150 group practices were contracted by 19 regional health agencies (agences régionales de santé, or RHAs). The total cost of the pilot was about €7 million a year. Primary care structures are expected to improve management of chronic diseases and effectiveness of care provision by shifting the focus from curative care for acute conditions toward preventive services and care coordination. They are also intended to improve accessibility of care (with longer opening hours), cooperation between professionals (in particular between GPs and nurses), and the range of services for patients. Preliminary results of group practice suggest that the quality of care (prevention, coordination) in most domains is better, with slightly lower health care or pharmaceutical consumption in group practices than among solo-practice GPs. It also seems that in locations where pilots were installed, the density of GPs 7 has grown more rapidly than in the control areas. Group practice appears to be more attractive for generalists than solo practice in rural or underserved areas (Chevillard et al. 2013). 4.3 REFORMING THE HOSPITAL SECTOR TO IMPROVE EFFICIENCY Activity-Based Payment Until 2005 two different funding arrangements were used to pay public and private hospitals. Public and most private not-for-profit hospitals had global budgets, mainly based on historical costs, while private for- profit hospitals had an itemized billing system with different components: daily tariffs covered the cost of accommodation, nursing, and routine care; and separate payments were made for each diagnostic and therapeutic procedure, with separate bills for costly drugs and physicians’ fees. This difference between public and private hospitals had always been controversial. Public hospitals considered global budgets an instrument of rationing, which strangled the most dynamic hospitals and were inflexible to changing demand. Private hospitals argued that global budgets rewarded inefficiency and prevented fair benchmarking—that is, they believed that although they were more efficient, they were not remunerated accordingly. DRGs—a form of activity-based payment (ABP)—were launched in 2004–05 to pay for acute care services (including home hospitalization) with the objective of improving efficiency, creating a level playing field for payments to public and private hospitals, and improving the transparency of hospital activity and management (box 4.2). Improvements in hospital sector efficiency have been observed since the introduction of DRGs. The number of public hospitals in deficit decreased, and the size of the overall deficit fell from €485 million in 2007 to €220 million in 2010 (Cour des Comptes 2009). There also appeared to be a positive trend in productivity of public hospitals after 2004, with a strong rise in case mix–weighted production; whereas in the private sector, a modification of activity (case mix) was observed rather than a significant increase in productivity (Or et al. 2013; Studer 2012). Overall, DRG-based payment addressed some chronic problems inherent in the hospital market and improved transparency and accountability of health care facilities and enhanced efficiency. DRGs allowed for gains in overall transparency of information on hospital activity. Moreover, by linking payments to actual activity, DRG-payment creates direct incentives for hospitals to increase productivity. 7. Over 2004–11, comparing trends before and after 2008 (start of the pilots) in different areas. 29 Box 4.2 DRGs à la Française Implementation. In the public sector (public and private not-for-profit hospitals), the share of activities paid for by ABP increased gradually, starting with 10 percent in 2004 and 25 percent in 2005 to reach 100 percent in 2008. Private for-profit hospitals, however, have been paid entirely by ABP since March 2005. A transition period (until 2012) was allowed where "national prices" were adjusted, with each provider taking into account its own historical costs and prices. Patient classification. Under ABP, the income of each hospital is linked directly to the number and case mix of patients treated, defined in terms of homogeneous patient groups (groupes homogènes des malades). The classification was inspired initially by the US Health Care Financing Group classification (HCFA-DRG) but adapted to the French system and modified regularly. The current version (v11) of the classification, introduced in 2009, accounts for 2,291 groups, distinguishing four levels of case severity for most DRGs, compared with 784 in the previous version. Price setting. The DRG prices (tariffs) for each service are set annually at the national level based on average costs. However, there are two different sets of tariffs: one for public (including private, not-for- profit) hospitals and one for private, for-profit hospitals. The tariffs for public hospitals cover all costs linked to a stay, while those for the private sector do not cover medical fees paid to doctors (which are paid FFS) or the costs of biological and imaging tests (for example, scanning), which are billed separately. Average costs per DRG (reference costs) are calculated from a voluntary set of hospitals. Reference costs are transformed into actual DRG prices by the Ministry of Health in a complex and opaque calculation, taking into account total hospital expenditure growth and public health priorities. Extra payments. Public hospitals (and private hospitals participating in so-called “public missions”) receive additional payments—Missions d'Intérêt Général et à l'Aide à la Contractualisation (General Interest Missions and Assistance Contracts)—to compensate for specific missions, including education, research, and innovation-related activities; activities of general public interest such as those meeting national or regional priorities (for example, developing preventive care); and investments in quality, contracted with the RHAs. The costs of maintaining emergency care and related activities are paid by fixed yearly grants, plus an FFS element accounting for the annual activity of providers. Finally, a restricted list of expensive drugs and medical devices is paid retrospectively, according to the actual level of prescriptions. Expenditure control. To contain hospital expenditure, national expenditure targets for acute care (with separate targets for the public and private sectors) are set by Parliament. If the actual growth in volume exceeds the target, prices subsequently decrease. However, this macro-level regulatory mechanism creates confusion and an extremely opaque environment for hospitals in which it is impossible to predict market trends and prices. Prices for homogeneous patient groups are set as a function of overall changes in hospital activity, increasingly independent of costs and their evolution at individual hospital level. Source: Authors. However, the quality of hospital care may be at risk: 30-day readmission rates for the main cardiovascular diseases and for cancer are reported higher since the introduction of DRG payment. Moreover, the sharp increase in standardized rates of certain interventions and procedures is suggestive of supplier-induced demand that is little justified. Increasingly, the quality and relevance of care provided under the DRG system are questioned, as ABP provides incentives to develop hospital activity, sometimes beyond what is medically necessary. Assuring pertinence of care is now a policy priority, with several institutions tackling the issue. Finally, it is largely recognized that ABP does not favor cooperation between hospitals or within a hospital between different departments to assure care coordination and quality. Future objectives include adjusting hospital payments based on a range of quality indicators, extending payments beyond acute hospital reimbursement (especially for chronically ill and multi-morbidity patients), and bundling payments for rehabilitative services. But these may be more difficult to achieve than initially thought due to lack of robust cost data across providers. 30 New Corporate Governance for Hospitals Influenced by lessons from other sectors, many health care policy makers concluded that the performance problems of public hospitals were grounded in the rigidity of hierarchical bureaucracy, the lack of control by managers, and absence of performance-based incentives. Having applied new public management techniques in other sectors, it was a natural step to consider applying similar reforms to the health sector (Preker and Harding 2003). Initially, the reform choice was to give hospitals some degree of autonomy. Limited success with this type of reform encouraged European policy makers to transform state-owned facilities into private firms or networks (such as Hospital Trusts in the United Kingdom). Along these lines, a middle course between the status quo and privatization, which consisted of retaining public ownership of the hospital but initiating organizational reforms in the form of “corporatization,” was introduced in France, as in neighboring countries, in the 2000s. Introducing new payment mechanisms alone would not be sufficient for hospitals to rationalize how they were organized: combining internal governance and funding reform was necessary to give public hospitals the requisite flexibility to respond to the incentives they were offered. Reforms in internal governance, included, first, reinforcing the role of the hospital chief executive