Challenges for health systems in member countries of the Organisation for Economic Co-operation and Development.

Challenges for health systems in Member Countries of the Organisation for Economic Co-operation and Development Jeremy Hurst1 For reasons of equity most OECD countries have chosen to base their funding of health care mainly on public sources. There is an almost universal problem of affordability in the health systems of these countries, arising from the tension between the willingness of populations to pay taxes and the eagerness of patients to use health services where these are free or heavily subsidized at the point of use. These tensions are likely to be exacerbated by a surge of new medical technologies adding to demands for health care. Some observers have predicted the breakdown of publicly funded systems of health care under new spending pressures. However, governments can deploy a range of policies for handling new demands. They can also take comfort from the fact that many of them have already coped with successive waves of technological change in health care without abandoning their core commitment to the public funding of health systems. Furthermore, if standards of living continue to rise, public and private insurers should find it easier to obtain the revenues needed to pay for the improved health care expected by consumers. Keywords: financing, health; financing, government; health expenditures; health services accessibility; health status; health care reform; biomedical technology; international agencies. Voir page 758 le re´sume´ en franc¸ais. En la pa´gina 759 figura un resumen en espan˜ol. Introduction In most respects the problems faced by health systems in countries of the Organisation for Economic Co-operation and Development (OECD) are less severe than those in other countries. The 29 Member Countries of the OECD are industria- lized countries with high or middle per capita incomes. On average their populations enjoy a relatively high health status. Their economies, for the most part, support adequate or more than adequate levels of health expenditure. Moreover, most OECD countries have organized the financing of their health care systems in such a way that the healthy support the sick, the young support the old, and the rich support the poor. On the demand side of health systems, people who reach old age tend to become sick as they become older. A high proportion of total health expenditure in OECD countries is devoted to patients who are within two years of death. Affluence can encourage unhealthy habits and pose threats to health, not least via side- effects on the environment. Some groups in OECD countries, for instance the unemployed and the poor, tend to be excluded from the general health advances enjoyed by the majority or do not benefit to the same extent. Meanwhile, among the majority, rising stan- dards of living, a growing cascade of new medical technologies and increasing consumerism raise ex- pectations about what can and should be done to improve health protection and alleviate the remaining burden of disease. On the supply side, economic growth and technological changes tend to raise the relative cost of protecting public health and of supplying populations with personal medical care and the latest medical advances. Mainly for reasons of equity the govern- ments of most OECD countries have become heavily involved in the regulation, financing and sometimes the provision of medical care. Conse- quently, the problem of reconciling rising demand and increasing costs falls mainly in the public sector. It follows that most, if not all, OECD governments carry heavy responsibilities for achieving, or con- tinuing to achieve, good performance (1) in their health systems. Achievements of OECD health systems Over the past 35 years the average life expectancy at birth in the Member Countries of the OECD has increased by about eight years for women and seven years for men. Fig. 1 shows trends in average life expectancy for men and women in 13 OECD countries during the period 1960–96. The increase was largely attributable to a halving in years of life lost before the age of 70. In recent years the threat of an 1 Head of Health Policy Unit, Organisation for Economic Co-operation and Development, 2, rue Andre´-Pascal, 75775 Paris Cedex 16, France. Ref. No. 00-0588 751Bulletin of the World Health Organization, 2000, 78 (6) AIDS epidemic, which might have interrupted or reversed some of the gains in life expectancy, has been contained. There has been much debate about whether improved life expectancy has been offset by increas- ing morbidity, particularly among the elderly. Evi- dence from household surveys in some OECD countries indicates that the contrary is true. Morbid- ity, in the form of severe disability, seems to have been declining for some years in the majority of OECD countries for which data are available, at least in the younger age groups among the elderly (2). The gains in health status are not solely attributable to improvements in health care and public health. Increasing educational status and rising standards of living over several generations may have been more important. Nevertheless, multiple regres- sion analysis of the determinants of premature mortality suggests that, in addition to such factors, increases in numbers of physicians per capita, admissions to hospital and real expenditure on pharmaceuticals have all been associated significantly with falling premature mortality (3, 4). There is only limited evidence as to whether health systems in OECD countries achieve the goal of being responsive to the legitimate expectations of their populations. In 1996, people in the 15 countries of the European Union were asked about their satisfactionwith health care systems (5). Over 60%of respondents in nine countries said they were very satisfied or fairly satisfied with the way health care was run. However, the corresponding figure was below 30% inGreece, Italy, and Portugal. There were clear signs of a positive association between the percentage of respondents saying they were very satisfied or fairly satisfied with health care and the level of health expenditure per capita. From a survey of this kind it is difficult to determine the extent to which responses reflect the personal experiences of users of health services. They may reflect impressions of a range of performance characteristics of health services among people who are mainly non-users. The measurement of respon- siveness requires a more targeted approach. In this connection there is a little evidence concerning one of the responsiveness characteristics proposed by