Printer Friendly

What can Americans learn from Europeans?

What can Americans learn from Europeans?


The United States, with its great plurality of organizational, ownership, and financing forms in health care, has been described as a laboratory for experimentation in health service organization and financing. The variety of institutions and financing mechanisms is enormous, and it has, over time, produced many interesting experiments that have been of great interest to European countries looking for models to reform their health care systems. During the last decade, concepts such as HMOs (health maintenance organizations), DRGs (diagnosis-related groups), and MTA (medical technology assessment) have become part of the standard vocabulary in most European countries.

It is therefore a somewhat strange situation, for an economist who has devoted a significant amount of time and effort to looking for opportunities to learn from the U.S. health care scene, to reverse his perspective and look for lessons that Americans can learn from European systems. Because there is so much variety in U.S. health care, it is also difficult to find new ideas and approaches that have not already been tried somewhere at some time in the United States. There is also the risk of being irrelevant. Means cannot be judged without reference to the goals. It is obvious that the moral bases or values on which health care systems are based differ between the United States and European countries. In the United States, personal responsibility, freedom of choice, and pluralism are the major moral commitments. In European countries, goals related to population health and equality of access to health services have been relatively more important. Centralized health planning and a large governmental role in health care financing have therefore been more generally accepted in Europe.

As has been pointed out by others (e.g., Culyer, Maynard, and Williams, 1981), it is unhelpful to seek to learn from a system that is seeking to accomplish different aims. However, this should not be overstated: After all, a means may serve more than one end or may be adopted more appropriately to serve another end. Therefore, even though the ideological bases may vary between countries, it does not follow that neither has anything to learn from the other. So without appearing presumptuous, I hope to identify some areas or issues in which I think the experience of European countries can be of value in the development of U.S. health policy. I concentrate on issues that I can substantiate with data from the Organization for Economic Cooperation and Development (OECD) data base on international differences in health care (Organization for Economic Cooperation and Development, 1989).

Common aspects of European systems

Europe, as well as the United States, shows a great variety in health care systems and consequently also in health care policies. In addition to learning from its own diversity, the United States can learn from the specifics of different European systems at different points in time. There are also many studies in which U.S. researchers have looked at particular types of policies, such as physician reimbursement and cost containment (Reinhardt, 1981) or have focused on individual countries. For Sweden alone, it is easy to find more than a dozen publications in which U.S. researchers have looked at various aspects of the Swedish health care system and made comparisons with the United States (e.g., Lembcke, 1959; Anderson, 1972; Navarro, 1974; and Rosenthal, 1986 and 1987).

In this article, I concentrate on a number of aspects that are common to most, if not all, European health care systems. There are many from which to choose. The most obvious to start with is the financing and delivery of health care to the elderly. The populations of Europe are significantly older than that of the United States; therefore, Europe can give interesting evidence of what is to come in the United States. As shown in Table 1, the proportion of the population 65 years of age or over is 13.8 percent in Europe, compared with 12.2 percent in the United States. In some countries, such as Sweden, it is more than 17 percent.

An increasing share of health care resources spent on the elderly dramatically alters the conditions for health care financing and delivery. The insurance function becomes more of an intertemporal allocation problem, similar to the pension system, than a traditional health insurance. In the Federal Republic of Germany (hereafter called Germany), the public pension funds transfer money to the sickness funds to subsidize health care expenses for retired members. In 1975, these transfers accounted for 27 percent of total revenues for sick funds (Henke, 1980.) An increasing amount of total health expenditures is spent on the last years of life, and almost everyone reaches the average life expectancy (Le Grand and Rabin, 1986).(1) For the provision of health care, this means that nursing home care becomes relatively more important, as does the integration of social and medical services. In order to study health care for the elderly in different countries, age-specific cost and utilization data are needed. These data are not yet available, so this topic will have to wait for further refinements of the international comparative health statistics.

Another aspect of great interest is the quality of care. Systems of quality control are different in Europe from those in the United States. The use of litigation to deal with medical malpractice is considerably more common in the United States than in Europe. This affects health care costs directly, through insurance premiums paid by doctors, and indirectly, through its effect on physician behavior--the practice of defensive medicine. It would be of great interest to know more about the consequences for quality of care of the different ways of practicing medicine. In addition, it would be helpful to look at measures of quality assurance in the United States are Europe. However, international comparative statistic provide only scant information for such exercises.

Instead, I concentrate on three general aspects of European systems for which there are existing international statistics: the comprehensiveness of these systems; the role of regionalization in achieving cost control, efficiency, and equity; and the management of new expensive medical technologies.

A comprehensive health care system

The most striking difference between European systems, taken as a whole, and that in the United States is the comprehensiveness of the European systems. In most European countries, everyone is eligible for coverage of medical expenses through a public plan--usually public both in the sense of finance and in the sense of provision. This is the case for hospital care, ambulatory care, and medical goods, though the element of copayment is usually higher for the latter two categories. Even in countries with a significant amount of private insurance, such as Switzerland and the Netherlands, the share of the population eligible for coverage through public plans is close to 100 percent for hospital care (Table 2). Those not covered are mainly the more affluent members of the population, who may choose not to participate in the public plans.