officer (CEO), closely associating medical staff in strategic decision making; and second, giving medical management more autonomy within their remit. The powers of the hospital CEO have been greatly extended, and he or she alone is accountable to the regulatory entities with the power to make spending and recruiting decisions. This expansion was associated with a dual management model, engaging the medical community in strategic decision making. In practice, the Medical Board (Commission médicale d’établissement) now has full powers for care safety, quality, and patient outcomes; its main role is to draw up the medical project (projet médical) of the hospital, for which it has full decision-making and implementation power. Additionally, the CEO must consult with the Medical Board in any decision that affects the hospital’s organization from the standpoint of medical care. Moreover, the chairperson of the Medical Board is also vice chair of the board of directors (the directoire) that assists the CEO and must be consulted on all key aspects of hospital management (financial accounts, investment choices, recruitment and staffing policy, and contracting with regulatory entities). Another structurally significant aspect of reforms for internal governance was the introduction of new units called medical “departments” (pôles). Their aim is to give senior medical personnel more managerial autonomy while making them more accountable for resources. Indeed, the everyday actions and strategic choices they make must comply with the activity and managerial performance specifications contained in the contract they sign (the department contract—contrat de pôle) with the hospital management. This internal contract reflects the objectives and indicators set out in the external contracting mechanisms (see below on external governance). The creation of such a department headed by medical staff with increased managerial autonomy has been a crucial step toward promoting a performance-oriented culture in hospitals. Another underlying objective is to encourage cooperation between medical wards and resource sharing (administrative and paramedical personnel, equipment, and R&D), thus helping to rationalize organization at hospital level. Changes required by law in hospitals’ internal organization were implemented quite rapidly. A few years after 2005, medical departments were in place and joint CEO/Medical Board governance structures had been set up. The principle of empowering the medical community is now widely accepted and is promoted in hospitals by a medico-economic approach to health care. A cornerstone of internal governance reforms has been the creation of medical departments that encompass several medical units or specialties and are financially accountable to the CEO. In principle, the rationale for defining boundaries should be based on medical logic or common projects such as R&D or patient pathways. In reality, however, decisions are sometimes based on preexisting logistical or organizational constraints such as common accountancy software or use of the same building. The most successful experiences occurred where departments’ boundaries matched hospitals’ strategic objectives and where the medical units they encompassed were tied to common goals. In less successful cases, heads of departments have had to give medical meaning ex post to artificially designed organizations (IGAS 2010). 31 SECTION 5: REFORMING HEALTH SYSTEM GOVERNANCE Summary: Traditionally, France’s health care system has had strong central governance and a hospital-centered model; the regions had limited power. New sector governance was introduced at the end of the 2000s, shifting power to recently created Regional Health Agencies (RHAs), as “one-stop shops.” In addition, a set of instruments was introduced to move from a supply-oriented approach and to better serve populations according to their actual needs and demand. 5.1 THE CREATION OF RHAS Metropolitan France’s 22 regions and 96 departments have limited degrees of administrative autonomy. Several initiatives for decentralizing executive powers have led to fragmentation of responsibility for health services among administrative levels (Larcher 2007). In 2009, a major reform package—the Law “Hôpital, patients, santé et territoires” (Hospital, Patients, Health and Territories)—transformed health governance, simplifying regional health management while strengthening coherence of health care policies at regional level. Until 2010, most hospital activities were controlled by regional hospital agencies (agences régionales d'hospitalisation), but several state agencies were responsible for the provision and financing of different types of long-term and social care