WHO, namely prompt attention to health needs. Information has been collected in some countries on waiting times for visits to specialists or admission to hospital for surgical operations (6). In one survey, for example, waiting times for visits to specialists after referral by general practitioners were, on average, less than two weeks in 8 of 13 countries but more than three weeks in the others (7). The available evidence suggests that average waiting times vary inversely and significantly with both health expenditure and physician numbers per capita. The great majority of OECD citizens have been freed from the worry of unaffordable bills for medical care. About a billion people, i.e. more than 90% of the population, are effectively insured against the cost of medical care that is basic or better than basic. Moreover, since the bulk of health insurance coverage is public, payment for it tends to reflect ability to pay, or at least is more in accordance with ability to pay than payment for private health insurance or out-of-pocket payments would be. Taking account of both public and private sources of finance, research conducted in 12 OECD countries during the late 1980s and early 1990s suggested that overall payment for health care was mildly progressive in five countries, mildly regressive in five and moderately regressive in two (8). In this context, ‘‘progressive’’ means that the rich paid a higher proportion of their income for health care than the poor, and ‘‘regressive’’ means the reverse. The suggestion is that the financing of health care is often more in accordance with ability to pay than would have been the case in a wholly private financing system. In certain countries there is also evidence that treatment is broadly in accordance with need, i.e. morbidity, in publicly financed systems (9). Since morbidity is associated with poverty it appears that many OECD health care systems act as a means of redistributing income from rich to poor. It is difficult to produce evidence on trends in efficiency across OECD countries in terms of the ratio of goal achievements to inputs, because we lack comprehensive information on the outputs of health care systems. Fig. 2 shows that inputs, in the form of total real health expenditure per capita, rose over sixfold between 1960 and 1997 in the 18 OECD countries for which data are available. Part of the reason for the increase was that real gross domestic product (GDP) per capita went up nearly threefold in these countries over the same period. Fig. 3 suggests that there were periods when costs were contained in the sense that the health expenditure share of GDP was stabilized. On average, for example, stabilization of the health expenditure share in the 18 countries occurred between 1992 and 1997. Remaining problems In discussing the many problems that remain in health care systems it should be noted that the Fig. 1. Average life expectancy at birth in 13 OECD countries, 1960–96 Special Theme – Health Systems 752 Bulletin of the World Health Organization, 2000, 78 (6) OECD represents groups of countries at somewhat different stages of economic development and with a variety of health care institutions. These groups face rather different problems in relation to their health systems. Mexico and Turkey are middle-income coun- tries according to the World Bank’s classification (10). They are still undergoing epidemiological transition, with infant mortality levels at more than twice the average in the OECD. Their respective health expenditures are under US$ 500 per capita at purchasing power parity exchange rates (10, 11). Their numbers of physicians per capita are less than half the OECD average. Significant minorities of the populations in both countries, mainly in rural areas, lack full basic health insurance coverage or access to adequate public care, although steps are being taken to improve this situation. The Czech Republic, Hungary, and Poland are alsomiddle-income countries according to theWorld Bank’s classification. However, their infant mortality rates are quite close to the levels observed in high- income OECD countries. Their respective health expenditures are under US$ 1000 per capita at purchasing power parity exchange rates. The num- bers of physicians per capita, however, are at or above the OECD average. There is universal access to public health care and there are high public shares of health expenditure. The health care systems have been reformed since 1989, with the result that they have changed from public integrated systems of the ‘Semashko’ design to social health insurance systems of the ‘Bismarck’ variety (12). Unfortunately, a legacy of informal, under-the-table payment for health care may, in varying degrees, continue to present barriers to access for the poor. Greece, the Republic of Korea, and Portugal are also classified by the World Bank as middle- income countries. Their infant mortality levels are close to the OECD average. Their respective health expenditures are between US$ 800 and US$ 1100 per capita at purchasing power parity exchange rates. While the Republic of Korea has less than half the OECD average number of physicians per capita, Greece and Portugal have more than the average. All three countries have relatively high private shares of health expenditure. The Republic of Korea, which combines universal public health insurance with high co-payments, has the highest reported private share of total health expenditure in the OECD at 55%. The remaining 21 Member Countries are classified by the World Bank as high-income countries. Among these, a group including Ireland, New Zealand, the Nordic countries, Spain and the United Kingdom, has levels of health expenditure between about US$ 1000 and US$ 2000 per capita. They have high public shares of total health expenditure and they tend to rely on general taxation as a major source of health care finance. Their hospitals are mainly public and they pay their hospital physicians, and sometimes their general practitioners, mainly by salary or capitation. These countries enjoy high health status and equity in access to health services but patients often face long waiting times for specialist care. There is a tendency to be concerned about the extra load that health expenditure imposes on general taxation, about waiting times for specialist care, and about low productivity in parts of the health care systems. A second group of high-income countries includes Australia, Austria, Belgium, Canada, France, Germany, Italy, Japan, Luxembourg, the Nether- lands, and Switzerland. They have levels of health expenditure between about US$ 1500 and US$ 2700 per capita. They