 INTERNATIONAL PERSPECTIVES FORUM 

The System Under French National : Lessons for Health Reform in the United States

The French | Victor G. Rodwin, PhD, MPH combines universal coverage with a public–private mix of hospital and ambulatory care THE FRENCH HEALTH CARE out reforming the overall man- levels of resources (Table 2), and and a higher volume of service system has achieved sudden no- agement and organization of the a higher volume of service provi- provision than in the United toriety since it was ranked No. 1 health system. This strategy has sion (Table 3) than in the United States. Although the system 32 is far from perfect, its indica- by the World Health Organiza- exacerbated tensions among the States. There is wide access to 1 tors of health status and con- tion in 2000. Although the state, the NHI system, and comprehensive health services sumer satisfaction are high; methodology used by this assess- health care professionals (princi- for a population that is, on aver- its expenditures, as a share of ment has been criticized in the pally physicians), tensions that age, older than that of the United gross domestic product, are Journal and elsewhere,2–5 indi- have long characterized the po- States, and yet ’s health far lower than in the United cators of overall satisfaction and litical evolution of French expenditures in 2000 were States; and patients have an health status support the view NHI.13–15 equal to 9.5% of its gross domes- extraordinary degree of choice that France’s health care system, Although the French ideal is tic product (GDP) compared with among providers. while not the best according to now subject to more critical scru- 13.0% of GDP in the United Lessons for the United these criteria, is impressive and tiny by politicians, the system States.17 States include the importance deserves attention by anyone in- functions well and remains an The health system in France is of government’s role in pro- viding a statutory framework terested in rekindling health important model for the United dominated by solo-based, fee-for- for universal health insurance; care reform in the United States States. After more than a half service private practice for ambu- recognition that piecemeal re- (Table 1). French politicians century of struggle, in January latory care and public hospitals form can broaden a partial pro- have defended their health sys- 2000, France covered the re- for acute institutional care, gram (like Medicare) to cover, tem as an ideal synthesis of soli- maining 1% of its population among which patients are free to eventually, the entire popula- darity and liberalism (a term un- that was uninsured and offered navigate and be reimbursed tion; and understanding that derstood in much of Europe to supplementary coverage to 8% under NHI. All residents are au- universal coverage can be mean market-based economic of its population below an in- tomatically enrolled with an in- achieved without excluding pri- 16 systems), lying between Britain’s come ceiling. This extension of surance fund based on their oc- vate insurers from the sup- “nationalized” health service, health insurance makes France cupational status. In addition, plementary insurance market. where there is too much ra- an interesting case of how to en- 90% of the population sub- (Am J . 2003;93: 31–37) tioning, and the United States’ sure universal coverage through scribes to supplementary health “competitive” system, where too incremental reform while main- insurance to cover other benefits many people have no health in- taining a sustainable system that not covered under NHI.33 An- surance. This view, however, is limits perceptions of health care other distinguishing feature of tempered by more sober ana- rationing and restrictions on pa- the French health system is its lysts who argue that excessive tient choice. Following an over- proprietary hospital sector, the centralization of decisionmaking view of the system, and an as- largest in Europe, which is acces- and chronic deficits incurred by sessment of its achievements, sible to all insured patients. Fi- French national health insurance problems, and reform, this article nally, there are no gatekeepers (NHI) require significant explores lessons for the United regulating access to specialists reform.9,10 States of the French experience and hospitals. Over the past 3 decades, suc- with NHI. French NHI evolved from a cessive governments have tin- 19th-century tradition of mutual kered with health care reform; THE FRENCH HEALTH aid societies to a post–World the most comprehensive plan CARE SYSTEM War II system of local demo- was Prime Minister Juppé’s in cratic management by “social 1996.11,12 Since then, whether The French health care system partners”—trade unions and governments were on the politi- combines universal coverage employer representatives—but it cal left or right, they have pur- with a public–private mix of hos- is increasingly controlled by the sued cost control policies with- pital and ambulatory care, higher French state.34 Although NHI

January 2003, Vol 93, No. 1 | American Journal of Public Health Rodwin | Peer Reviewed | International Perspectives Forum | 31  INTERNATIONAL PERSPECTIVES FORUM 

TABLE 1—Health Status and Consumer Satisfaction Measures: France, United States, Germany, United Kingdom, Japan, and Italy

France US Germany UK Japan Italy

Health status Infant mortality (deaths/1000 live births), 1999a 4.3 7.2b 4.6 5.8 3.4 5.1 LEB (female), 1998a 82.2 79.4 80.5 79.7 84.0 81.6c LEB (male), 1998a 74.6 73.9 74.5 74.8 77.2 75.3c LE at 65 (female), 1997a 20.8 19.2 18.9 18.5 21.8 20.2 LE at 65 (male), 1997a 16.3 15.9 15.2 15.0 17.0 15.8 Severe disability-free life expectancy (female), 1990/1991d 14.8 NA NA 13.6 14.9 NA Severe disability-free life expectancy (male), 1990/1991d 18.1 NA NA 16.9 17.3 NA Potential years of life lost per 100000 population (female), 1993e 2262 3222 2713 2642 1914 2136 Potential years of life lost per 100000 population (male), 1993e 5832 6522 5752 4688 4003 4873 Consumer satisfaction, % Only minor changes needed, 1990f 41 10 41 27 29 12 Very satisfied, 1996g 10 NA 12.8 7.6 NA 0.08 Fairly satisfied, 1996g 55.1 NA 53.2 40.5 NA 15.5