services. The core measure in the reform was the creation of RHAs, which, along the lines of one-stop shops, brought together seven different public agencies. The agencies have the mission to define health care needs at regional level and guarantee fair access to care, and to improve coordination between hospital, ambulatory, and social care providers. They oversee public and private hospitals, as well as nursing homes. The RHAs were set up in 2010. Their directors are nominated by the government’s Council of Ministers (cabinet). They are to invest in health promotion and prevention as well as to regulate care supply to meet the needs of the local population while assuring quality and efficiency. They are increasingly important actors in shaping health care organization. Another significant structural change is the creation of new legal entities called local hospital communities (communautés hospitalières de territoire), in view of quality problems in low-volume, small hospitals for complex surgery (Or 2007). They are formed by grouping large and small hospitals based on the complementarity of their medical competencies. The idea is to concentrate complex surgical interventions in high-volume hospitals and transfer less complex medical and medico-social care to small, local hospitals. In principle, hospitals in such a community will be able to share their patients as well as their health care resources. 5.2 GREATER LOCAL REGULATION OF HEALTH CARE SUPPLY In the last few decades in all high-income countries, technological and organizational progress has changed the role that modern hospitals play within health care economies and territories (often, regions), in particular with the advent of lighter ambulatory care techniques. Unlike other countries in Europe such as Sweden or the United Kingdom, which have a strong primary care infrastructure, France has traditionally been a hospital-centered system with weak ties to other health care providers. On the eve of reforms, there was agreement that the last several decades of supply-side planning pursued centrally by the Ministry of Health had led to an expensive, over-hospitalized health care system to the detriment of other health care providers. Reforms therefore aimed at readjusting hospitals’ functions within a territory in relation to other providers, and at rationalizing overall health care supply to address local needs, while avoiding duplication and waste. Ultimately, the policy objective was (and is) to promote the emergence of hybrid forms of health care organization that cut across hospital boundaries and enable better links between hospitals as well as with ambulatory, primary, and social care. Previous forms of strict quantitative planning with quotas were abandoned in 2003 and replaced by more flexible approaches based on fulfilling local health needs rather than defining strict targets for supply (that is, a demand-oriented rather than a supply-side approach). A regional organization plan (schéma régional 32 d'organisation sanitaire), drawn up by each RHA, defines each region’s health care supply needs simultaneously for hospital, ambulatory, and social care on the basis of regional demographic and epidemiological characteristics. The shift toward more decentralized and flexible forms of governance is also apparent in the increased use of contracting mechanisms between individual hospitals and the RHAs. Indeed, RHAs must sign a five-year contract with each hospital (contrat pluriannuel d’objectifs et de moyens), defining service, quality objectives, and resources required, so that hospitals have to justify their resource use. Despite this shift, RHAs’ capacity to manage financial resources remains weak. They have hardly enough funding on which to base rational supply-side restructuring. The Ministry of Health remains the last-resort decision maker on most sensitive issues, such as hospital restructuring. In 2013, RHAs had €3 billion to finance some hospital and social services—a very small proportion of public health care expenditure in the hospital sector (less than 4 percent). Moreover, RHAs do not have the data on which to make rational decisions. Indeed, most strategic data on patients’ characteristics and health care consumption are contained in the claims data that are managed by health insurance funds. In theory, this information should be shared with RHAs and analyzed by local teams to identify care-consumption patterns and shortcomings in care supply. But in reality, access to health data is a major problem (box 5.1), and the lack of an appropriate (open) structure presenting health data in a meaningful way as well as the scarcity of expertise in health information statistics among RHA staff reduce the agencies’ capacity for regional guidance. Another issue for improving regional efficiency was that various types of suppliers were regulated nationally by different bodies, although locally they were recently under the umbrella of RHAs. Indeed at national level, all ambulatory care providers were mainly piloted by the Social Security administration with collective contracting mechanisms, while hospitals were within the ministry’s remit; and home and social care are in the hands of locally elected authorities. On the ground, this institutional setting has not yet been conducive to cooperation across boundaries of care, nor to the emergence of new forms of hybrid care supply with multidisciplinary teams responsible for health care pathways of patients. Box 5.1 The Importance of Health Information Systems for Regulation Almost all residents in France have an electronic data card (the Carte Vitale) that contains their identification and confirms their right to statutory health insurance benefits. This card is used by almost all health care providers (hospitals, home-based GPs and specialists, pharmacists, diagnostic laboratories, etc.) to electronically bill the health insurance funds. Thanks to this electronic third party payment system, patients do not have to directly pay health care professionals, as was previously the case. However, the Carte Vitale does not have any medical information and does not allow one to follow patients’ medical history—it is only used for billing. Reimbursement data on ambulatory care and pharmaceutical products—payment of which is based on an itemized FFS system—are registered in a data repository managed by health insurance funds called the SHI inter-scheme system (Système National d’Information Inter-régime de l’Assurance Maladie). It contains about 1.2 billion yearly reimbursement records providing valuable information on prescriptions, physicians, and patients, and can be linked to data from other sources such as hospitals and private insurers. The data are collected primarily for reimbursement, follow-up, and management by SHI. Use of these data is strictly regulated with few public agencies and institutions allowed to exploit certain data. Private for-profit organizations are forbidden access, and use of insurance data for commercial purposes is prohibited. However, processing these data for general use is not part of the funds’ mission, and hence little information is available to the public for comparing the quality and cost of health care services. Currently, there is intense demand both from the industry and the research community for further sharing the reimbursement data (the “open data initiative”). 33 On hospital care, where activity is measured by DRGs, the major information system is an activity database (the Programme de Médicalisation des Systèmes d’Information) managed by a specialized technical agency of the Ministry of Health (the Agence Technique de l’Information Hospitalière). This agency is also responsible for setting the DRG tariffs, based on the cost structure of a sample of hospitals. Source: Authors. 34 CONCLUSION France’s experience presents several potentially useful lessons for other countries aiming for UHC. The mix of mandatory and voluntary insurance to cover copayments of basic health services seems to have been effective in covering the entire population for a comprehensive set of goods and services and against costs of illness. However, the introduction of a state-funded insurance scheme for the poorest was necessary to ensure that individuals with variable incomes actually benefit from the same health coverage. Moreover, a series of complementary schemes and measures were necessary to deal with negative effects of cost sharing for basic health services for low socioeconomic groups and improve equitable access to care over time. Still, inequalities in health and access to specialist services persist. Thus France’s experience with private complementary insurance suggests that reliance on voluntary insurance for financing basic health services is problematic for equity and redistribution. A wider mandatory system covering all necessary services would probably be a more equitable approach. The initial employment-based funding system proved somehow unsustainable in a constrained economic environment with high unemployment and an aging population. Funding sources have been broadened in the past 10 years, generating additional resources through different taxes applied to a broader range of incomes, including that from financial assets. On the expenditure side, a series of measures, including national targets, to curb escalating trends have helped monitor health expenditure more closely and demonstrated some promising