In Switzerland, three partners share the expenditures for the health care system. In 1984, public funds accounted for 6.2 billion Swiss francs, sickness funds and social insurances for 5.1 billion francs, and privately insured and uninsured persons for 5.2 billion francs (Gygi and Frei, 1986). In the Netherlands, about 70 percent of the population is insured with the sickness funds, which operate the social insurance system of the Sickness Funds Insurance Act. Individuals with an income of more than a certain amount (approximately 43,000 Dutch guilders in 1982) have to acquire private health insurance, but employees with an annual income under that level are mandatorily insured by sickness fund insurance. Private financing accounted for about 25 percent of total health expenditures in 1980. Insurance for greater risks, i.e., exceptional expenses, is governed by the General Special Sickness Expenses Act. This is a national insurance plan applying in principal to all residents of the Netherlands (Rutten, 1982). At present, the health care system in the Netherlands is in a period of transition. In October of 1988, the Dutch parliament approved the implementation of the first steps in the new direction. Two major issues are the introduction of national health insurance ("basic insurance") and regulated competition among insurers and among providers. For a review of the recent developments, see van de Ven (1989).

In the United States, the coverage rate of public plans is estimated to be only 40 percent of the population for hospital care and 25 percent for ambulatory care. Even if a higher degree of private insurance, subsidized through tax exemptions for employers or individuals, partly compensates for this, a significant part of the population lacks coverage or is inadequately covered for health care. It has been estimated (Davis, 1989) that as many as 37 million people, 15 percent of the population, lack adequate insurance coverage; many lack coverage of any kind and are dependent on private charity. However, other sources show that 14.5 percent of the poor, 15.6 percent of the near-poor, 11.9 percent of other-low-income people, and 4.5 percent of all others lack insurance coverage (Kasper, 1986). This is a total of 17 million people, or 7 percent of the population, who are uninsured.

The coverage rate is only one aspect of the comprehensiveness of a system. We also have to look at what benefits are provided. In Table 3, the average or typical percent of a bill paid by public insurance is shown for several European countries and the United States. It can also be seen that the copayment rate is lower in European countries than in the United States.

The advantages as well as the problems associated with a system that provides insurance coverage for everyone probably provide the most significant lessons for the United States.

Differences in total health expenditures

A higher coverage rate and a greater share of public financing in Europe may be thought to imply significantly higher total expenditures. However, this is not the case. On the contrary, the health expenditures share of the gross domestic product (GDP) is significantly higher for the United States than it is for any European country (Table 4). (Gross domestic product is the gross market value of the goods and services attributable to labor and property located in a given nation.)

In 1986, total health expenditures, as a percent of GDP, were 10.9 percent in the United States, compared with a European average of 7.2 percent. Sweden, the European country with the highest share, had 9.1 percent. Also in absolute terms, using exchange rates or purchasing power parities (PPPs) for conversion of national currencies, the expenditures in the United States are significantly higher than in any European country.

Does this mean that the level of expenditures in the United States is totally inconsistent with the experiences of the European countries? To answer this question, we must look more closely at the determinants of health care expenditures. The major determinant of such differences is GDP per capita. (See Gerdtham et al., 1988, for a review and some new estimates.) Taking into account that GDP per capita is higher in the United States than in Europe, are the expenditures still higher than would be expected?

In Table 5, the actual and predicted 1986 health expenditures for 18 European countries and the United States are shown. A logarithmic function has been used (Gerdtham et al., 1988). The regression equation is: HEXP = -5.99 + 1.36 GDP ([R.sup.2] = 0.89) t-ratio (-5.6) (11.9) The constant elasticity is 1.36, which means that a 1-percent increase in GDP will give a 1.36-percent increase in total health expenditures. Both the elasticity and the constant are strongly significant. As can be seen in Table 5, the actual health expenditures for the United States are 28 percent higher than predicted. This is close to the upper bound of a 95-percent confidence interval. (The comparability of health care expenditures data among countries is, despite significant improvements, still not exact. The lower-than-predicted levels of expenditures for Denmark, Luxembourg, and Norway can be explained partly by underreporting of certain types of expenditures, mainly nursing home care.)

The expenditure levels of 18 European countries were used as a basis to predict health expenditures for the United States (Table 6). The result is very similar for the logarithmic specification of the regression function, although the elasticity is slightly reduced. For the linear specification, the prediction is lower and the level of actual expenditures is significantly outside the 95-percent confidence interval.

We can conclude that the higher share of public financing in European countries does not result in higher expenditures than in the United States. In fact, there is evidence to the contrary.(2) Details on the cost-containment policies used in Europe are covered in the article by Culyer in this issue; therefore, they are not discussed here. Instead, let us look at other aspects of health policy that relate to the comprehensive nature of European systems. However, it is necessary to relate these aspects to expenditure restraint, because this is the major mechanism by which resources are allocated in European systems.