have high public shares of total health expenditure but mostly tend to rely on compulsory health insurance contributions for the bulk of their health care finance. Australia and Canada, however, rely on general taxation for this purpose. There is a greater dependence on private, often non-profit, hospitals than in the previous group and there is a tendency to pay physicians on a fee-for- service basis. As in the previous group, there are high levels of health status and equity in access. Waiting times for specialist care are shorter, sometimes much shorter, than in the previous group.However, there is a tendency to be concerned about the burden of health insurance contributions on employers and employees, supplier-induced demand, and the con- trol of rising health expenditure. Fig. 2. Average per capita real total expenditure on health and real GDP indices in 18 OECD countries, 1960–97 Fig. 3. Average total expenditure on health as a percentage of GDP in 18 OECD countries, 1960–97 Challenges for health systems in OECD countries 753Bulletin of the World Health Organization, 2000, 78 (6) Finally, the USA’s level of health expenditure is over US$ 4000 per capita. The most important source of health care finance in America is voluntary, employer-related, private health insurance, although there are large public insurance programmes for the elderly and the poor. Approximately 15% of citizens lack health insurance cover. The bulk of hospitals are private and physicians have traditionally been paid mainly on a fee-for-service basis. Americans were concerned for many years about rising health care costs and incomplete insurance coverage. However, the 1990s brought a big shift in private health insurance from open-ended, fee-for-service arrange- ments to closed-ended, health maintenance organi- zations practising managed care for cost control. In recent years, health expenditure has consequently stabilized as a share of GDP. A new concern among many insured Americans is about restrictions on access and allegations about reductions in the quality of care arising from the managed care approach. Up to 100 million OECD citizens, concen- trated in a small number of countries, still lack access to basic health insurance coverage or to adequate ‘free’ public care. The number of such people has been dwindling in some countries as the public share of health expenditure has grown. Fig. 4 indicates that, in the six countries with the lowest public share of health expenditure in 1970, the public share rose from 40% in 1970 to 58% in 1997. Clearly, the levels of health expenditure and physicians per capita vary widely. The evidence on both health outcomes and waiting times suggests, if only superficially, that countries tend to get what they pay for. Premature mortality declines as physician numbers increase, and waiting times for specialist care decline as expenditure per capita rises. Common underlying difficulties Some underlying difficulties in health care systems are common to all, or nearly all, OECD countries. Equality in health status has not been achieved even where there has been universal access to reasonable standards of health care formany decades. It is true that there has been increased equity in health status almost everywhere in the sense that there has been convergence in respect of the average age of death. Increasing numbers of OECD citizens are living what has conventionally been regarded as a full life span. In most countries, however, a minority of the population, usually composed of a dwindling number of people, has not kept up with the general advances in health status. The growing evidence of the persistence of health inequalities, notwithstanding universal access to adequate medical care, has been one of the reasons for a revival of interest in public or population health measures. Another reason has been the growth of evidence about the importance of, for example, educational, lifestyle and environmental factors in the etiology of disease. In response, many countries are developing preventive and population health strate- gies to tackle the root causes of the remaining disease burden. Patients using publicly funded health care systems have incentives to demand more medical care and expenditure than would have been the case had they remained uninsured. There is ‘moral hazard’ when health services are free of charge or heavily subsidized at the point of use. Moreover, if providers are paid on a fee-for-service basis there is no incentive for them to economize if a third party is responsible for payment. Patients and providersmay, in effect, collude in expanding the volume and price of medical care. Meanwhile, those paying the taxes that support the systems have every incentive to resist higher taxes. The conflict between the propensity of patients to demand more health services and the unwilling- ness of taxpayers to supply additional public finance for health care is likely to intensify as the redis- tributive character of publicly funded health services increases. Although many transfers of health ex- penditure are reciprocated over the life spans of individuals, at any one time those paying the bills tend to be relatively wealthy and healthy, while those using the services tend to be relatively sick and poor. This creates a potential for political conflict, especially when health expenditure is tending to rise, unless feelings of social solidarity are strong. Governments seek ways of controlling expen- diture if they are trying to limit the use of price in decision-making on access to and allocation of health care. On the demand side, the strategies include limiting what is covered in the basic package of care, imposing partial cost-sharing on patients, and imposing global budgets on all or part of health expenditure. On the supply side, they include various volume and cost controls. The most successful cost- control device inmany countries has proved to be the global budget, making health expenditure predictable for both payer and provider. If the predicted level is not exceeded, managers and professionals are relatively free to make purchasing and provision decisions. Global budgets were deployed many decades ago in most of the public systems funded by general taxation. They were often introduced during the 1980s in public systems funded by social insurance (13, 14). More recently, private insurers Fig. 4. Public share in total spending on health as a percentage of GDP in OECD countries, 1970–97 Special Theme – Health Systems 754 Bulletin of the World Health Organization, 2000, 78 (6) have introduced, under managed care, cost contain- ment devices analogous to