results in curbing health spending in very recent years. Like many other countries, efficiency and quality of care have been a continuous concern for France’s health system. Investing in primary care by strengthening the role of GPs in delivery and coordination of care has been a key strategy in recent years. The experience with pay-for-performance contracts for GPs (CAPI) has shown advantages over traditional FFS for improving the efficiency and accountability of providers. However, the French experience with gatekeeping has also suggested that payment reforms built on FFS are, on their own, unable to radically change provider behavior. France has also been exploring alternative payment models incentivizing collaborative work in multidisciplinary group practices that emphasize prevention and care coordination. At hospital level, a tailored DRG system introduced in the mid-2000s has reached half its goals by boosting productivity, but challenges remain for ensuring quality and pertinence of care and improving performance overall. France’s experience also shows that to deal with adverse effects of a DRG payment system, a strong information system monitoring costs and quality of hospital services is essential, as is flexible and transparent governance supporting continual fine-tuning of the incentive structure. To improve equity of access to care, as well as quality and efficiency, the authorities also had to modify the traditional centralized and fragmented governance model. Health governance has been thoroughly reformed over the past decade, through the creation of regional entities overseeing all health care providers (hospital, ambulatory, and social care) to meet the needs of the local population, and by remodeling hospital corporate governance, leading to a new deal between hospital administration and the medical community. The conflicting pressures to curb rising health expenditures while ensuring equity of access and quality of care characterize France’s experience with UHC—highlighting the crucial role of complex, perpetual reforms to sustain UHC. 35 REFERENCES Allonier C., P. Dourgnon, and T. Rochereau. 2008. “Enquête sur la santé et protection sociale 2008.” IRDES, Paris. Arnould M. L, and G. Vidal. 2008. “Typologies des contrats les plus souscrits des complémentaires en 2006.” Etudes et Résultats, DREES. Askenazy P., B. Dormont, P-Y. Geoffard, and V. Paris. 2013. “Pour un système de santé plus efficace,” Les notes du conseil d’analyse économique, No. 8, July. Assurance maladie. 2010. “Point d’information.” 16 (September). http://www.ameli.fr/fileadmin/user_upload/documents/Dp_capi_16_09_2010_vdef.pdf. Bellanger, M., and P. R. Mosse. 2005. “The Search for the Holy Grail: Combining Decentralized Planning and Contracting Mechanisms in the French Health Care System.” Health Economics 14: S119–S132. Boisguerin, B., and B. Haury. 2008. “Les bénéficiaires de l’AME en contact avec le système de soins.” Etudes et Résultats 645 (July): 1–8, DREES. Bourgueil, Y., and Y. Berland. 2006. “Health Care Human Resource Policy in France.” Eurohealth 12 (3): 24–26. Bréchat, Pierre-Henri, Antoine Leenhardt, Marie-Christine Mathieu-Grenouilleau, et al. 2010. “Des pôles d’activités hospitaliers entre gestion de la crise et mise en œuvre d’une politique de santé publique.” Santé Publique 22 (5): 571–80. Cambois, E., and F. Jusot. 2007. “Ampleur, tendance et causes des inégalités sociales de santé et de mortalité en Europe: une revue des études comparatives.” Bulletin Epidémiologique Hebdomadaire, Thématique—Les inégalités sociales de santé en France en 2006: éléments de l'état des lieux 2–3 (January): 10–14. Cases, C. 2009. “Evaluation of the Cancer Plan.” Health Policy Monitor. http://hpm.org/survey/fr/a13/1. Chevillard, G., J. Mousquès, V. Lucas-Gabrielli, and Y. Bourgueil. 2013. “Répartition géographique des maisons et pôles de santé en France et impact sur la densité des médecins généralistes libéraux. Deuxième volet de l’évaluation des maisons, pôles et centres de santé dans le cadre des expérimentations des nouveaux modes de rémunération (ENMR).” Question d’Economie de la Santé, IRDES, No. 193. Chevreul, K., I. Durand-Zaleski, S. B. Bahrami, C. Hernandez-Quevedo, and P. Mladovsky. 2010. “France: Health System in Review.” The Health Systems in Transition (HIT) Profiles 12 (6). European Observatory on Health Systems and Policies, Copenhagen. CNAMTS (Caisse nationale de l'assurance maladie des travailleurs salariés). 2011. Les Comptes de l’Assurance Maladie, Chiffres et Repères. http://www.ameli.fr/l-assurance-maladie/statistiques-et- publications/index.php. Com-Ruelle, L., P. Dourgnon, and V. Paris. 2006. “Can Physician Gate-Keeping and Patient Choice Be Reconciled in France? Analysis of Recent Reform.” Eurohealth 12 (1): 17–19. Couffinhal, A., and M. Perronnin. 