Preventive versus curative services

The most important question for any health care system is how well it is achieving the basic health objectives of the population. The difficulties in comparing the objectives of health care systems are well known. Health is difficult to measure and is determined to a major extent by factors outside the control of the health care system. However, one area in which health care can make a directly measurable difference is in the reduction of infant mortality, although it is difficult to document exactly the relative contribution of various factors.

The World Health Organization (1981) states that the infant mortality rate "is a useful indicator of the health status not only of infants but also of whole populations and of the socioeconomic conditions under which they live. In addition, the infant mortality rate is a sensitive indicator of the availability, utilization, and effectiveness of health care, particularly perinatal care." Waaler and Sterky (1984), examining trends in infant mortality, perinatal mortality, and gross national product in four Scandinavian countries, suggest that perinatal mortality (late fetal and neonatal deaths per 100 live and still births) is preferable to infant mortality (death rates of infants under 1 year of age per 100 live births) as an indicator of the quality of heath care. However, examining the relationship between infant mortality and perinatal mortality in 1986 and changes from 1960 through 1986 in European countries and the United States gives no support for this hypothesis. The differences between countries and over time are very similar with both measures (Table 7). As can be seen in Table 7, several European countries have an infant mortality rate that is 30-40 percent lower than that of the United States. Smaller countries can be expected to have lower rates because of greater cultural homogeneity, but it may be observed that the rates are also lower for the big European countries. Taking into account the high GDP per capita and the high rate of health expenditures, one would expect the United States to have significantly lower infant and perinatal mortality rates than the Federal Republic of Germany, France, and the United Kingdom. To the extent that the differences between Europe and the United States can be attributed to medical action, the lower mortality rates must have been achieved mainly by preventive measures. The resources for neonatal intensive care and the frequency of cesarian delivery are higher in the United States, but this obviously does not result in a lower infant mortality rate.

My hypothesis is that it is easier to allocate resources to programs like prenatal care and vaccination for children in a more comprehensive system. Considering the relative efficiency of health care services, it is difficult to allocate resources to interventions with dubious value, when even basic services with proven or obvious cost effectiveness are not being provided. The underprovision of such inexpensive but cost-effective services is a serious inefficiency in a health care system.

The commitment to primary health care and prevention in many European countries must be considered in the shaping of the future U.S. health care system. But this policy has not been without its problems, and, in many instances, it has been only a verbal commitment. In other instances, it has resulted in overoptimistic expectations about its ability to solve health problems or even to control escalating health care costs. But there is much to be learned from the mistakes as well as the successes. Focusing on health problems from a population perspective, rather than an individual perspective, gives new insights into the relative efficiency of different interventions. It also indicates that more health services are not always a solution to social problems; rather the solutions to social problems can have a significant impact on health. A comprehensive system enforces this broader perspective.

Regionalization, global budgets, and planning

One important common aspect of European health care systems is regionalization. Typically, regions, rather than the country as a whole, are the basis for the allocation of health resources. However, regionalization is achieved in different ways in each country. In Sweden, for example, the regions (counties) have total responsibility for both the health of the population and the provision of health care. The average population of a region is 350,000 inhabitants, and the decisions about health services provision are made by elected representatives (community councils) in the region. The financing of services comes from a local proportional income tax.

Similar systems can be found in the other Scandinavian countries of Denmark, Norway, and Finland. However, in Finland, the local communities are the basic regional unit, and in Norway, the regions (fylke) are responsible for hospital care and the local communities for primary care. In Denmark, the responsibility for hospital services, general practitioners, and practicing specialists is decentralized to 16 regions (counties or "amt"), the typical population being 250,000-300,000 inhabitants. Local health and social services, including home care, are run by communities of varying size. The Copenhagen and Frederiksberg communities, in the Copenhagen metropolitan area, have community as well as county obligations.

In contrast, in the United Kingdom, regional authorities are not elected, nor do they have tax powers, nor do the districts beneath them, which are responsible for the provision of hospital care to the locality. Instead, their finances are directly controlled by cash-limited budgets allocated from the center. Since 1977, the allocation of resources among regions has been based on the Resource Allocation Working Party (RAWP) formula, with the objective of securing geographical equity in the availability of resources (Department of Health and Social Security, 1976). Different indicators of need determine the per capita-based funding for the regions. RAWP has been controversial, and there is an extensive literature of criticism and comment (Mays and Bevan, 1987). But after 10 years, the gap has narrowed substantially between the regions receiving the most and the least funds.

Regionalization is also strong in insurance-based systems such as those in Germany, the Netherlands, and Switzerland. In the last country, health care is by law a responsibility for the regions (cantons), some of them very small. Although the majority of financing comes from private and public insurance (Krankenkassen), it is the canton that has the overall responsibility for the provision of health care resources. The cantons also have to underwrite the deficit in public hospitals operated by them. This gives them a strong influence over the allocation of resources.