Note. US=United States; UK=United Kingdom; LEB=life expectancy at birth; LE=life expectancy; NA=not available. aSource. Organization for Economic Cooperation and Development.6(p27) b1998. c1997. dDefined as life expectancy with the ability “to perform those activities essential for everyday life without significant help.”6(p27,31) eSource. Organization for Economic Cooperation and Development.6(p30) fSource. Harvard–Louis Harris Interactive 1990 Ten-Nation Survey, cited by Blendon et al.7 gSource. Eurobarometer Survey, 1996, cited in Mossialos.8

consists of different plans for dif- Unlike Medicare, however, of coverage. Following its original lowing an agrarian metaphor—as ferent occupational groups, they French NHI coverage increases passage in 1928, the NHI pro- a set of 3 sprouting branches: all operate within a common as individual costs rise, there are gram covered salaried workers in (1) pensions, (2) family allow- statutory framework.35–37 Health no deductibles, and pharmaceu- industry and commerce whose ances, and (3) health insurance insurance is compulsory; no one tical benefits are extensive. In wages were under a low ceil- and workplace accident cov- may opt out. Health insurance contrast to Medicaid, French ing.38,39 In 1945, NHI was ex- erage.20 The first 2 are managed funds are not permitted to com- NHI carries no stigma and pro- tended to all industrial and com- by a single national fund, while pete by lowering health insur- vides better access. In summary, mercial workers and their the third is run by 3 main NHI ance premiums or attempting to French NHI is more generous families, irrespective of wage lev- funds: those for salaried workers micromanage health care. For than what a “Medicare for all” els. The extension of coverage (Caisse Nationale d’Assurance ambulatory care, all health in- system would be like in the took the rest of the century to Maladie des Travailleurs Salariés, surance plans operate on the United States, and it shares a complete. In 1961, farmers and or CNAMTS), for farmers and traditional indemnity model— range of characteristics with agricultural workers were cov- agricultural workers (Mutualité reimbursement for services ren- which Americans are well ac- ered; in 1966, independent pro- Sociale Agricole, or MSA), and dered. For inpatient hospital ser- quainted—fee-for-service prac- fessionals were brought into the for the independent professions vices, there are budgetary allo- tice, a public–private mix in the system; in 1974, another law (Caisse Nationale d’Assurance cations as well as per diem financing and organization of proclaimed that NHI should be Maladie des Professions Indépen- reimbursements. The French in- health care services, cost shar- universal. Not until January dentes, or CANAM). In addition, demnity model allows for direct ing, and supplementary private 2000 was comprehensive first- there are 11 smaller funds for payment by patients to physi- insurance. dollar health insurance coverage workers in specific occupations cians, coinsurance, and balance granted to the remaining unin- and their dependents, all of billing by roughly one third of NATIONAL HEALTH sured population on the basis of whom defend their “rightfully physicians. INSURANCE residence in France.40 earned” entitlements.41 Like Medicare in the United NHI forms an integral part of The CNAMTS covers 84% of States, French NHI provides a NHI evolved, in stages, in re- France’s social security system, legal residents in France, which great degree of patient choice. sponse to demands for extension which is typically depicted—fol- includes salaried workers, those

32 | International Perspectives Forum | Peer Reviewed | Rodwin American Journal of Public Health | January 2003, Vol 93, No. 1  INTERNATIONAL PERSPECTIVES FORUM 