those long deployed by governments in public systems. If cost containment does not involve cost- sharing it leads to non-price rationing. Thismaymean restrictions on services, hasty impersonal treatment and queuing. As indicated above, there is evidence that waiting times for specialist care in particular tend to vary inversely with health expenditure. Further- more, rationing is unpopular, especially if it leads to absolute reductions in services rather than restraint in the rate of growth of services. This applies to cost containment by both private and public insurers. During the recent period of cost containment of health expenditure in OECD countries there have been well-publicized adverse reactions among con- sumers to increased rationing in systems as different as the publicly funded National Health Service in the United Kingdom and the privately funded health maintenance organizations in the USA. Two errors can be made by governments when setting health expenditure: spending too much or spending too little. In principle, health expenditure is optimal at the point where the marginal benefits of health care are equal to the marginal costs. However, there is always uncertainty about where this point lies, particularly when there is a complex public/private mix of financing. In practice it becomes a matter of judgement, based on a reading of public opinion, as it is difficult to measure and value health outcomes. Such judgement may, of course, be informed by evidence about factors such as the effectiveness of particular services at the margin or the comparative lengths of waiting lists. The wide variations in health expenditure between OECD countries with similar standards of living suggest that different judgements have been made. Moreover, judgements may change with time. As health care costs have risen, for example, some governments have allowed the private share of health expenditure to increase slightly, usually by increasing cost-sharing among patients or by limiting the scope of services covered by public schemes to those regarded as basic or core in character. Fig. 4 indicates that, in the six OECD countries with the highest public shares of health expenditure in 1970, the average public share fell from 90% in that year to 84% in 1997. In this instance the smallness of the change is what is significant. There remain widespread perceptions of inefficiency in health care delivery in most, if not all, OECD health care systems. For example, there is much evidence of large and, on the basis of what is known about morbidity, inexplicable variations in the rates of utilization of medical care between small geographical areas in particular countries. The usual explanation is that in the absence of objective measures of need and without agreed protocols for care, clinicians in different areas develop their own habits relating to treatment rates on the basis of subjective impressions of need and convictions about clinically appropriate treatments. The goal of improving the efficiency and performance of health care systems has great appeal for governments and for insurers generally. Increas- ing the productivity of health care offers the promise of easing the tensions between the demands for better health care and the pressures for lower costs. OECD countries have consequently introduced many reforms of their health care systems. These reforms have usually focused on one or more of five key areas: incentives, information, management, regulation, and the mix of services. Attempts to change incentives have included: the separation of purchasers and providers, adopted in the 1990s by several formerly integrated publicly funded health systems; encouragement of competition between providers and, sometimes, between insurers; and privatization of selected sources of supply, such as non-clinical support services in hospital and long- term nursing care. There has also been much tinkering with methods of paying health care professionals. Attempts to improve information have included the adoption of evidence-based medicine; investment in better costing and pricing of services; and the growing use of health technology assessment for new medical technologies. Attempts to improve management have included the growing professionalization of the management function and the importation of private sector managers into public systems. Attempts to improve regulation have included the adoption of performance management of health services. Changes in the mix of services have included widespread shifts away from inpatient hospital care and towards outpatient and day care in acute services, and towards nursing home and domiciliary care in long-term care. There is growing interest in moving expenditure towards population health measures with a view to preventing disease, particularly among disadvantaged groups. Health care reforms have often disappointed their instigators. Attempts to change incentives have sometimes had perverse side-effects. The acquisition of relevant information can be very costly. Increased performance management has been seen as under- mining clinical freedom and has sometimes been demotivating. Central to this issue are well-known difficulties such as the asymmetry of knowledge betweenmanagers and professional staff, particularly doctors, and the monopoly power often acquired by providers. Unfortunately, there has been a lack of rigorous evaluation of many reforms and a con- sequent lack of evidence about their effects on efficiency, apart from ‘favourable’ indications de- rived from process measures such as increasing activity, increasing use of day-case surgery, and falling lengths of stay in hospital. The general effect of many of the changes and reforms outlined above has been a mild convergence in health care institutions between OECD countries. A good number of formerly integrated public systems have introduced the purchaser/provider separation that was formerly found mainly in social insurance systems, many of which have adopted Challenges for health systems in OECD countries 755Bulletin of the World Health Organization, 2000, 78 (6) global budgets such as were formerly typical of the integrated systems. Countries with a high public share of health spending have sometimes trimmed it by shifting some financing to the private sector. Countries with low public health insurance coverage have been increasing the public share of their health expenditure. Public insurers have been looking for ways to contract out some of their services to private providers. Meanwhile, some private insurers have been adopting cost containment devices formerly found only in public systems. New