In the Netherlands, hospitals, whether owned by local communities or lay boards of trustees, operate on a fixed predetermined budget since 1984. The budget is negotiated with the third-party payers. The central government has a strong influence on construction of facilities and acquisition of major medical equipment through licenses, which are issued on the basis of regional and national health sector planning.

The strict planning systems for hospital care in Switzerland and the Netherlands, the two European systems most similar to those of the United States, are the major reason why expenditures are constrained in these countries, and the share of GDP that goes to health care is kept below that of the United States. In practice, these countries have global budgets for the most expensive part of health care, hospital services.

In Germany, the hospitals are financed through prospective, hospital-specific, all-inclusive per diems negotiated between the hospital and regional associations of sickness funds. These rates are subject to approval by the State governments, which also must approve and finance capital investments, based on statewide hospital planning. The State governments therefore control the capacity of the hospital system.

In Figure 1, a first attempt to develop a way to classify regionalized health care systems is shown. The first question to be asked is whether regionalization is based on a federal structure of the country as in Canada, Switzerland, and Germany. This aspect is of particular relevance for the United States, being a federal country. In federal countries, the constitution usually states the division of responsibilities in health care between the federation and the individual States. The constitution determines the possibilities and limitations for regionalization. In Switzerland, the constitution clearly states that the cantons forming the federation are responsible for health care within their borders.

The second aspect of regionalization concerns the size of the population in the region, because this determines how regionalization works in practice. Very large regions must be divided into smaller regions in order to find a suitable size to manage health care institutions. An example of this is the Regional Health Authorities in the United Kingdom, which are divided into districts that provide health services. Very small regions (such as some cantons in Switzerland) must rely heavily on cooperation with other regions for provision of specialized services.

If the region is governed by elected representatives, there is a political process behind the allocation of resources. This is, of course, the case in the federal countries, but also in the Scandinavian countries. However, in the United Kingdom, the members of the RHAs and District Health Authorities (DHAs) are appointed by the government. A system with elected representatives is usually combined with the right to levy regional taxes. But elected representatives and the power to tax do not always go together. In Norway, the regions (fylke) receive 50 percent of their budgets from the central government; once they reach the cap on their own tax power, they cannot spend more than the amount set by the government. In Sweden and Denmark, the regions (landsting and amt, respectively), can finance health care through a proportional income tax.

A region can be both a purchaser and a provider of care. Usually the two functions go together. However, in Germany, the states finance capital costs, but do not themselves provide any services. In Sweden, Denmark, and Switzerland, the regions are also the main providers of hospital services, but there is also a "market" for services, mainly tertiary care, between regions. In the United Kingdom, there is discussion about separating the regions' roles of purchaser and provider. Similar discussion is found in Denmark and Sweden.

In most regionalized systems, the region operates a global budget for health services. Although this is not the case in, for example, Germany, there have been attempts to form a voluntary agreement, negotiated between the different interest groups, to contain costs through a global budget. For a description of the German Health Care Cost Containment Act of 1977, see Stone (1979). In Sweden, the global budget includes all health expenditures except drugs and dental care, but in other countries, such as Switzerland and the Netherlands, it includes only hospital expenditures.

For many regions, health care is their main responsibility. In other instances, they have a wider responsibility for public services and transfer payments. This obviously has implications for the tradeoff between health care and other public expenditures. The percentage of the total budget that comes from central funds can also have significant implications for the relationship between central and regional governments and for total spending on health care, as well as for the allocation of resources to different services. Even in countries in which regions have the power to tax, the central government exercises influence through different types of government grants. In Sweden, the tax equalization plan and transfers from the social insurance system to the counties are of great financial importance for the regional budget.

Another significant aspect of regionalization is the distribution of regulatory power between the central government and the regions. This is done very differently in different countries. Usually the central government regulates fees and capital investments. However, regions can also have important regulatory power. In Sweden, the county councils regulate how many private practitioners are allowed to practice under the health insurance plan and how many patient visits they can have during a year. In Germany, there are regional negotiations of reimbursement rates.

Regionalization is one way of controlling total health care expenditures as well as guaranteeing everyone access to basic and effective health services. Public choice theory tells us that, the larger the population served by a given budget, the more difficult it is to manage that budget. Rent-seeking from different interest groups becomes impossible to handle when the budget serves a country as large as the United States. It is impossible to make rational decisions about allocation of resources based on knowledge of local health needs.

In Europe, it is only the United Kingdom that has for a long time operated a global budget for the whole country. But there the total budget is allocated to the Regional Health Authorities, according to a weighted population formula, and it is the RHAs that determine the allocations within these geographical areas. This global cost containment has obviously been successful, if we look at total health care expenditures and value for money. However, we must not forget that a major reason behind the lower share of GDP allocated to health care in the United Kingdom, compared with the United States, is the significantly lower GDP per capita.