TABLE 2—Health Care Resources: France and United States, 1997–2000 to cost sharing and selected ele- ments of la médecine libérale Resources France US (private practice): selection of Active physicians per 1000 population 3.3a (1998) 2.8a (1999) the physician by the patient, Active physicians in private, office-based practice per 1000 population 1.9b (2002) 1.7c (1999) freedom for physicians to prac- General/family practice, % 53.3b (2002) 22.5c (1999) tice wherever they choose, clin- Obstetricians, pediatricians, and internists, % 7.5b (2002) 35.6c (1999) ical freedom for the doctor, and Other specialists, % 39.2b (2002) 41.0c (1999) professional confidentiality. It is Nonphysician personnel per acute hospital bedd 1.9 (2001)e 5.7 (2000/01)f wrongly invoked, however, in Total inpatient hospital beds per 1000 populationg (1998) 8.5a 3.7a the case of fee-for-service pay- Short-stay hospital beds per 1000 population 4.0h (2000) 3.0i (1998) ment with reimbursement Share of public beds, % 64.2h (2000) 19.2i (1999) under universal NHI, for such a Share of private beds, % 35.8h (2000) 80.8i (1999) system is more aptly character- Proprietary beds as percentage of private beds (1999), % 56j 12i ized as a bilateral monopoly Nonprofit beds as percentage of private beds (1999), % 44j 88i whereby physician associations Share of proprietary beds, % 27k (1998) 10.7i (1999) accept the monopsony power of the NHI system in return for a 17 Source. Organization for Economic Cooperation and Development. the state’s sanctioning of their bSource. CNAMTS.18 cSource. National Center for Health Statistics.19 (These figures exclude federally employed physicians as well as all anesthesiologists, monopoly power. In the hospi- pathologists, and radiologists.) tal sector, liberalism provides d Nonphysician personnel include all hospital employees—administrative, technical, and clinical—excluding physicians.Among the category of the rationale for the coexis- physicians in the United States, we included chiropractors and podiatrists. eSource. CREDES.20 tence of public and proprietary fSource. Acute care beds: American Hospital Association21; nonphysician personnel: Bureau of Labor Statistics.22 hospitals, the latter accounting g These differences reflect the use of long-term care beds in French hospitals—public and private nonprofit—as nursing homes. for 27% of acute beds in hSource. DRESS.23 iSource. American Hospital Association.21 France in contrast to 10.7% in jSource: DRESS.24 the United States (Table 2). k 25 Source. DRESS. Also, unit service chiefs in pub- lic hospitals have the right to use a small portion of their who were recently brought into for fraud and abuse, and pro- ding to their occupational cate- beds for private patients. the system because they were vide a range of customer ser- gories. In France, the commit- The French tolerance for or- uninsured, and the beneficiaries vices for their beneficiaries. ment to universal coverage is ac- ganizational diversity—whether of 7 of the smaller funds that French NHI covers services cepted by the principal political it be complementary, competi- are administered by the ranging from hospital care, out- parties and justified on grounds tive, or both—is typically justi- CNAMTS.33 The CANAM patient services, prescription of solidarity—the notion that fied on grounds of pluralism. Al- and MSA cover, respectively, drugs (including homeopathic there should be mutual aid and though ambulatory care is 7% and 5% of the population, products), thermal cures in spas, cooperation between the sick dominated by office-based solo with 4% covered by the remain- nursing home care, cash bene- and the well, the active and the practice, there are also private ing 4 funds. fits, and to a lesser extent, den- inactive, and that health insur- group practices, health centers, All NHI funds are legally pri- tal and vision care. Among the ance should be financed on the occupational health services in vate organizations responsible different NHI funds, there re- basis of ability to pay, not actu- large enterprises, and a strong for the provision of a public ser- main small differences in arial risk.42 public sector program for mater- vice. In practice, they are quasi- coverage. nal and child health care. Like- public organizations supervised Smaller funds with older, ORGANIZATION wise, although hospital care is by the government ministry that higher-risk populations (e.g., OF HEALTH CARE dominated by public hospitals, oversees French social security. farmers, agricultural workers, including teaching institutions The main NHI funds have a net- and miners) are subsidized by The organization of health with a quasi-monopoly on med- work of local and regional funds the CNAMTS, as well as by the care in France is typically pre- ical education and research, that function somewhat like fis- state, on grounds of what is sented as being rooted in prin- there are, nevertheless, opportu- cal intermediaries in the man- termed “demographic compensa- ciples of liberalism and plural- nities for physicians in private agement of Medicare. They cut tion.” Retirees and the unem- ism.32,42 Liberalism is correctly practice who wish to have part- reimbursement checks for ployed are automatically invoked as underpinning the time hospital staff privileges in health care providers, look out covered by the funds correspon- medical profession’s attachment public hospitals.

January 2003, Vol 93, No. 1 | American Journal of Public Health Rodwin | Peer Reviewed | International Perspectives Forum | 33  INTERNATIONAL PERSPECTIVES FORUM 