challenges In all OECD countries the demand for health care is expected to rise. To some extent the determinants are demographic and epidemiological. Between a third and a half of health expenditure is incurred by people aged 65 and over in many of these countries. The proportion of such people is expected to rise, especially during the second decade of the 21st century. This implies an increase in cases of chronic diseases associated with ageing, such as dementia, cardiovascular disease and musculoskeletal condi- tions. Sharp reductions in age-specific morbidity would be necessary to offset these trends. Increasing incidences of diseases associated with affluence, such as alcohol abuse, asthma and obesity-related diabetes, can also be expected to add to demands. Costs can be expected to continue rising, partly because health services are labour-intensive. Their costs tend to rise in line with average wages unless there are productivity improvements. The available evidence indicates that there was amean relative price rise of 0.7% per annum in the cost of health care across 16 OECD countries between 1980 and 1996 (15). That could merely suggest that the price indicators for health care services do not adequately measure improvements in outcome and quality, and hence in productivity. Alternatively, it may be that productivity improvements in health care tend to lag behind improvements in productivity elsewhere in OECD economies. However, the main reason for rising demands and costs is likely to be a flood of new medical technologies, many of them representing additional services with a potential for tackling the remaining burden of disease. Some of the most important advances are likely to come in the field of biotechnol- ogy, the subject of a whole programme of work at the OECD (16). For example, the human genome project is mapping the entire human genetic code, and this is likely to point the way to treatments for many diseases that are currently incurable. The increasing scope for genetic manipulation is likely to cause great ethical difficulties for insurers and providers, as well as adding new demands for health care interventions. Parallel developments in deciphering the genetic make-up of new or resurgent infectious diseases such as AIDS, malaria and tuberculosis may lead to new ways of controlling or eliminating them. The arrival of new technologies is likely to coincide with enhanced expectations among patients and potential beneficiaries. Already there are many signs that consumers are becoming better informed and more assertive about diagnostic and treatment choices. Improved vehicles of communication, such as the Internet, can be expected to favour the rapid dissemination of news about therapeutic advances to potential beneficiaries. There is also likely to be a rise in consumerism and a growing tendency for patients to enter into litigation if they suspect that they have been victims of rationing or medical negligence. Some of the new technologies may lead to reductions in costs. However, most are likely to have the reverse effect, at least in the short run. Many of them offer extra opportunities to intervene in what was previously regarded as the natural course of unavoidable diseases. Many are bound to be more expensive than the technologies they replace. For example, many of the products of the biotechnology revolution are likely to be owned by commercial enterprises, allowing monopoly prices to be charged for them until the patents expire. Insurers have the opportunity to negotiate with companies over prices, but if enough OECD countries were to force prices down towards the costs of production the incentives for biotechnology companies to make further innovations would be undermined. The combination of these trends in the demand for and cost of health services is likely to mean an increase in the desired share of publicly financed health care in GDP. However, tax revenues are likely to increase at the same rate as national income if tax policies remain unchanged. Conse- quently, the conflict between the demands for and supply of public finance for health care is likely to intensify in the absence of deficit financing or cuts in other areas of public expenditure. This conflict could be particularly acute in middle-income countries, many of which are already spending a higher proportion of their health care budget on imported drugs and medical equipment than is the case inmost higher-income countries (17). Moreover, some middle-income countries can be expected to try to improve access to health care, at public expense, for those people who are still uninsured. Implications These persistent problems and new challenges present governments, voters and consumers with increasingly difficult choices. Some observers are predicting breakdown for those countries with high public shares of total health expenditure. These observers point out that many governments took on the onerous responsibility of financing universal medical care at a time when there were relatively few effective medical procedures and care was mainly palliative. With the growing range and sophistication of medical techniques, many of these systems are Special Theme – Health Systems 756 Bulletin of the World Health Organization, 2000, 78 (6) already in difficulties. Constrained by fiscal con- servatism, they will be unable to afford the expected surge of new medical technology. As the services offered by such systems fall behind what consumers expect, governments are likely to come under strong pressure to abandon or reduce sharply their funding role in favour of the private sector. Themore affluent citizens of such countries may increasingly vote with their feet and desert public systems for the private alternative. It should be noted, however, that private insurers can be expected to face problems similar to those confronted by governments. If the expected new technologies are to be paid for, health insurance premiums are bound to rise. To an extent depending on their perception of the benefits of the new procedures, consumers may resist such rises. However, predictions of the breakdown of publicly funded health care systems are probably exaggerated. Governments will be able to choose among a number of policy options for dealing with the anticipated new spending pressures. Few of these options may be attractive in themselves, but theymay be less unattractive than the option of abandoning the goal of providing access to adequate health care for all citizens. Governments may seek to limit the package of basic care or to increase the rationing of care in some other way. Health expenditure may be held at a