The major advantage of regionalization is that it provides a forum for discussion about priorities in health care and value for money for different services. It also gives an opportunity to identify the total resources for health care spent on a population and to assess the appropriateness of these expenditures in relation to population needs. Because the need (defined as services with a positive marginal product on health) is for all practical purposes unlimited, a tradeoff between health care and other goods and services has to be made. Regionalization is a way to make this tradeoff more transparent and responsive to local needs. The region also forms the basic unit for planning of facilities and mobilization of resources. It is also an important mechanism for ensuring geographical equity in the availability of services.

Regionalization, freedom of choice, and competition

I have introduced concepts such as regions, priorities, value for money, planning, and global budgets in relation to population needs. Does this not amount to socialized medicine? Without very precisely defining what "socialized medicine" is, I will argue that this is not the case. My main argument is that regionalization can take many different forms and is consistent with different degrees of freedom of choice and competition among providers. The wide variety of European health care systems, all more or less based on the concept of regionalization, shows that regionalization can be combined with different financing systems, different reimbursement systems, different mixes of private and public providers, and different regulatory mechanisms.

A comparison between Sweden and Switzerland can illustrate the different forms that regionalization can take. In Switzerland, ambulatory health care is provided by private practitioners working on a fee-for-service basis. The costs are reimbursed by private or public insurers. The copayment is very small and almost everyone is covered. Physicians can locate where they wish, but there is an incentive plan to stimulate location in remote (mountain) areas. The majority of the hospitals are owned and operated by the cantons. There are also private and community-owned hospitals. The hospitals are reimbursed from the insurance plans on a per diem basis. The public hospitals, which constitute the majority, run deficits that are covered by regional taxes. This provides the raison d'etre for a strict planning and budgeting system for hospital services. Doctors in hospitals are salaried but can take private patients both as inpatients and outpatients, and share the revenue for these patients with the hospital. Patients can choose their doctors as well as the hospitals where they are treated.

In Sweden, the county councils are responsible for all health care for a defined population--a county or region. Ambulatory care is delivered in public health centers as well as in outpatient clinics at the hospitals. The number of private practitioners is small, and their activities are highly regulated, both in terms of the number of patients they can treat and the reimbursement their patients can receive from health insurance. There is a small market for private care outside the public reimbursement system. The choices are limited for the patients, who are assigned a specific health center and a hospital and seldom have any influence over which doctor treats them.

The Swiss experience tells us that an insurance-based system with both private and public insurance and total coverage of the population can be combined with a strict planning system for hospital care at the regional level. The strict planning system is a prerequisite for controlling the health care budget funded by the canton. In terms of cost containment, it has been very successful. In dollars (PPP), Sweden and Switzerland spend almost exactly the same amount on health care (Table 3), which is far less than the United States spends, both in absolute terms and in relation to GDP.

The Swedish experience shows that a regional monopoly can be successful in creating access for everyone to health services and in containing costs. The quality of care is also judged to be high. However, there are problems with that type of organization that should not be overlooked and from which we can learn. First, among the people, there is dissatisfaction with not being able to make any choices in health care. But this lack of choice is not inherent in either regionalization or regulation. It can be changed in different ways within the system. Saltman and von Otter (1987) have suggested that opportunities for consumers to choose which health care center to use, which doctor to see, and which hospital to use should be increased. This will, however, have very little impact on efficiency unless the choices have financial consequences for the providers. We still do not know if the budget system can be made flexible enough to accommodate such choices. However, the experiences in the United Kingdom show that strict regulation can be combined with freedom for patients to choose a doctor and hospital. In the United Kingdom, the patient is free to select any general practitioner (GP) and the GP is likewise free to accept the patient or not. GPs are also free to refer patients to consultants of their choice, including those in hospitals outside the region. The GP's income comes partly from capitation and partly from fee for service. In contrast to Sweden, the global regional budget does not include ambulatory care; therefore, the Swedish system can be seen as financially more tightly controlled than the system of the United Kingdom.

Second, there is a problem with incentives for providers working on a fixed budget without competition. One way to improve productivity and efficiency among providers is to separate financing from provision of services in order to create competition. In Sweden, it has been suggested that the purchase of health care should be transferred from the county councils to the local communities (Jonsson and Rehnberg, 1987). The county councils should continue to be primarily providers of hospital services to the local communities. The ideas behind such a division of health care services relate closely to the discussion and research on HMOs and managed competition in the United States during this decade (Enthoven, 1980). In this issue, Culyer's article presents similar ideas for improving internal efficiency in the National Health Service (NHS). The advantage of separating financing from delivery is mainly that public providers must compete with each other and with private providers. One can, of course, question whether local communities can efficiently carry out the two roles as sponsors (Enthoven, 1988) and financers of health care, but it is a first step in creating managed competition in a regionalized system such as those in the United Kingdom and Sweden.

Today, in both Sweden and Switzerland, there is a well-functioning market for specialized (tertiary) care, in which one region buys services from providers in other regions. This gives smaller regions access to high-quality specialized care. For some services, they are buyers, and for others, they are sellers. This division of tasks is created with planning and competition, the latter increasing in importance over time. To an increasing extent, the payments are based on (prospective) prices that have been negotiated using a DRG-type framework. There is no reason why this market could not also be extended to referrals, not only between hospitals, but also between primary care centers and hospitals. Discussion about competition as a way to improve health services has taken place primarily in the United States. However, the reforms being implemented in Europe will provide not only interesting information about the problems of and opportunities for managed competition, but also examples for decentralized experiments in the United States.