2004. “Accès à la couverture complémentaire maladie en France: Une comparaison des niveaux de remboursements.” CREDES. Cour des Comptes. 2009. “La Sécurité Sociale.” In La mise en place de la T2A: Bilan à mi-parcours. Paris. ———. 2013. “Le médecin traitant et le parcours de soins coordonnés: une réforme inaboutie”, Rapport public annuel, Chapter 3, February. Paris. Culyer, A. J., and A. Wagstaff. 1993. “Equity and Equality in Health and Health Care.” Journal of Health Economics 12: 431–57. Daudigny, Y., J. Le Menn, and A. Milon. 2012. “Refonder la tarification hospitalière au service du patient.” Rapport d’information au Sénat, No. 703. Mission d’Evaluation et de Contrôle de la Sécurité Sociale, Paris. Debrosse, D. 2010. “La Nouvelle gouvernance hospitalière.” Actualité et Dossier en Santé Publique (ADSP) 72 (September). 36 Desprès, C., S. Guillaume, and P. Couralet. 2009. “Le refus de soins à l’égard des bénéficiaires de la Couverture maladie universelle complémentaire à Paris: une étude par testing auprès d’un échantillon représentatif de médecins (omnipraticiens, gynécologues, ophtalmologues, radiologues) et de dentistes parisiens.” Rapport IRDES pour le Fonds CMU, July. Devaux, M., and M. de Looper. 2012. “Income-related Inequalities in Health Service Utilization in 19 OECD Countries 2008-2009.” OECD Health Working Paper, Paris. Dourgnon, P., Z. Or, and C. Sorasith. 2012. “Les inégalités de recours aux soins en France, retour sur une décennie de réformes.” Actualité et Dossier en Santé Publique (ADSP) 80: 33–35. ———. 2013. “L’impact du dispositif des affections de longue durée (ALD) sur les inégalités de recours aux soins ambulatoires entre 1998 et 2008.” Questions d'économie de la santé 183 (January). IRDES, Paris. Garnero, M., and M-O. Rattier. 2011. “Les contrats les plus souscrits auprès des complémentaires santé en 2009.” Etudes et Résultats, DREES. Grignon M., and M. Perronnin. 2003. “Impact de la CMUC sur la consommation de soins.” Questions d’économie de la santé. IRDES, Paris. Guillaume, S., and T. Rochereau. 2010. “La protection sociale complémentaire collective: des situations diverses selon les entreprises.” Questions d’économie de la santé 155. IRDES, Paris Guthmuller, S., F. Jusot, J. Wittwer, and C. Desprès. 2012. “Faire valoir ses droits à l’Aide complémentaire santé: les résultats d’une expérimentation sociale.” Economie et Statistique 455–56. Guthmuller, S., and J. Wittwer. 2012. “L’effet de la Couverture Maladie Universelle - Complémentaire (CMU-C) sur le nombre de visites chez le médecin: Une analyse par régression sur discontinuities,” Revue d'Economie Publique 28–29. Hanratty, B., T. Zhang, and M. Whitehead. 2007. “How Close Have Universal Health Systems Come to Achieving Equity in Use of Curative Services? A Systematic Review.” International Journal of Health Services 37 (1): 89–109. IGAS (Inspection Générale des Affaires Sociales). 2010. Bilan de l’organisation en pôles d’activité et des délégations de gestion mises en place dans les établissements de santé. IGF (Inspection Générale des Finances). 2012. Evaluation de la tarification des soins hospitaliers. Report 2011-M-056-01. http://www.igf.finances.gouv.fr/webdav/site/igf/shared/Nos_Rapports/documents/2012/2011-M-056- 01.pdf. IRDES (Institut de Recherche et Documentation en Economie de la Santé). 2000. “La complémentaire maladie en France: qui bénéficie de quels remboursements?” Bulletin d’information en économie de la santé 32 (October). Kambia-Chopin, B., and M. Perronnin. 2010. “Les franchises ont-elles modifié les comportements d’achats de médicaments?” Questions d’économie de la santé 158 (October). IRDES, Paris. Kunst, AE, F. Groenhof, and J.P. Mackenbach. 1998. “Mortality by Occupational Class among 30-64 years in 11 European Countries.” Soc Sci Med 46 (11): 1459-76. Larcher, G. 2007. “Role and Place of the Hospital: An Inventory Prepared by the Consultative Commission on Hospital Missions.” 14/12/2007. Paris: Ministry of Health, Youth, Sport and Community Organisations (in French). Lucas-Gabrielli, V., and M. Coldefy. 2007. “Regional Healthcare Planning.” Health Policy Monitor. http://hpm.org/survey/fr/a10/4. Mackenbach, J.P. et al. 2008. “Socio-economic Inequalities in Health in 22 European Countries.” N Engl H Med June 5, 538 (23): 2468-81. Ministère de la Santé. 2011. “Evaluation de la Loi CMU.” Report No. 5, November. MECSS (Mission d’Evaluation et de Contrôle de la Sécurité Sociale). 2012. “Refonder la tarification hospitalière au service du patient.” Rapport d’information au Sénat, No. 703. Mousquès, J., and M. Naidich. 2008. "Regulating Nurses—Follow Up." Health Policy Monitor (October). http://www.hpm.org/survey/fr/a12/1. Mousquès, J., and D. Polton. 