TABLE 3—Use of Health Services: France and United States, 1997–2000 [3.3%], a specific tax on the [0.8%], Use France US and subsidies from the state Physician office visits per capitaa (1999) 6.0b 2.8c [4.9%].) The general social con- Specialist visits per capita (1999) 1.9b 1.4c tribution, a supplementary in- Hospital days per capita (1999) 2.4d 0.9d come tax (5.5% of wages and all Short-stay hospital days per capita (1999) 1.1d 0.7d other earnings) raised specifi- Admission rate for short-stay hospital services per 1000 population 170.1e (2000) 118.0f (1998) cally for NHI, was introduced in Average length of stay for all inpatient hospital services (1999) 10.6b 7.0g 19 91 to make health care fi- Average length of stay in short-stay beds (1999) 6.2e 5.9f nancing more progressive and to Per capita spending on pharmaceuticals, PPP,$ (1999) 484h 478h increase NHI revenues by en- MRIs per million population 2.5i (1997) 7.6i (1998) larging the tax base. As a share of total personal health care ex- Note. $PPP=purchasing power parity; MRI=magnetic resonance imaging unit. penditures, French NHI funds fi- aOrganization for Economic Cooperation and Development (OECD) Health Data has traditionally published a figure of around 6 physician consultations per capita for the United States.According to the 2002 edition, this figure is based on the National Health Interview Survey of the nance 75.5%, supplementary National Center for Health Statistics.This source, however, includes telephone contacts with physicians, as well as contacts with physicians in private insurance covers 12.4% hospital outpatient departments and emergency rooms (ERs).The French figure includes consultations with all physicians in private practice (7.5% for the nonprofit sector including health centers (5.4) and home visits by physicians (0.6). It excludes all telephone contacts and hospital outpatient and ER consultations.Thus, to obtain comparable data, the US figure is taken from the National Ambulatory Medical Care Survey (NAMCS), a survey of mutuelles and 4.9% for commer- visits to physicians’ offices, hospital outpatient departments, and ERs.According to the 1995 NAMCS, visits to physician offices account for 81% cial insurers), and out-of-pocket of ambulatory care use, and visits to emergency rooms and hospital outpatient departments account, respectively, for 11.2% and 7.8% of expenditures represent 11.1%.44 ambulatory care use.Taking these proportions into account, as well as the fact that patients are seen by physicians in only 71% of outpatient department visits, the 1999 per capita rate of physician visits would only increase to 3.04. Physicians in private practice bSource. CREDES.20 (and in proprietary hospitals) are c 19 Source. National Center for Health Statistics. (These figures exclude federally employed physicians as well as all anesthesiologists, paid directly by patients on the pathologists, and radiologists.) dSource. OECD.27 basis of a national fee schedule. eSource. Ministry of Health and Social Affairs.28 Patients are then reimbursed by f 29 Source. National Center for Health Statistics. their local health insurance gSource. National Center for Health Statistics.30 hThese figures, cited in Reinhardt et al,31 understate differences in the per capita volume of prescription drugs sold because increases in drug funds. Proprietary hospitals are prices have been significantly higher in France than in the United States since 1980.When expenditure data on prescription drugs in France reimbursed on a negotiated per and the United States are adjusted by the OECD index of pharmaceutical price inflation in both nations, the volume of prescription drug diem basis (with supplementary purchases in France exceeds that in the United States by a factor of 2. Source: OECD Health Data 1999, cited in S. Chambaretaud.26 iSource. OECD Health Data 2001. fees for specific services) and public hospitals (including pri- vate nonprofit hospitals working The private hospital sector in have developed a strong capac- agement and quality assurance in partnership with them) are France (both nonprofit and pro- ity for cardiac surgery and radi- activities. paid on the basis of annual prietary hospitals) has 36% of ation therapy. global budgets negotiated every acute beds, including 64% of all The number of nonphysician FINANCING AND year between hospitals, regional surgical beds, 32% of psychiat- personnel per bed is higher in PROVIDER agencies, and the Ministry of ric beds, and only 21% of med- public hospitals than in private REIMBURSEMENT Health. As for prescription ical beds.24 The nonprofit sec- hospitals; in the aggregate, it is drugs, unit prices allowable for tor, which operates only 9% of 67% lower than in US hospitals In 2000, roughly half of reimbursement under NHI are all beds, has over 44% of pri- (Table 2). This difference in hos- French NHI expenditures were set by a commission that in- vate long-term care beds.24 Pro- pital staffing may reflect a more financed by employer payroll cludes representatives from the prietary hospitals, typically technical and intense level of ser- taxes (51.1%) and a “general so- Ministries of Health, Finance, smaller than public hospitals, vice in US hospitals. It also re- cial contribution” (34.6%) and Industry. have traditionally emphasized flects differences between an levied by the French treasury on In contrast to Medicare and elective surgery and obstetrics, NHI system and the US health all earnings, including invest- private insurance in the United leaving more complex cases to system, which is characterized by ment income.43 (Remaining States, where severe illness usu- the public sector. Over the past large numbers of administrative sources of financing for the ally results in increasing out-of- 15 years, however, although and clerical personnel whose CNAMTS and its affiliated pocket costs, when patients be- there has been no change in its main tasks focus on billing many health insurance funds included come very ill in France their relative share of beds, the pro- hundreds of payers, documenting payroll taxes on employees health insurance coverage im- prietary sector has consolidated, all medical procedures per- [3.4%], special taxes on automo- proves. For example, although and many proprietary hospitals formed, and handling risk man- biles, tobacco and alcohol coinsurance and direct payment