Introduction of new medical technology

One important aspect of a health care system is its ability to control the introduction and diffusion of new, and often expensive, medical technologies. The European countries provide good examples of how new medical technology is managed in a comprehensive health care system based on regionalization. Because systems differ, the pattern of introduction and diffusion differs among technologies and countries, providing an interesting variety for study and analysis. Because information about new medical technology is more or less simultaneously available in all industrialized countries, the same opportunities for adoption and diffusion exist in all countries. Differences in rates of adoption and levels of diffusion can be explained primarily by characteristics of the national health care systems, including availability of resources and cultural patterns. The European countries therefore provide a laboratory in which the United States can observe the management of new medical technology.

The introduction of dialysis and transplantation for treatment of end stage renal disease (ESRD) in the 1960s can serve as an example of how the introduction and diffusion of a new technology in managed in European countries compared with the United States. In 1970, the number of treated patients per 1 million of population was about 25 in the United States, which was only one-half the number treated in Denmark, Sweden, and Switzerland, and about the same as in the United Kingdom. Two years later, when the Social Security Act was amended to authorize funding for the treatment of ESRD under Medicare, the rate of treatment was also significantly higher in countries such as Belgium, Finland, France, and the Netherlands than in the United States.

When treatment of ESRD became available, it was obvious that many Americans could not afford it and people were dying, for lack of adequate insurance coverage. There was no way that the new technology could be accommodated within the existing system. A separate reimbursement program had to be enacted to provide equity of access to these costly new technologies. When reimbursement became available, the number of persons treated increased rapidly. By 1975, the treatment rates in the United States were higher than in all but two European countries, and, in 1980, the treatment rate in the United States was the highest in the world (Table 8). In the 1980s, the treatment rate has continued to increase and is now far higher than in most European countries, although Belgium and Switzerland seem to be catching up.

The European countries experienced the same emotional debate about equity, access, and costs, but the situation was different in that the new technology could be accommodated within the existing systems of resource allocation. However, tradeoffs had to be made between this new technology and treatment for other conditions. The need to establish priorities had a significant impact on the choice of technology for treatment of ESRD. Thus, in the United Kingdom, a high proportion of patients was treated with home dialysis. Before the establishment of the Medicare ESRD program, 40 percent of treatment in the United States was home based. Because home dialysis support services furnished by nonphysicians were not covered in the original legislation, physicians had no incentive to steer patients toward this treatment. Therefore, the proportion of patients treated by home dialysis in the United States declined to 12 percent by 1978 (Drucker, Parsons, and Maher, 1986). Also the age distribution of patients shifted, so that the proportion of patients 55 years of age or over increased from 7 percent in 1967 to 45 percent in 1978 (Drucker, Parsons, and Maher, 1986). In 1981-82, the rate of treatment was 82.4 per 1 million of population in the United States, compared with 25.8 on average for the 32 countries in the European Registry (Marine and Simmons, 1986). High-rate countries, such as Belgium and Switzerland, had 44.1 and 44.9, respectively. The share of patients 60 years of age or over at the start of treatment was 26 percent in the 32 countries, compared with 38 percent for the United States. The introduction of reimbursement has obviously shifted the indications for treatment, without any assessment of the marginal costs and benefits.

Another important difference between the United States and Europe is the share of patients treated with transplants. It is well documented that this treatment is the cost-effective alternative when possible. It also gives the patient a better quality of life. The proportion of patients with functional transplants in the United States is difficult to document, but best estimates give a figure of about 20 percent of all patients treated for ESRD (Bonair, 1988; Eggers, 1988), which is less than one-half the rate of the Scandinavian countries, the United Kingdom, the Netherlands, and Switzerland (Table 9). Efforts have also been made to provide incentives for more transplants in the United States (Eggers, 1988).

The effect on cost per case is difficult to assess for each system. However, there is reason to believe that the incentives in the United States have had an impact on not only the number of treated cases but also the cost per case. Many European health care systems seem to have performed well, compared with the United States, when we look at both cost and cost effectiveness of treatment for ESRD--that is, they have achieved a lower treatment rate and a higher share of transplants. (Also, for many years during the introduction of the technologies, the rate of treatment was higher in many European countries than it was in the United States.) The higher incidence of treated renal failure in the United States is, of course, not necessarily a bad thing. One could argue that the United States is providing better access to a life-saving technology than are the European countries. But the marginal costs and benefits of the extended indications for intervention are largely unknown, and a comparison with European countries is one way to answer the question. The comparative studies that have been undertaken between the United States and the United Kingdom (Aaron and Schwartz, 1984; Marine and Simmons, 1986), do not give a full account of how new technology is managed in Europe. By looking at a greater number of European countries, a better perspective on the U.S. allocation of resources to new technologies can be achieved.