2004. “Sickness Funds Reform: New Governance.” Health Policy Monitor. http://www.hpm.org/en/Surveys/IRDES- France/04/. 37 OECD (Organisation for Economic Co-operation and Development). 2013. Health at a Glance 2013: OECD Indicators. OECD, Paris. Or, Z. 2005. “Hospital Payment Reform.” Health Policy Monitor 5. http://www.irdes.fr/EspaceAnglais/International/Bertelsmann2005.html. ———. 2007. “Hospital 2007.” Health Policy Monitor (October). http://hpm.org/survey/fr/a9/1. ———. 2008. “Changing Regional Health Governance in France.” Eurohealth 14 (4): 7–8. ———. 2010a. “Measures to Curbing Health Expenditure.” Health Policy Monitor (October). ———. 2010b. "P4P for Generalists: First results." Health Policy Monitor (October). http://www.hpm.org/survey/fr/a16/3. Or, Z., and G. De Pouvourville. 2006. “French Hospital Reforms: A New Era of Public-Private Competition?” Eurohealth 12 (3): 21–23. Or, Z., C. Cases, M. Lisac, K. Vrangbaek, U. Winblad, and G. Bevan. 2009. “Are Health Problems Systemic? Politics of Access and Choice under Beveridge and Bismarck Systems.” IRDES Working Paper, Paris. Or, Z., and N. Sirven. 2011. “The French Health Care System.” In Health Care Systems: A Global Survey, ed. H. S. Rout, 152–78. New Delhi: New Century Publications. Or, Z., J. Bonastre, F. Journeau, and C. Nestrigue. 2013. “Activité, productivité et qualité des soins des hôpitaux avant et après la T2A.” Questions d’économie de la santé 186 (April). IRDES, Paris. Pichetti, S., and C. Sermet. 2011. “Le deremboursement des médicaments en France entre 2002 et 2011: éléments d’évaluation.” Questions d’économie de la santé 167 (July). IRDES, Paris. Preker A., and A. Harding. 2003. Innovations in Health Service Delivery. Washington, DC: World Bank. “Rapport d’information au Sénat.” 2013. No. 335. Mission d’Evaluation et de Contrôle de la Sécurité Sociale, Paris. Rodwin, V. et al. “Universal Health Insurance in France: How Sustainable?” Essays on the French Health Care System, Office of Health and Social Affairs, Embassy of France in Washington, DC. http://wagner.nyu.edu/health/universal.pdf. Sandier, S., V. Paris, and D. Polton. 2004. Health Care Systems in Transition (HIT): France, Copenhagen, WHO Regional Office for Europe on behalf of the European Observatory on Health Systems and Policies. Studer, N. 2012. “Quelles évolutions récentes de la productivité hospitalière dans le secteur public?” Document de travail Série Études et Recherche 114, DREES. Wagstaff, A., E. van Doorslaer, H. van der Berg, S. Calonge, T. Christiansen, and G. Citoni. 1999. “Equity in the Finance of Health Care: Some Further International Comparisons,” Journal of Health Economics 18: 263–90. 38 While Universal Health Coverage (UHC) offers a powerful goal for a nation, all countries—irrespective of income—are struggling with achieving or sustaining UHC. France is a high-income country where health coverage is in effect universal. Health-related costs are covered by a mix of mandatory Social Health Insurance (SHI) and private complementary schemes, while benefit packages are comprehensive, uniform and of good quality. France provides some of the highest financial protection among countries in the OECD. Still, under pressure to sustain UHC without compromising equity of access, the system has been fine-tuned continually since inception. Much can be learned from France’s experience in its reforms toward better fiscal sustainability, equity and efficiency. This study assesses the major challenges that France has faced for sustaining UHC, and highlights its experiences and lessons in addressing system bottlenecks to benefit less developed countries as they embark on the path to UHC. ABOUT THIS SERIES: This series is produced by the Health, Nutrition, and Population Global Practice of the World Bank. The papers in this series aim to provide a vehicle for publishing preliminary results on HNP topics to encourage discussion and debate. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations or to members of its Board of Executive Directors or the countries they represent. Citation and the use of material presented in this series should take into account this provisional character. For free copies of papers in this series please contact the individual author/s whose name appears on the paper. Enquiries about the series and submissions should be made directly to the Editor Martin Lutalo (mlutalo@ worldbank.org) or HNP Advisory Service (healthpop@worldbank.org, tel 202 473-2256). For more information, see also www.worldbank.org/hnppublications. 1818 H Street, NW Washington, DC USA 20433 Telephone: 202 473 1000 Facsimile: 202 477 6391 Internet: www.worldbank.org E-mail: feedback@worldbank.org