level perceived as affordable, and rigorous health technol- ogy assessment may be employed so that the general introduction is allowed of only those new technol- ogies that pass a test of cost-effectiveness, applied in such a way as to exhaust the affordable budget but no more. For example, new technologies might be assessed for their potential to add quality-adjusted life years per national currency unit. Technologies with effectiveness/cost ratios exceeding a certain bench- mark would be eligible for general public funding and made widely available, and they would be incorpo- rated into clinical guidelines and protocols. Technol- ogies failing to reach the benchmark might be provided only in specialized centres or for carefully selected patients on an experimental basis. If new technologies were introduced in this way and within a budgetary constraint they would necessarily displace older, less cost-effective ones. Clearly, non-price rationing would tighten for some patients. Governments may increase cost-sharing by patients in the public system. Desirable effects of cost-sharing include the curbing of unnecessary care and the raising of revenue for the system. Undesirable effects include the deterrence of necessary care. If the sick are not to face inequitable financial barriers to access it is necessary to exempt the poor from cost- sharing. That may be administratively cumbersome. Selective charging is also open to the objection that charges leviedonly on themiddle classes penalize those who are already paying disproportionately for health care through their taxes. An alternative is to make charges fairly nominal. However, since cost-sharing incurs additional administrative costs, revenues from nominal charges may not be worth collecting. Governments may succeed in some of their efforts to improve the efficiency of health care systems. Gains in efficiency have the great advantage of lessening any growth in the tension between the demand for and supply of health care finance. Governments may withdraw public insurance cover from households with incomes above a certain level. Germany and the Netherlands already exempt those with incomes above a given ceiling, who are well able to afford private insurance, from compul- sory health insurance schemes. However, such changes would undermine solidarity. The people exempted on grounds of high income would, on average, be those best able to pay taxes in support of the public system. At the same time, as good risks, they would be those least likely to make demands on it. Solidarity is popular in many OECD countries, probably because ofwidespread altruism in respect of access to health care. Many people with high incomes may prefer to live in countries where they help to fund health care for those with low incomes. Social and economic welfare could be reduced rather than increased by the introduction or intensification of two-tier health care. Finally, governments may choose to review their stance on the share of GDP taken by public expenditure (or to review other spending areas) and to raise taxes and health insurance contributions so as to make new procedures affordable within public schemes. Health technology assessment may be applied to the new interventions and it may be found that, for many of them, quality-adjusted life years per national currency unit are greater than the average for existing technologies. This may be a sign that health care spending should be increased so as to accommodate the new procedures. In effect, govern- ments would be acting as consumers would have done if faced by private spending decisions on new products whose perceived benefit/cost ratios ex- ceeded those for existing consumption. Many governments, of course, would want to choose a combination of the above options in order to cope with new spending pressures. Opportunities There are likely to be further improvements in the health status of OECD populations irrespective of the pace of developments in medical care. For example, standards of living and education have been improving for many years in all of the countries concerned. Given that such factors take a long time to affect health it is likely that future generations will be healthier than their predecessors, even if health expenditure were to stabilize. This could delay the demand for care from part of the population but could not eliminate it entirely because of the inevitability of death and the morbidity that typically precedes death. Gains in health status can also be expected to the extent that the new technologies are effective and Challenges for health systems in OECD countries 757Bulletin of the World Health Organization, 2000, 78 (6) can be afforded. Some gains may be achieved through improved public health techniques and some through improved personal health care. Although, initially, new technologies tend to be more expensive than the ones they displace, in the medium to long term their relative prices usually decline and the rate of take-up is enhanced. On the question of affordability there is cautious optimism in some quarters about the prospects for a long period of strong economic growth — a sustainable ‘long boom’ in the world economy (18). The reasons for this include expected improvements in productivity associated with devel- opments in computing and the Internet, and further cumulative effects on world trade of earlier reduc- tions in tariff barriers and non-tariff barriers. Nevertheless, such optimism could be mis- placed. There is no guarantee that future growth rates will be higher and more sustained than those of the past two decades. If growth rates do not increase the affordability of health care systemswill continue to be a major concern. There should be opportunities, however, to improve the efficiency of publicly funded health services. Thus, with regard to incentives for efficiency, governments that continue to be heavily involved in the provision, as opposed to the financing, of services, may wish to review their stance. Whereas the arguments in favour of govern- ment involvement in the financing of health care are strong, those in favour of government provision of health care are more debatable. One of the risks of government involvement on the supply side aswell as on the demand side is that it can create a conflict of interest, i.e. provider capture, which can undermine the role of government as purchaser. Public health insurers who contract with independent providers may obtain better results for patients than those who produce health care themselves. As far as improving information for efficiency is concerned, mention has already been made of the key role that health technology assessmentmight play in guiding decisions on the deployment of new medical technologies. Since health technology as- sessment is costly andmuch of its scientific content is transferable across countries, this is an obvious area for greater international cooperation. In addition, there is much scope for better monitoring of performance and outcomes. Several OECD coun- tries are already developing performance-monitoring systems for their health systems. TheOECD is giving priority in its health policy work to comparisons of health system performance between its Member Countries. This work is backed up by the WHO performance framework presented elsewhere in this issue of the Bulletin. Further opportunities can be expected for pursuing efficiency gains through the development of the public and private versions of managed care. Caution is required in order tominimize adverse side- effects such as higher administrative costs. Success with managed care may require further, sometimes painful, adjustments both to consumer expectations and to professional habits associated with self- regulation. Conclusion Although the problems faced by health systems in OECD countries are less severe than those encoun- tered elsewhere, they continue to present challenges. For reasons of equity, most OECD countries have chosen to base their funding of health care mainly on public sources. There is a widespread problem of affordability, arising from the tension between the willingness of the population to pay taxes and the eagerness of patients to use health services if these are free or heavily subsidized at the point of use. These tensions are likely to be exacerbated by a surge of new medical technologies adding to demands for health care. Some observers have predicted the breakdown of publicly funded systems of health care under new spending pressures. However, governments can deploy a range of policies for handling new demands. They can also take comfort from the fact thatmany of themhave already coped with successive waves of technological change in health care without abandoning their core commitment to the public funding of health systems. Furthermore, if standards of living continue to rise, public and private insurers should find it easier to obtain the revenues needed to pay for the improved health care expected by consumers. n Acknowledgements The author is grateful for statistical assistance from Andrew Devlin and for advice from John Martin, Peter Scherer, Manfred Huber, Ste´phane Jacobzone, Melissa Jee, Gaetan Lafortune and Jan Bennett. Re´sume´ Enjeux pour les syste`mes de sante´ des Etats Membres de l’Organisation de Coope´ration et de De´veloppement e´conomiques Les difficulte´s auxquelles se heurtent les syste`mes de soins de sante´ des pays de l’OCDE sont moins graves que celles des autres pays. Les 29 Etats Membres de l’OCDE sont des pays industrialise´s ou` le revenu par habitant est e´leve´ ou moyen. Leurs populations ont un niveau ge´ne´ral de sante´ relativement e´leve´, qui continue de s’ame´liorer. Leurs e´conomies, pour la plupart d’entre elles, sont en mesure de supporter des de´penses de sante´ d’un niveau suffisant, voire plus que suffisant. La majorite´ des pays de l’OCDE ont en outre organise´ le financement de leur syste`me de soins de sante´ de telle sorte que les personnes en bonne sante´ subventionnent les malades, que les Special Theme – Health Systems 758 Bulletin of the World Health Organization, 2000, 78 (6) jeunes subventionnent les personnes aˆge´es et que les riches subventionnent les pauvres. Il n’en subsiste pas moins de nombreuses difficulte´s. Du coˆte´ de la demande, les personnes aˆge´es tendent a` tomber malades apre`s avoir atteint le stade de la vieillesse. Une proportion importante des de´penses de sante´ est consacre´e aux malades dans les deux anne´es pre´ce´dant leur mort. La richesse peut en outre favoriser des habitudes malsaines et faire peser des menaces sur la sante´, notamment par ses effets secondaires inde´sirables sur l’environnement. Certains groupes, dans les pays de l’OCDE, comme les choˆmeurs et les pauvres, ne profitent pas de l’ensemble des progre`s sanitaires dont be´ne´ficie la majorite´, ou du moins pas dans la meˆme mesure. Entre- temps, l’ame´lioration du niveau de vie, le de´ferlement des nouvelles technologies me´dicales et le de´veloppe- ment de la socie´te´ de consommation suscitent des attentes croissantes dans la majorite´ au plan de la protection de la sante´ et de l’alle´gement de la charge de morbidite´ restante. Du coˆte´ de l’offre, la croissance e´conomique et les progre`s technologiques tendent a` accroıˆtre le couˆt relatif de la protection de la sante´ publique et de l’acce`s aux soins me´dicaux individuels et aux dernie`res avance´es me´dicales. Les gouvernements de la plupart des pays de l’OCDE, principalement par souci d’e´quite´, sont de´sormais e´troitement associe´s a` la re´glementation, au financement et parfois a` la fourniture des soins me´dicaux. C’est donc surtout au secteur public qu’e´choit la difficile taˆche de concilier l’accroissement de la demande et la hausse des couˆts. Il s’ensuit que dans la plupart, sinon dans la totalite´, des pays de l’OCDE, c’est au gouvernement qu’incombe la lourde responsabilite´ d’assurer, ou de maintenir, la bonne performance du syste`me de sante´. De nombreuses re´formes ont e´te´ motive´es par cet objectif. Certaines d’entre elles, par exemple dans le domaine de l’endiguement des de´penses, ont re´ussi, mais d’autres ont eu des conse´quences qui ont de´c¸u leurs initiateurs. Certains observateurs ont pre´dit la faillite des syste`mes de soins de sante´ publics sous l’effet des nouvelles de´penses auxquelles ne peut se soustraire l’Etat. Face aux nouvelles demandes, les gouvernements peuvent toutefois recourir a` tout un e´ventail de politiques. C’est ainsi que les syste`mes publics d’assu- rance-maladie qui passent des contrats avec des prestateurs prive´s, ou avec des prestateurs publics aux conditions du marche´, peuvent obtenir de meilleurs re´sultats pour les malades que les syste`mes publics d’assurance-maladie qui fournissent eux-meˆmes les soins de sante´. Constat rassurant pour les gouvernements, un grand nombre d’entre eux ont de´ja` fait face a` plusieurs vagues successives de progre`s technologiques dans le domaine des soins de sante´, sans pour autant renier leur engagement fondamental de financer les syste`mes de sante´. De plus, si le niveau de vie continue de s’e´lever, les syste`mes publics et prive´s d’assurance-maladie recueil- leront d’autant plus facilement les recettes ne´cessaires au financement des soins de sante´ ame´liore´s