Extracorporeal shock wave lithotripsy

A more recent example of a new medical technology is extracorporeal shock wave lithotripsy (ESWL) for treatment of kidney stones and lately also gallstones. This technology disintegrates stones through the use of shock waves and does not require an incision. It is an equipment-embodied technology and the equipment cost is several million United States dollars. A single ESWL unit can serve a large population, analogous to the specialized services of a heart surgery center or a burn unit. The cost per treated patient is dependent on the number of patients treated per year. ESWL was developed in Germany, and the first patients were treated there in 1980. The lithotriptor was approved for widespread clinical use in the United States by the Food and Drug Administration in December 1984, at which time six experimental sites had lithotriptors (Bloom et al., 1989). By July 1, 1986, there were 84 ESWL units in operation in the United States. At that time, 31 States had at least one unit; there were 11 in California, 6 each in Illinois and Texas, and 5 in North Carolina (Bloom, et al., 1989). In the spring of 1989, the number of lithotriptors had increased to about 225. In relation to the size of the population, it is about the same as in countries such as Belgium, Italy, Germany, Spain, and Switzerland (Table 10). However, there are wide variations among European countries; the United Kingdom has by far the lowest number in relation to the population.

The introduction of ESWL has brought up a number of policy issues relating to the introduction of new medical technology in both the United States and Europe. In the United States, the discussion has centered around the problems of adapting the Medicare payment system to this new technology (Office of Technology Assessment, 1986). Because it is neither "medical" nor "surgical" technology, there is no appropriate DRG to which the treatment can be assigned. If the surgical DRG is used, it will grossly overpay the procedure. If the most appropriate medical DRG is used (324-urinary stones, patient age under 70 years, without complications or comorbidities), it only covers about 60 percent of costs (Cotter et al., 1986). The compromise seems to have been to assign all patients treated with ESWL to DRG 323 (urinary stones, patient age over 69 years, with complications or comorbidities), which comes closer to the estimated costs for the procedure (Cotter et al., 1986). However, it has also been suggested that a special DRG should be established for ESWL.

Unlike the situation for many other technologies, the European countries can provide comparative information on both costs and charges for ESWL. In Table 11, a detailed account of the costs for treating the average patient with ESWL in Sweden in 1985 is shown. These costs (U.S. $3,900) include physician salaries and can be compared with an average of $5,700 in the United States (Bloom et al., 1989). The difference in costs can be studied in detail and can give important lessons about the way this technology is used and managed in the different countries. Even more interesting is to look at how costs vary among different types of patients. In Table 12, the cost per patient for treatment of different sizes of kidney stones is shown. Costs increase with the size of the stone, especially for stones larger than 30 millimeters. Detailed comparisons of costs and charges can give interesting information of relevance in assessing the consequences of the reimbursement system in the United States and can suggest ideas for reform. However, there are also other aspects of health systems that must not be overlooked. The regionalized systems in Europe can provide population-based data on utilization of medical technology that can be used to study the proper indications for as well as the consequences of new medical technology. The role of medical technology assessment in the management of new technology can also be an interesting area for comparison. The European countries do not have a formal approval system for medical devices. Nevertheless, many countries in Europe have performed more comprehensive evaluations, including not only safety and efficacy, but also cost effectiveness and quality of life, of ESWL than have been undertaken in the United States. Such comprehensive evaluations of ESWL for the treatment of gallstones are now under way in both Sweden and the United Kingdom. These assessments form part of the policy for controlling the diffusion of medical technology. In addition, the wide diversity of policies used in the European countries can yield interesting lessons for the management of the same technologies in the United States.

Are health care systems converging?

European health care systems obviously differ from the U.S. system in important aspects. There are also important differences among European systems. One important difference is the open-endedness (that is, the lack of budget restrictions) of the systems. However, the need to contain costs has forced open-ended systems to find different ways to restrict total expenditures for health care. It seems clear that a "free" market cannot solve the basic resource allocation problems in health care: efficiency and equity in health care production and consumption. Public insurance systems, tax subsidies to private insurance, asymmetric information between producers and consumers, and provider monopolies through licensing (doctors) are inherent factors in a health care system that make free competition an ineffective policy; competition has to be "managed."

In the United States, the discussion of managed competition has centered on HMOs and their ability to provide total coverage for health care at a lower cost than traditional fee-for-service arrangements. The jury is still out, but the discussion has already changed the perception of what is needed to contain costs and increase efficiency in health care. The idea of prospective payment lies behind the development of DRGs to classify patients. This system, developed at Yale University, is as close as one can come to what Oscar Lange (1938) called "market socialism." Hospitals compete against a set of predetermined administrative prices. So far, this payment method has been used only for Medicare-covered inpatient acute hospital care, but work is under way to extend it to ambulatory care. Another example of how the health care market has changed in the United States is the introduction of medical technology assessment (MTA) as a tool for health policy. The role of MTA in U.S. health policy has been controversial, but it was invented as a response to a need for more information for policymakers--this information will be needed even more in the future. Today, when the era of rapidly expanding health care resources has come to an end, new medical technology is the major dynamic factor in health care. Clearly, future policy will be aimed at control of introduction and diffusion of new medical technology. Medical technology assessment, based on explicit cost-effectiveness and cost-benefit studies, will certainly have a major role in the development of those policies.