34 | International Perspectives Forum | Peer Reviewed | Rodwin American Journal of Public Health | January 2003, Vol 93, No. 1  INTERNATIONAL PERSPECTIVES FORUM 

is symbolically an important part drugs. Owing to strict controls on the United States. Add to this the early 1970s.52 Much like the pro- of French NHI, patients are ex- capital expenditures in the health enormous choice of health deliv- spective payment system for empted from both when (1) ex- sector, France has fewer scanners ery options given to consumers, Medicare in the United States, penditures exceed approximately and magnetic resonance imaging the low level of micromanage- France has imposed strong price $100, (2) hospital stays exceed units than the United States. But ment imposed on health care control policies on the entire 30 days, (3) patients suffer from France stands out as having professionals, and the higher health sector. Greater cost con- serious, debilitating, or chronic more radiation therapy equip- level of population health status tainment has been achieved illness, or (4) patient income is ment than the United States, achieved by the French, and through such controls in France below a minimum ceiling, Japan, and the rest of Europe. some would argue that the than in the United States.32 thereby qualifying them for free In contrast to Great Britain French model is a worthy export Although the level of health supplementary coverage. and Canada, there is no public product. Others, however, would services use is high in France Charges for services provided perception in France that health emphasize the problems that ac- (Table 3), prices per service unit by health professionals—whether services are “rationed” to pa- company this model. are exceedingly low by US stan- in office-based practice, in outpa- tients. In terms of consumer satis- First, despite the achievement dards, and this has led to increas- tient services of public hospitals, faction (Table 1), a Louis Harris of universal coverage under NHI, ing tensions (physicians’ strikes or in private hospitals—are nego- poll placed France above the there are still striking disparities and demonstrations) between tiated every year within the United Kingdom, the United in the geographic distribution of physician associations and their framework of national agree- States, Japan, and Sweden.7 A health resources and inequalities negotiating partners—the NHI ments concluded among repre- more recent European study re- of health outcomes by social funds and the state. The allow- sentatives of the health profes- ports that two thirds of the popu- class.45,47,48 In response to these able fee for an office visit to a sions, the 3 main health lation is “fairly satisfied” with the problems, there is a consensus GP, for example, is only 20 €, insurance funds, and the French system.8 that these issues extend beyond and one half of all French physi- state. Once negotiated, fees must France also ranks high on health care financing and organi- cians are GPs. Physician special- be respected by all physicians ex- most measures of health status zation and require stronger pub- ists also receive low fees (23 €), cept those who have either cho- (Table 1). A recent report by the lic health interventions.49 except for cardiologists (46 €), sen or earned the right to engage Organization for Economic Coop- Second, there is a newly per- psychiatrists (36 €), and those in extra billing, typically special- eration and Development ceived problem of uneven qual- who do not accept assignment. ists located in major cities. In- (OECD), for example, indicates ity in the distribution of health The $55000 average net annual deed, in Paris, up to 80% of phy- that France is well above the services. In 1997, a reputable income of French physicians— sicians in selected specialties OECD average on a range of key consumer publication issued a salaried hospital-based doctors as engage in extra billing, in con- indicators.12 A more critical view list of hospitals delivering low- well as GPs and specialists in pri- trast to the national average of would emphasize that France has quality, even dangerous care.50 vate practice—is barely one third 20% among general practition- high rates of premature mortality Even before this consumer that of their US counterparts ers (GPs). In consulting these compared with the rest of Eu- awareness, there was a growing ($194000)53,54 (C. LePen and E. physicians, patients are reim- rope, but most analyses of this recognition that one aspect of Piriou, written communication, bursed only the allowable rate by phenomenon suggest that it has quality problems, particularly August 2002). In addition to NHI; supplementary insurance less to do with health care ser- with regard to chronic diseases price controls, capital controls on schemes cover the remaining ex- vices than with inadequate pub- and older persons, is the lack of the health system are stringent. penditures, with different limits lic health interventions to reduce coordination and case manage- They include limits on the num- set by each plan. alcoholism, violent deaths from ment services for patients. These ber of medical students admitted suicides and road accidents, and problems are exacerbated by the to the second year of medical SERVICES, PERCEPTIONS, the incidence of AIDS.45,46 anarchic character of the French school, controls on hospital beds AND HEALTH STATUS health system—what might be and medical technologies, impo- ACHIEVEMENTS, called the darker side of laissez- sition (since 1984) of global Existing data (Table 3)— PROBLEMS, AND faire.51 budgets on hospital operating ex- whether they come from surveys REFORM Third, although, compared penditures, and the more recent or are byproducts of the adminis- with the United States, France Juppé plan that imposed annual trative system—indicate consis- The French health care system appears to have controlled its expenditure targets for all NHI tently that, compared with Amer- delivers a higher aggregate level health care expenditures, within expenditures. icans, the French consult their of services and higher consumer Europe, France is still among the Prime Minister Juppé’s plan doctors more often, are admitted satisfaction with a significantly higher spenders. This has led the and more recent reforms have to the hospital more often, and lower level of health expendi- Ministry of Finance to circum- addressed the problems noted purchase more prescription tures, as a share of GDP, than in scribe health spending since the above; none of them, however,

January 2003, Vol 93, No. 1 | American Journal of Public Health Rodwin | Peer Reviewed | International Perspectives Forum | 35  INTERNATIONAL PERSPECTIVES FORUM 

have been solved. The Juppé out a “single-payer” system. To do evidence in support of this Note. A bibliography in English on the government established a slew this, however, will still require a proposition. Of course, it is easier French health care system is available on the author’s Web site at http://www.nyu. of national public health agen- statutory framework and an ac- to achieve this model before the edu/projects/rodwin/main.html. cies to strengthen disease sur- tive state that regulates NHI fi- emergence of a powerful com- veillance and monitor food nancing and provider reimburse- mercial health insurance industry Acknowledgments safety, drug safety, and the envi- ment. Of course, French NHI was such as exists in the United States I thank the R.W. Johnson Foundation ronment.55 It organized a new not designed from scratch as a today. Nevertheless, so long as for a Health Policy Investigator Award national agency, the Agence Na- pluralistic, multipayer system pro- NHI covers the insurance func- that enabled me to explore this topic and others. tional d’Accréditation et d’Évalu- viding universal coverage on the tions, why prevent the private in- I am grateful to Dr Robert Butler, ation en Santé, to promote basis of occupational status. It is surance industry from providing president and CEO, ILC-USA, and to my health care evaluation, prepare the outcome of sociopolitical useful services, on a contractual colleagues in the New York Group on Rekindling Health Care Reform for hospital accreditation proce- struggles and clashes among basis, under a NHI program? sponsoring the seminars and lecture on dures, and establish medical trade unions, employers, physi- Fourth, coverage of the re- which this article is based and for help- practice guidelines.56,57 It also cians associations, and the state. maining 1% of the uninsured in ful discussion during its preparation. I thank Claude LePen, William Glaser, set up regional hospital agencies This suggests that NHI in the France suggests that national re- Michael Gusmano, and Marc Duriez for with new powers to coordinate United States could similarly sponsibility for entitlement is their insights; Birgit Bogler, Gabriel public and private hospitals and emerge from our patchwork accu- more equitable than delegating Montero, and Eric Piriou for precious re- search assistance; and 3 anonymous 58 allocate their budgets. mulation of federal, state, and em- these decisions to local authori- French reviewers for provocative and In addition, the Juppé plan in- ployer-sponsored plans so long as ties. This lesson is consistent with thoughtful comments. cluded measures to modernize we recognize the legitimate role the experience of Medicare ver- the French health care system by of government in overseeing the sus Medicaid in the United States, References improving the coding and collec- rules and framework within as exemplified by the differences 1. World Health Report 2000. Avail- able at: http://www.who.int/whr/2001/ tion of information on all ambu- which these actors operate. among states and counties in archives/2000/en/index.htm. Accessed latory care consultations and pre- Second, the evolution of French dealing with the uninsured. October 18, 2002. scriptions and by allowing NHI demonstrates that it is possi- Finally, and perhaps most im- 2. Coyne JS, Hilsenrath P. The World experiments to improve the coor- ble to achieve universal coverage portant for the United States, the Health Report 2000: can health care systems be compared using a single dination of health services. This without a “big bang” reform, since French experience suggests that it measure of performance? Am J Public represents an emerging form of this was accomplished in incre- is possible to solve the problem of Health. 2002;92:30, 32–33. French-style managed care—a mental stages beginning in 1928, financing universal coverage be- 3. Navarro V. The World Health Re- centrally directed attempt to ra- with big extensions in 1945, fore meeting the challenge of port 2000: can health care systems be compared using a single measure of tionalize the delivery of health 19 61, 1966, 1978, and finally in modernizing and reorganizing the performance? Am J Public Health. 2002; 51 services. The institutional barri- 2000. Of course, the extension of health care system for the 21st 92:31, 33–34. ers to such reform are consider- health insurance involved political century. The Clintons’ plan at- 4. Murray C, Frenk J. World Health able, but whatever transpires in battles at every stage.13 , 3 8 In the tempted to do both and failed. Report 2000: a step towards evidence- the future, the French experience United States, since it is unlikely France may be more prepared based health policy. Lancet. 2001;357: 1698–1700. with NHI may be instructive for that we will pass NHI with one and willing to implement the Clin- 5. Navarro V. World Health Report the United States. sweeping reform, we may first tons’ plan than the United States. 2000: a response to Murray and Frenk. have to reject what Fuchs calls the The United States would do bet- Lancet. 2001;357:1701–1702. LESSONS FOR THE “extreme actuarial approach” of ter to follow the French example 6. A Caring World: The New Social UNITED STATES our private health insurance sys- in solving the tough entitlement Policy Agenda. Paris: Organization for 60 Economic Cooperation and Develop- tem and then accept piecemeal issues before restructuring the en- ment; 1999:27. Perceptions of health systems efforts that extend social insurance tire health care system. 7. Blendon R, Leitman R, Morrison I, abroad can become caricatures coverage to categorical groups be- Donelan K. Satisfaction with health sys- of what we wish to promote or yond current beneficiaries of pub- tems in ten nations. Health Aff(Mill- wood). 1990;9(2):185–192. avoid at home. It is thus a risky lic programs. About the Author Victor G. Rodwin is with the Wagner venture to derive lessons from Third, French experience dem- 8. Mossialos E. Citizens’ views on School, New York University, New York, health care systems in the 15 member the French experience for health onstrates that universal coverage NY, and the World Cities Project, New states of the European Union. Health care reform in the United States. can be achieved without exclud- York, a joint venture of NYU Wagner and Econ. 19 97;6:109–116. the International Longevity Center-USA. 9. de Kervasdoué J. Pour une Révolu- Nonetheless, I set forth 5 propo- ing private insurers from the sup- Requests for reprints should be sent to tion sans Réforme. Paris, France: Galli- sitions to provoke further debate. plementary insurance market. Victor G. Rodwin, PhD, MPH, 4 Wash- mard; 1999. First, the French experience The thriving nonprofit insurance ington Sq North, New York, NY 10003 (e-mail: [email protected]). 10. Le Pen C. Les Habits Neufs d’Hip- demonstrates that it is possible to sector (mutuelles)aswell as com- This article was accepted September pocrate. Paris, France: Calmann-Lévy; achieve universal coverage with- mercial companies (e.g., Axa) are 10, 2002. 1999.

36 | International Perspectives Forum | Peer Reviewed | Rodwin American Journal of Public Health | January 2003, Vol 93, No. 1  INTERNATIONAL PERSPECTIVES FORUM 

11. Sorum P. Striking against managed 26. Chambaretaud S. 2000. La Con- Circonscription de Caisse au 31 Décembre 55. Jourdain A, Duriez M, eds. Les care: the last gasp of la médecine sommation de Médicaments dans les 1999. Paris, France: Caisse Nationale agences dans le système de santé. Actu- libérale? JAMA. 1998;280:659–664. Principaux Pays Industrialisés. Paris, d’Assurance Maladie des Travailleurs alité et Dossier en Santé Publique. De- France: Direction de la Recherche, des Salaries (CNAMT); May 2001. Dossier cember 2001;37:18–60. 12. Imai Y, Jacobzone S, Lenain P. The Etudes, de l’Évaluation et des Statis- Études et Statistiques no. 48. Changing Health System in France. Paris, 56. Matillon Y, Loirat P, Guiraud- tiques (DRESS); 2000. Études et Résul- France: Economics Department, Organi- 42. Rodwin V. The marriage of na- Chaumeil B. Les rôles de l’ANAES dans tats no. 47. zation for Economic Cooperation and tional health insurance and la médecine la régulation du système de santé Development; November 2000. Work- 27. OECD Health Data, 2001. Paris, libérale: a costly union. Milbank Mem Français. Actualité et Dossier en Santé ing Paper 268. France: Organization for Economic Co- Fund Q Health Soc. 1981;59:16–43. Publique. December 2001;37:46–50. operation and Development; 2001. 13. Hatzfeld H. Le Grand Tournant de 43. Rapport de la Commission des 57. Durieux P, Chaix-Couturier C, Du- la Médecine Libérale. Paris, France: Edi- 28. Programme de Médicalisation des Comptes Nationaux de la Santé, 2000. rand-Zaleski I, Ravaud P. From clinical tions Ouvriéres; 1963. Systèmes d’Information. Paris, France: Paris, France: Direction de la recommendations to mandatory prac- Ministry of Health and Social Affairs; Recherche, des Études de l’Évaluation tice: the introduction of regulatory prac- 14 .Wilsford D. Doctors and the State: 2000. et des Statistiques (DRESS); 2001. tice guideline in the French healthcare The Politics of Health Care in France and system. Int J Technol Assess Health Care. the United States. Durham, NC: Duke 29. 1998 National Hospital Discharge 44. Commission des Comptes de la Sécu- 2000;16:969–975. University Press; 1991. Survey. Washington, DC: National Cen- rité Sociale. Paris, France: Direction de ter for Health Statistics, Centers for Dis- la Recherche, des Études de l’Évalua- 58. Coudreau D. Les agences ré- 15. Immergut E. Health Politics: Inter- ease Control and Prevention; 1999. tion et des Statistiques (DRESS); 2002. gionales de l’hospitalisation dans le sys- ests and Institutions in Western Europe. tème de santé. Actualité et Dossier en Cambridge, England: Cambridge Uni- 30. Health in the United States. Wash- 45. Haut Comité de la Santé Publique. Santé Publique. December 2001;37: versity Press; 1992:chap 3. ington, DC: National Center for Health La Santé des Français. Paris, France: La 50–54. Statistics, Centers for Disease Control Découverte; 1999. 16. Grignon M. Quel filet de sécurité and Prevention; 2001. 59. Lancry PJ, Sandier S. Rationing pour la santé? Une approache 46.Rodrigue JM, Garros B. Regards health care in France. Health Policy. économique et organisationelle de la 31.Reinhardt U, Hussey P, Anderson sur la Santé des Français. In: de Kervas- 1999;50:23–38. couverture maladie universelle. Revue G. Cross national comparisons of health doué J, ed. Le Carnet de Santé de la Française des Affaires Sociales. 2002;2: systems using 1999 OECD data. Health France en 2000. Paris, France: Mutual- 60.Fuchs V. What’s ahead for health 145–176. Aff (Millwood). 2002;21(3):169–181. ité Française; 2000, insurance in the United States? N Engl J Med. 2002;346:1822–1824. 17. OECD Health Data, 2002. Paris, 32. Rodwin V, Sandier S. Health care 47. Salem G, Stephane R, Jougla E. Les France: Organization for Economic Co- under French national health insurance. Causes de Décès. London, England: John operation and Development; 2002. Health Aff (Millwood). 19 93;12(3): Libbey Eurotext; 1999. Atlas de la 113–131. Santé en France; vol 1. 18. Carnets Statistiques no. 108. Paris, France: Caisse Nationale de l’Assurance 33. Sandier S, Polton D, Paris V, 48. Leclerc A, Fassin D, Grandjean H, Maladie des Travailleurs Salariés Thompson S. France. In: Dixon A, Kaminski M, Lang T, eds. Les Inégalités (CNAMTS); 2002. Mossialos E, eds. Health Care Systems in Sociales de Santé. Paris, France: La Dé- Eight Countries: Trends and Challenges. couverte/INSERM; 2000. 19. 1999 National Ambulatory Medical London, England: London School of Care Survey. Washington, DC: National 49. Got C. Risquer Sa Peau. Paris, Economics and Political Science; 2002: Center for Health Statistics, Centers for France: Bayard; 2001. 30–45. Disease Control and Prevention; 1999. 50. La Liste Noir des Hôpitaux. Paris, 34. Catrice-Lorrey A. Dynamique In- 20. Eco-Santé 2001. Paris, France: France: Sciences et Avenir; October terne de la Sécurité Sociale. Paris, France: 19 97. Centre de Recherche, d’Étude et de Economica; 1982. Documentation en Economie de la 51.Rodwin V. The rise of managed Santé (CREDES); 2001. 35. White J. Competing Solutions: care in the United States: lessons for American Health Care Proposals and In- 21. Hospital Statistics 2000. Chicago, French health policy. In: Altenstetter C, ternational Experience. Washington, DC: Bjorkman JW, eds. Health Policy Reform, Ill: American Hospital Association; Brookings Institute; 1995:chap 4 and 5. 2001. National Variations, and Globalization. 36. Glaser W. Health Insurance in Prac- New York, NY: St Martin’s Press; 1997: 22. National industry specific occupa- tice. San Francisco, Calif: Josey Bass; 39–58. tional employment and wage estimates, 19 92. 52. Rodwin V. Management without specific industry code 806, hospitals, objectives: the French health policy US Dept of Labor, Bureau of Labor Sta- 37. Dupéyroux JJ. Sécurité Sociale. gamble. In: McLachlan G, Maynard A, tistics. Available at: www.bls.gov/oes/ Paris, France: Dalloz; 1997 eds. The Public/Private Mix for Health. 2000/oesi3_806.htm. Accessed Octo- 38. Galant H. Histoire Politique de la London, England: Nuffield Provincial ber 29, 2002. Sécurité Sociale Française. Paris, France: Hospitals Trust; 1982:289–325. 23. L’activité des Établissements de Santé Armand Colin; 1955. 53. Data from the Socio-Economic Mon- en 2000. Paris, France: Direction de la 39. Duriez M, Lancry JP, Lequet-Slama itoring System 1984–1999. Chicago, Ill: Recherche, des Études, de l’Évaluation D, Sandier S. Le Système de Santé en American Medical Association. Avail- et des Statistiques (DRESS); 2002. France. Paris, France: Presses Universi- able at: http://www.ama-assn.org/ama/ Études et Résultats no. 177. taires de France; 1996. pub/category/7801.html. Accessed No- 24. Les Établissements de Santé en 40. Boisguerin B. La CMU au 31 Mars vember 20, 2002. 1999. Paris, France: Direction de la 2002. Paris, France: Direction de la 54. Audric S. Les Disparités de Revenus Recherche, des Études de l’Évaluation Recherche, des Études de l’Évaluation et de Charges des Médecins Liberaux. et des Statistiques (DRESS), Ministère et des Statistiques (DRESS); July 2002. Paris, France: Direction de la de l’Emploi et de la Solidarité; 2001. Études et Résultats no. 179. Recherche, des Études de l’Évaluation 25. Annuaire des Statistiques Sanitaires 41. La Population Protégée par les et des Statistiques (DRESS), Ministère et Sociales 1999. Paris, France: DRESS Régimes de Sécurité Sociale: Répartition de l’Emploi et de la Solidarité; 2001. Collection Études et Statistiques; 2000. Géographique par Département et par Études et Résultats no. 146.

January 2003, Vol 93, No. 1 | American Journal of Public Health Rodwin | Peer Reviewed | International Perspectives Forum | 37