qu’atten- dent les usagers. Resumen Retos para los sistemas de salud de los Estados Miembros de la Organizacio´n de Cooperacio´n y Desarrollo Econo´micos Las dificultades con que tropiezan los sistemas de atencio´n sanitaria en los paı´ses de la OCDE son menos graves que en otras partes del mundo. Los 29 Estados Miembros de la OCDE son paı´ses industrializados con ingresos per ca´pita altos o medianos. Sus poblaciones gozan, como promedio, de un nivel de salud relativa- mente elevado y en aumento. La mayor parte de sus economı´as efectu´an gastos de salud suficientes o ma´s que suficientes. Por otra parte, la mayor parte de los paı´ses de la OCDE han organizado la financiacio´n de sus respectivos sistemas de atencio´n sanitaria de tal manera que los sanos financian a los enfermos, los jo´venes a los ancianos, y los ricos a los pobres. No obstante, au´n quedan muchos problemas por resolver. Con respecto a la demanda, quienes sobreviven hasta una edad avanzada tienden a enfermarse con los an˜os. Una elevada proporcio´n de los gastos de salud totales se destina a pacientes a quienes les quedan menos de dos an˜os de vida. Adema´s, la abundancia puede alentar ha´bitos malsanos y crear amenazas para la salud, entre ellas efectos secundarios indeseables en el medio ambiente. Algunos grupos de poblacio´n de los paı´ses de la OCDE, como los desempleados y los pobres, tienden a quedar excluidos de los progresos sanitarios generales de los que goza la mayorı´a, o bien a no beneficiarse en la misma medida. Entretanto, en lo que atan˜e a la mayorı´a, el aumento del nivel de vida, una sucesio´n creciente de nuevas tecnologı´as me´dicas y el consumismo en aumento elevan las expectativas acerca de lo que se puede y se debe hacer para mejorar la proteccio´n de la salud y aliviar la carga restante de morbilidad. En lo concerniente a la oferta, el crecimiento econo´mico y los cambios tecnolo´gicos tienden a aumentar los costos relativos de proteger la salud pu´blica y ofrecer a la poblacio´n una atencio´n me´dica personal y los u´ltimos adelantos de la medicina. Principalmente por razones de equidad, los gobiernos de la mayor parte de los paı´ses de la OCDE han pasado a participar intensamente en la reglamentacio´n, la financiacio´n y a veces la prestacio´n de la atencio´n me´dica. Por consiguiente, el problema de conciliar el aumento de la demanda con el de los costos afecta principalmente al sector pu´blico. De ello se desprende que la mayor parte, si no la totalidad, de los gobiernos de la OCDE tienen la gran responsabilidad de conseguir o seguir consiguiendo un buen desempen˜o de sus sistemas de salud. Se han introducido muchas reformas para alcanzar este objetivo. Algunas de ellas han dado Challenges for health systems in OECD countries 759Bulletin of the World Health Organization, 2000, 78 (6) buenos resultados, por ejemplo en cuanto a la reduccio´n de los costos, pero otras han tenido efectos colaterales adversos y han desalentado a sus promotores. Algunos observadores han predicho el quebranta- miento de los sistemas de salud financiados con fondos pu´blicos como resultado de las nuevas presiones en materia de gastos. Sin embargo, los gobiernos pueden aplicar una variedad de polı´ticas para atender las nuevas demandas. Por ejemplo, las entidades pu´blicas de seguro me´dico que contratan a prestadores de servicios independientes, o a prestadores pu´blicos en condiciones de libre competencia, pueden obtener mejores resulta- dos para los pacientes que las entidades pu´blicas de seguro me´dico que prestan ellas mismas atencio´n sanitaria. Los gobiernos pueden considerar reconfortante el hecho de que muchos de ellos han afrontado ya olas sucesivas de cambios tecnolo´gicos en el a´mbito de la asistencia sanitaria sin abandonar su compromiso central con la financiacio´n pu´blica de los sistemas de salud. Por otra parte, si sigue aumentando el nivel de vida, deberı´a resultar ma´s fa´cil para los aseguradores pu´blicos y privados obtener los ingresos necesarios para pagar la mejor asistencia sanitaria que esperan los consumidores. References 1. Frenk J, Murray C. A WHO framework for health system performance assessment. Bulletin of the World Health Organization, 2000, 78: 717–731. 2. Jacobzone S et al. Is the health of older persons in OECD countries improving fast enough to compensate for population ageing? OECD Economic Studies, 2000, 30: 149–190. 3. Or Z. Out-performing or under-performing: an analysis of health outcomes in France and other OECD countries. Health and System Science, 1997, 1: 321–344. 4. Or Z. Determinants of health outcomes in industrialised countries: a pooled, cross-country, time-series analysis. OECD Economic Studies, 2000, 30: 53–78. 5. Mossialos E. Citizen’s view on health care systems in the 15 Member States of the European Union. Health Economics, 1997, 6: 109–116. 6. Hurst J. What causes variations in the performance of health care systems? Paris, OECD (Labour Market and Social Policy, Occasional Paper, forthcoming). 7. The European study of referrals from primary to secondary care. London, Royal College of General Practitioners, 1992 (Report to the Concerted Action Committee of Health Services Research for the European Community, Occasional Paper No. 56). 8. Wagstaff A et al. Equity in the finance of health care: some further international comparisons. Journal of Health Economics, 1999, 18: 263–290. 9. Van Doorslaer E, Wagstaff A, Rutten F. Equity in the finance and delivery of health care, an international perspective. Oxford, Oxford University Press, 1993. 10. Cichon M. Modelling in health care finance. Geneva, Interna- tional Labour Office, 1999. 11. OECD. Health data 99. Paris, OECD, 1999. 12. Marre´e J, Groenewegen P. Back to Bismarck: Eastern European health care systems in transition. Aldershot, Acebury, 1997. 13. The reform of health care, a comparative analysis of seven OECD countries. Paris, OECD, 1992. 14. The reform of health care systems, a review of seventeen OECD countries. Paris, OECD, 1994. 15. Huber M. Health expenditure trends in OECD countries, 1970–1997: an update on data and methodology. Health Care Financing Review, 1999, 21 (2): 99–118. 16. Arber W, Brauchbar M. Biotechnology for the 21st century. In: 21st century technologies. Paris, OECD, 1998. 17. Jacobzone S. How can pharmaceutical policy reconcile social objectives and industrial efficiency? A view from the recent experience of OECD countries. Paris, OECD, 2000 (Labour Market and Social Policy, Occasional Paper No. 40). 18. The future of the global economy: towards a long boom? Paris, OECD, 1999. Special Theme – Health Systems 760 Bulletin of the World Health Organization, 2000, 78 (6)