The convergence theory implies that planning will play an increasing role in market economies. Developments in the United States during the 1980s cannot accurately be described as increased health care planning. But they certainly represent an increase in public control over the health care system. This can best be understood by looking at the great attraction HMOs, DRGs, and MTA have had for health researchers and policymakers in Europe. These ideas have fit in well with the more comprehensive and planned systems in Europe. They have been seen as a way to increase the role of markets and competition within systems in which traditional planning has proven impotent to adapt to a situation of slower resource growth with continued introduction of new medical technology. In this way we can talk about a convergence of systems.

This convergence creates interesting opportunities for the future. When new ideas, such as HMOs, DRGs, and MTA, are transferred to European countries, it is possible for the United States to learn how they work under different regulatory conditions. How will HMOs work when based on regional, rather than voluntary, participation? How will HMOs that include the elderly work? It has been suggested that the evidence for cost savings from HMOs in the United States is not relevant for Europe, because the majority of HMO members are under 65 years of age. This may be the case, but one hypothesis could be that HMOs will yield even greater savings for the elderly, who have higher consumption rates of care and for whom there are more alternatives for intervention. These questions can be answered, if experimentation on a broad basis can be started in Europe. Changes in the health care systems of the Netherlands and the United Kingdom, based on Enthoven's ideas, are already under way. The first HMO is also about to be started in Switzerland.

The DRG system has been introduced, at least on the research agenda, in most European countries from Sweden in the north to Portugal in the south. There is also a very interesting experiment under way with financial incentives for hospital efficiency in Leningrad, U.S.S.R. (Hakansson et al., 1988). The research and the experiments in Europe will give not only interesting comparative data on DRG groupings, but also information about how, for example, physician costs can be integrated in the DRG payment and how this will change the weights.

Medical technology assessment, including consensus conferences, has also been imported into Europe from the United States. The Netherlands and Sweden have been in the forefront of this movement and have established special government committees for MTA. The European Regional Office for the World Health Organization has a special program for MTA, and introduction of MTA is one of the targets for "Health for All" by the year 2000. Within the European Common Market, a special committee for MTA has been set up within the health service research committee. It seems that today, MTA is more vital and growing in Europe than in the United States. This will, in the longer term, produce important information on medical technologies that are also used in the United States and will give lessons on how MTA can be implemented as part of health policymaking.


The single most important lesson from the European health care systems during the 1980s concerns the role of central government. The European systems have been able to reduce the total costs of health care not so much through central planning and regulation as through global budgets at the regional level. In fact, the role of the central government in health care policy has never been as strong in Europe as is perhaps thought in the United States, and during the 1980s there has been a strong trend toward decentralization. The reason for this is the obvious difficulty in managing such a huge and complicated system from the center. Compared with Europe, the U.S. Federal Government seems to have less control over the totality of the system, at the same time that it is more directly involved in detailed regulation of efficacy, safety, and price setting. Leadership and control of global expenditures and decisions regarding the comprehensiveness of the system must come from the center, but planning and management should be left to the regional level. Decentralization can be combined with internal markets and competition among providers. Planning and markets are not necessarily antithetical; they can work together to create better health services. [Tabular Data 1 to 12 Omitted] [Figure 1 Omitted]

(1)The results shown by Le Grand and Rabin (1986), a decline during the past 50 years in the Gini coefficient for variation in the age of death, can be described as a "rectangularization" of mortality. For a discussion of rectangularization of morbidity, see Fries (1980). Please note that rectangularization is not caused by changes in demography but by changes in the epidemiology of disease, partly caused by medical interventions.
COPYRIGHT 1989 U.S. Department of Health and Human Services
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1989 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:International Comparison of Health Care Financing and Delivery: Data and Perspectives; Symposium: International Comparisons of Health Care Systems; includes respondents
Author:Jonsson, Bengt; Henke, Klaus-Dirk; Reinhardt, Uwe E.; Davis, Karen; Meyer, Jack A.
Publication:Health Care Financing Review
Date:Jan 1, 1989
Previous Article:What can Europeans learn from Americans?
Next Article:Health care expenditure and other data.

Related Articles
Overview of international comparisons of health care expenditures.
What can Europeans learn from Americans?
Health expenditures in major industrialized countries, 1960-87.
International infant mortality rankings: a look behind the numbers.
What Can the U.S. Learn from National Health Accounting Elsewhere?
Health Expenditure Trends in OECD Countries, 1970-1997.
Health-Care Quality Promotion through Infection Prevention: Beyond 2000.
High health care utilization and costs associated with lower socio-economic status: results from a linked dataset.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters