Journal Issue

Hungary: Selected Issues

International Monetary Fund
Published Date:
November 1997
  • ShareShare
Show Summary Details

IV. Health Care Reform in Hungary48

A. Introduction

54. The purpose of this chapter is to review the progress in the structural reform of the public health care sector, which began during the 1990s, and led to substantial structural changes. Lately, however, the pace of reform has lost momentum, as the recent legislation in this area represents only a limited improvement.

55. The first section of this chapter describes the main features of the Hungarian health care system, including the changes introduced in the early 1990s. The second section describes the reforms implemented during the last two years.

B. The Public Health Care Sector

Institutional characteristics

56. The Hungarian health care system is mainly public and insurance-based. Employers and employees pay health insurance contributions to the Health Insurance Fund (HIF) and, starting in 1997, everyone, including the adult nonworking population, also pays a flat per capita fee to be eligible for health services. The public purchaser of health care services is the HIF, which also subsidizes a portion of the retail price of pharmaceuticals and medical supplies and disburses several sickness-related payments, including sick pay.49 In addition, the central government, mainly through the Ministry of Social Welfare and various Central Budgetary Institutions (CBIs) (for investments, teaching and county hospitals, education, and research), and the local governments (mainly for investment purposes, as they own the local hospitals) perform a variety of health-related functions.50

57. The terms and conditions of purchase of health services by the HIF are spelled out in service contracts that, every year, are signed by the HIF, the hospitals, and the Medical Chambers (physicians’ associations)—though local governments (LGs) and trade unions also participate in the negotiations. The remuneration mechanisms differ for each category of provider and by type of service provided (see Appendix I). For example, while hospitals are essentially remunerated on the basis of output indicators,51 the family doctors (or general practitioners, henceforth, GPs) are remunerated on a capitation basis.

58. Public expenditure for health care varied between 6.5 percent and 7.4 percent of GDP from 1991 to 1994, and then declined to 6.3 percent of GDP in 1996 (Table 4). In addition, household expenditure on pharmaceuticals steadily increased during the 1990s in parallel with their price and import liberalization and the reduction of subsidy rates (see below). Total private expenditures on pharmaceuticals increased by 0.4 percent of GDP to 1.1 percent between 1993 and 1996.52 Public expenditure on health is not large for Hungary’s level of economic development—as measured by the level of GDP per capita—in comparison both with the OECD countries (Figure 11) and with the central European countries (Figure 12). Similarly, total (public plus private) expenditure on health in Hungary is not large in comparison with those of other OECD countries (Figure 11).5354

Figure 11.Hungary: Health Expenditures vs GDP in OECD Countries, 1995

(Per capita)

Sources: OECD, 1997.

Figure 12.Hungary: Public Health Expenditure Per Capita vs GDP Per Capita Selected Central European Countries, 1994 1/

Sources: Goldstein and others, 1996; WEO, 1997; and staff estimates.

1/ Countries: Hungary, Albania, Bulgaria, Croatia, Czech Republic (1993), FYR Macedonia (1993), Lithuania, Poland (1993), Romania, Slovakia, Slovenia.

Table 4.Hungary: Health Expenditures, 1991–96(In billions of forint)
1HIF total221.4265.5333.1397.8443.4509.0
Of which:
2Health services93.4115.5133.0172.1192.7228.8
3Pharmaceutical subsidies 1/39.443.054.269.681.697.6
drugs only39.449.561.670.084.2
4Pregnancy and childbirth benefits6.
5Sick pay27.428.235.340.839.833.0
6Disability pensions32.238.645.057.868.179.3
7Other social transfers\17.821.823.
8Administrative + Other expenditures5.210.97.815.113.723.4
9Transfers to PIF0.
In percent of GDP (1)
Health expenditures in percent of GDP (2+3) 2/
10Central plus local governments26.847.
11Ministry of Welfare4.319.622.631.135.646.8
12Other ministries7.49.710.114.513.113.0
13Investments by CB and LGs 3/15.117.619.526.623.319.2
In percent of GDP (10)
ITotal health expenditures of general government (2+3+10) 2/162.8212.2243.8323.6355.0422.5
In percent of GDP6.
IITotal private expenditure on drugs7.711.624.
In percent of GDP0.
Of which:
Co-payments on prescriptions7.711.615.220.530.437.5
In percent of GDP0.
IIITotal health expenditure (I+II)170.6223.9267.8355.6406.0493.5
In percent of GDP6.
Memorandum items
HIF + CB + LGs + Private outlays (1+10+11)256.0324.1409.2502.1566.3659.1
In percent of GDP10.211.011.511.510.29.9
Sources: CSO; 1993; Health Insurance Fund data; and staff estimates; 1991–92 private expenditures on drugs are from Ho and Orosz, 1994.

Includes medical supplies

Includes proportional allocation of administrative and other expenditure (line 8); and HIF’s outlays for hospital debt in 1996 (about 0.1 percent of GDP).

Excludes local government support for current costs and gratuity payments.

Sources: CSO; 1993; Health Insurance Fund data; and staff estimates; 1991–92 private expenditures on drugs are from Ho and Orosz, 1994.

Includes medical supplies

Includes proportional allocation of administrative and other expenditure (line 8); and HIF’s outlays for hospital debt in 1996 (about 0.1 percent of GDP).

Excludes local government support for current costs and gratuity payments.

Table 4.Hungary. Health Expenditures in Hungary, 1991–96(In percent of GDP)
1HIF total8.
Of which:
2Health services3.
3Pharmaceutical subsidies 1/
Drugs only1.
4Pregnancy and childbirth benefits0.
5Sick pay1.
6Disability pensions1.
7Other social transfers0.
8Administrative + other expenditures0.
9Transfers to PIF0.
Health expenditures (2+3) 2/
10Central plus local governments1.
11Ministry of Welfare0.
12Other ministries0.
13Investments by CB and Lgs 3/
ITotal health expenditures of general government (2+3+10) 2/
IIPrivate expenditure on drugs0.
Co-payments on Prescription0.
IIITotal health expenditure (I+II) 3/
Memorandum items
HIF + CB + LGs + Private outlays (1+10+11)
Sources: CSO; 1993; Health Insurance Fund data; and staff estimates; 1991–92 private expenditures on drugs are from Ho and Orosz, 1994.

Includes medical supplies

Includes proportional allocation of administrative and other expenditure (line 8); and HIF’s outlays for hospital debt in 1996 (about 0.1 percent of GDP).

Excludes local government support for current costs and gratuity payments.

Sources: CSO; 1993; Health Insurance Fund data; and staff estimates; 1991–92 private expenditures on drugs are from Ho and Orosz, 1994.

Includes medical supplies

Includes proportional allocation of administrative and other expenditure (line 8); and HIF’s outlays for hospital debt in 1996 (about 0.1 percent of GDP).

Excludes local government support for current costs and gratuity payments.

Main reforms in the early 1990s

59. Under the socialist system, health care was granted free of charge to every citizen, and was financed from the general revenue of the state and local councils. In the late 1980s, the government openly acknowledged the growing health crisis and the inefficiencies of the health care system (Makara, 1994). In the first half of the 1990s, important and radical measures were taken to reform the institutional, financing, and ownership arrangements (Appendix II). During this relatively short period, Hungary moved from a centrally planned public health care system, financed through general budgetary revenue, to a delegated and decentralized system, financed through a combination of social insurance payments (to cover the outlays of the HIF), general budgetary revenue (for the outlays of the State and LGs), and patients’ outlays for drugs. The main thrusts of the reforms are discussed below.

60. As a first step toward establishing an effective system of primary care, the government attempted to raise the profile of public health and primary services with the creation of the Public Health Service (PHS) in 1991 and of the Family Physician Service in 1992. The PHS assumed responsibility for local public health stations formerly under control of the LGs, with the objective of implementing national public health policies, supervising health professionals, and controlling quality in the delivery of health services. In 1992, physicians were allowed to become private GPs on the basis of contracts signed with both the LG and the HIF. Patients were also allowed to choose their own GP—with the objective of improving quality of care through competition. In order to reduce demand for expensive specialist and hospital services, all non-emergency cases (with the exception of gynecological cases) required written referral from the GP. The government hoped that the referral system would allow GPs to perform their gate-keeping function. In fact, this has not happened, as GPs were willing to prescribe medicines, tests, and specialist visits for the requesting patient.

61. The pharmaceutical market was progressively liberalized. In 1989, the government discontinued budgetary producer subsidies on drugs. Subsidies were paid by the Social Security Fund directly to the pharmacists and were set as a percentage of retail price—depending on the social and medical value of the drug. In 1991, the government replaced controls on the retail price of drugs with negotiated profit margins and lifted restrictions on medicine imports. Overnight, a free pharmaceutical market was created, and patients gained access to more, and better quality, drugs. Nevertheless, public health expenditure on pharmaceuticals declined by 0.1 percent of GDP to 1.3 percent from 1993 to 1996 (Table 4). This was achieved through continuous revisions of the subsidization rates and of the number of medicines in each subsidy group. As a result, between 1991 and 1996, the average of co-payment on prescription drugs increased by 10 percentage points, to 29 percent, and households’ outlays on prescription drugs doubled, to 0.6 percent of GDP (Table 5).55

Table 5.Hungary: Cost Sharing of Prescription Drugs, 1992–97
199219931994199519961997 1/
(In billions of forint)
Central government2.
(In percent)
Central government3.
(As percent of GDP)
Central government0.
(In billions of forint)
Memorandum item:
Gross Domestic Product2,942.73,548.34,368.35,561.96,660.08,025.0
Source: HIF.

January to April only.

Source: HIF.

January to April only.

62. The government also introduced performance-based remuneration, in an effort to increase efficiency by linking payments to performance. Before these changes, health providers were remunerated on the basis of salaries (doctors) or of historic costs (inpatient and outpatient services).56 The new methods for paying health providers were broadly based on the following principles: capitation-based pay for family physician services; historic cost plus fee for service for outpatient specialist care; Diagnostic Related Groups for short-term inpatient care; fee for service for special treatments during short-term inpatient care; and per diem for long-term care (Appendix I). These methods of payment are fraught with potential complications, however, and unless carefully designed and implemented, could undermine the objective of enhancing efficiency that they were originally established to accomplish (see below).

63. In 1992, the HIF was established as an autonomous purchaser of health services. The existence of a single purchaser is not in itself a prerequisite for efficient health service delivery, although it may offer some important advantages that could result in more efficient/less expensive service delivery (Oxley and MacFarlane, 1995).57 For these advantages to materialize, the purchaser should be able to assess the relative merits and cost effectiveness of different treatment strategies, and to selectively buy health services from potential suppliers. At the same time, the purchaser should be accountable to consumers (through parliament) for the quantity and quality of services provided. These conditions do not materialize in Hungary (see below).

64. During the early 1990s, policymakers were able to first contain public health expenditures and then to reduce them (¶58). As part of the financing reforms of the early 1990s, the government established spending caps for three main outlays of the HIF: family physician services, outpatient care, and hospital care. These cost control mechanisms have managed to keep a cap on expenditures even as performance-based remuneration was introduced and the structure of ownership changed (Table 4). The expenditure restraints in the health sector have also resulted in a reduction of more than 20 percent in the real earnings of non-manual workers in the public health and social welfare sector between 1992 and 1996 (Figure 13). The squeeze on earnings in the health and social welfare sector has been larger than the one experienced by the public administration as a whole. In fact, the ratio between gross earnings in the health sector and the public administration decreased by about 5 percentage points, to 65 percent, between 1992 and 1996 (Figure 13).

Figure 13.Hungary: Wages and Average Monthly Gross Earnings, 1992–96

Sources: Monthly Bulletin of Statistics, various issues; Statistical Yearbook of Hungary, 1995; and staff estimates.

65. Two important sources of cost escalation remain, however: high-technology services, which are remunerated on a fee-for-service basis, and pharmaceuticals.58 These are among the main reasons that explain why, every year, the expenditures of the HIF have been higher than budgeted (Table 6)—other less important reasons are post-budget increases in wages of health sector employees and underestimates of disability pension outlays.

Table 6.Hungary: HIF Approved Budgets and Outcomes, 1992–1996(In billions of forints)
Total Revenue238.6235.8270.5280.3336.4379.7436.0422.9489.5465.5
Total Contributions219.3212.0236.6251.0283.8294.9344.2332.4381.0353.7
Transfers from central budget1.
Other social security related revenues8.19.111.612.614.115.812.812.714.412.6
Other Revenue0.
Transfer between funds10.410.113.511.321.354.756.856.766.966.1
Operational revenues0.
Total Expenditures238.6257.5287.0306.0336.4397.8436.0445.1491.0509.0
In-kind provisions141.6156.2174.8186.9225.8241.5270.4275.7304.1326.1
Curative and preventive care110.6112.1129.6131.6164.5170.5192.3190.1217.5226.8
Pharmaceutical expenditures31.039.438.049.550.761.667.770.072.084.2
Medical supplies subsidy3.
Travel reimbursement1.
Cash benefits73.472.981.388.698.9108.0116.3118.2131.1121.5
Under retirement age disability36.036.543.
Maternity and child allowance6.
Sick pay29.128.929.035.334.140.834.839.842.833.0
Joint expenditure of the funds23.325.
Transfers between the funds32.936.537.140.037.2
Other expenditures0.
Percent of GDP0.
Source: Health Insurance Fund.
Source: Health Insurance Fund.

66. The successful capping of total public expenditures may have created some negative side effects. Among the unintended side effects are informal rationing, deterioration in quality of care, and arrears accumulation. While there is no hard evidence that either informal • rationing increased or quality of care deteriorated in Hungary,59 lower quality and longer waiting times for surgery and specialty visits are not an uncommon occurrence in health care systems with output caps (OECD, 1994). There is instead evidence of accumulation of hospital arrears to suppliers in Hungary during 1995 and 1996.60 The health reform experiences of OECD countries suggest that constraints on overall expenditure levels alone have rarely encouraged greater efficiency and effectiveness of providers. In some cases, overall expenditure caps may have weakened the achievement of these objectives, especially if the caps have not been accompanied by micro-level reforms to improve efficiency (Oxley and MacFarlane, 1995). These micro-level reforms are the main object of the 1997 legislation on the health care sector in Hungary (Section C).

The challenges ahead

67. While Hungary’s expenditure on health is not high by international standards, and despite the systemic changes in the health care sector during the first half of the 1990s, the efficiency of the health care system is a matter of continuous discussion in Hungary. This is sometimes triggered by the evident deterioration of the health status of the population—often referred to as “the health crisis” (Makara, 1994). The extent of this deterioration, the efficiency of the health sector, and their implications for health policy, are discussed in this section.

68. The health status of Hungary’s population, as measured in terms of life expectancy at birth, is by far the lowest among the OECD countries and one of the lowest among the central European economies in transition (Figure 14).

Figure 14.Hungary: Life Expectancy Comparisons

Sources: World Bank, 1997; Goldstein and Others, 1997; and OECD, 1997.

1/ Countries: Hungary, Albania, Bulgaria, Croatia, Czech Republic (1993), Poland (1993), FYR Macedonia (1993), Romania, Slovakia, Slovenia.

69. This was not the case before the mid-1970s. The life expectancy of a Hungarian male is two years lower now than in 1970–75 (Figure 15), a unique occurrence among OECD countries.61 While the life expectancy estimates for Hungarian and Austrian males moved together during the period 1955–75, a gap of six years has developed between the two since then. The life expectancy gap has widened also with respect to males of central European countries in transition (Figure 15), albeit not by as much as with respect to OECD countries. Life expectancy for women has continued to improve, but not at the same rate as in other European countries (Figure 15).

Figure 15.Hungary: Life Expectancy at Birth in Selected Countries, 1950–95

Sources: World Bank, 1997; United Nations, 1996.

1/ The three countries selected are representative of mortality developments of Northern Europe/ Scandinavia (Sweden), Western Europe (Austria), and the Mediterranean (Spain), Jozan (1994).

70. The main causes of death are related to lifestyle,62 and these causes affect the lower social classes disproportionately more than the upper classes in Hungary (Jozan, 1994). Male mortality is also substantially higher in rural areas. These mortality patterns offer important pointers for designing an improved health policy. Lifestyle may be influenced by targeted education and information campaigns aimed at reducing smoking and alcohol consumption, and by high excises on alcoholic beverages and tobacco. The main target groups of preventive public health and primary care in Hungary would be the non-urban population and the lower income quintiles. Expecting and lactating mothers would be other obvious candidates for targeting through an integrated approach involving pre- and post-natal monitoring by GPs, nutrient distribution, and education for new mothers.

71. As discussed above, the poor health status of the Hungarian population cannot be entirely ascribed to the level of efficiency and effectiveness of the health care system. Low efficiency nevertheless tends to increase expenditures and/or results in inferior health outcomes and social welfare—insofar as the public is not satisfied with the level and quality of care—and/or unequal distribution of health services. The government’s efforts to reform the health care sector are indeed motivated by its perceived low level of inefficiency and effectiveness—in terms of health outcomes. The remaining part of this section discusses the efficiency of the health care system.

72. The level of efficiency in health care delivery can be gauged by examining some indicators of capacity and input utilization: hospital beds, doctors, nurses, and contacts with physicians. What transpires is an inadequate organization of supply, characterized by: excessive reliance on inpatient care, and, in particular, expensive acute care—rather than chronic or rehabilitative—very low ratio of nurses to doctors, and very high utilization of specialist services.63 What this indicates is that the Hungarian health care system is geared toward more expensive and less effective curative and specialized care, rather than preventive care and basic services.

73. The capacity of the Hungarian health sector, as measured in terms of hospital beds per 1,000 people, was 37 percent higher than the OECD average in 1995 (Figure 16), even though Hungary has lower levels of both health expenditure and GDP per capita than the average OECD country. The number of days in hospital per 1,000 people is 19 percent higher than the OECD average, indicating that recourse to hospitalization is relatively high (Figure 16). However, given the large number of available beds, the occupancy rate remains 2.5 percentage points below the OECD average of 77 percent (Figure 16). This picture is more accentuated for acute care beds. In 1995, Hungary had 52 percent more acute care beds and 50 percent more days of stay in acute beds than the average OECD country (Figure 17). There is, therefore, an imbalance between acute and chronic (long-term) care beds in Hungary. Institutions caring for the elderly are almost entirely missing, and hospital wards for rehabilitation or for treating long-term diseases are in short supply. As a consequence, many beds in acute care departments are instead used for chronic care patients (Ho and Orosz, 1994). Until 1994, Hungary had no facilities for outpatient surgery.

Figure 16.Hungary: Inpatients, 1995

Source: OECD, 1997.

1/ The relative position of Hungary does not change despite the reduction of the number of beds in 1996. The total number of beds in 1995 is 93842 (9.3 per 1000); the number of beds in 1996 is estimated at 89491 (8.9 per 1000).

Figure 17.Hungary: Acute Care, 1995

Source: OECD, 1997.

74. The level and combination of manpower inputs in Hungary is even more striking. Hungary has by far the highest number of specialists and practicing physicians per 10,000 people than any other OECD country (Figure 18). Moreover, it also has 32 percent fewer nurses than the average OECD country (Figure 18). This input mix suggests that nurses are not employed as “screeners” or “first-contact doctors” to reduce physicians’ and specialists’ time with patients, which is a common method to reduce costs without a loss in effectiveness.

Figure 18.Hungary: Number of Medical Professionals, 1995 1/

(Per 1,000 population)

Source: OECD, 1997.

1/ Or latest available period 1990–95.

75. Physicians function at extremely high turnover rates, notwithstanding expenditure levels that are lower than in other OECD countries. Despite the extremely high number of practicing physicians in Hungary, each physician conducts 41 percent more patient visits—including outpatient visits in hospital wards—than the average physician in the other eight OECD countries for which data are available (Figure 19). Moreover, physicians’ consultations per capita are more than twice the average of the 12 OECD countries for which data are available (Figure 19). These data suggest that patients’ contacts with physicians may result in a low content of effective care—for example, patients may see doctors just to get a written referral to a higher level of care, including referrals for medical tests.64

Figure 19.Hungary: Patients’ Contacts with Physicians, 1994

Source: OECD, 1997.

1/ Australia (1991), France (1990) and Canada (1993).

2/ Includes outpatient treatments in hospital wards.

76. Why is the Hungarian health care system characterized by excess capacity and relatively poor indicators of performance? The roots of the problem may be traced to the inadequate incentive mechanisms facing health care providers, including their method of payment, and in the institutional setup (Ho and Orosz, 1994; Schneider and Jacobs, 1996; World Bank, 1997b).

77. Remuneration formulas have important implications for expenditure control and efficiency and, without careful design and implementation, they could undermine the objectives they were designed to achieve. For example, the DRGs, used for inpatient acute care, should facilitate data collection for a central policy agency, foster competitive contracting for treatments, and provide hospitals with incentives to increase turnover (i.e., reducing length of stay). However, the benefits that may be reaped by using the DRGs are not maximized because of the way the DRGs are used in Hungary. First, there is no competitive contacting in Hungary, as the HIF is forced to contract with each provider for levels of service supply that are set by law (Section C). Second, since there is a fixed total budget for financing hospitals in the HIF, allocations are made proportionally to individual hospitals retrospectively every month on the basis of their output share, as accounted in terms of DRGs.65 Therefore, DRGs are used more as an accounting instrument than as instrument to reward output.

78. The capitation system for GPs, while allowing control over the overall level of primary health expenditure, does not provide sufficient incentives to improve the content of family medicine, since remuneration remains the same, regardless of the services provided. Moreover, capitation-based salary tends to have a negative impact on the gate-keeping function of GPs, thereby increasing outlays for secondary care. In fact, capitation fees induce GPs to register too many patients and to underservice them, select the better risks, and refer them too quickly to more expensive secondary providers—specialists and hospitals (Oxley and MacFarlane, 1995). A lesser role for GPs in primary care reduces the production efficiency of the health care system, by shifting the mix of care toward higher levels of service.

79. The institutional setup in Hungary does not provide for a clear separation of responsibilities between the purchaser of health services (ultimately the State through the HIF) and the providers of these services. This separation is widely regarded as a prerequisite to improving the quality and efficiency of care (Schneider and Jacobs, 1996), which can be pursued through negotiations between purchasers and providers and competition among providers. In the legal setup in Hungary, the State guarantees: access to health services for the population; the provision of deficit financing for the HIF;66 and remuneration for health care providers. The municipalities (therefore, the general government) are the owners of hospitals. By law, the HIF has to purchase health services for the population. As a consequence, the State and the HD? together are ultimately responsible for financing the production of health care services by the providers.67 Therefore, the purchaser of health care services (the HIF) is effectively also the lender of last resort of health institutions. The separation of responsibilities between health providers and service purchasers is therefore blurred.

80. The low effectiveness of the health care system has been traced back by some authors to the system’s inability to adapt to changing demographic and epidemiological conditions of the population (Orosz, 1993; SZDSZ, 1993; Makara, 1994). Any health care system needs to be flexible to rearrange service delivery to meet the changing conditions of the population it serves. This will become a more important requirement in the coming decades, when the Hungarian population will age considerably (Figure 20). To adapt to these trends, health care providers would have to reduce considerably the supply of short-term beds (including maternity wards), while increasing the supply of geriatric facilities. Traditional types of home and nursing care could also be promoted by the State. The latest legislative changes have in part addressed these issues (Section C).

Figure 20.Hungary: Structure of the Population, 1955–2035

Sources: United Nations, Demographic Yearbook, various issues; World Bank Population Projections, 1995.

C. Structural Reform in the Health Sector During the Last Two Years

81. The purpose of this section is to review to what extent the legislation approved during the last two years addresses the sources of inefficiency in Hungary’s health care delivery system. Reforms in health care delivery have been approved in two distinct periods in the last two years. First, the Act on Capacity Reduction was adopted in July 1996.68 Second, in June-July 1997, Parliament approved a set of laws aiming at containing costs while improving the efficiency of health care delivery.69 This body of legislation comprised: (i) the Act on Social Insurance Eligibility and Financing, which regulates revenue of both the HIF and the Pension Fund; (ii) the Act on Health Insurance, which defines the health benefits available under the health insurance package;70 and (iii) amendments (referred to as Institutional Law) to the 1991 Law on the self-government of Social Insurance that redefine the procedures for selecting the representatives of trade unions and employers to the governing body of the Social Insurance Institution. Additionally, the government has presented to Parliament a draft of the Health Care Act, which would regulate the organization of the health sector. This section discusses only those aspects of these laws (Act on Capacity Reduction, the draft Health Care Act, and the Act on Health Insurance) that deal with the structural reform of the health care delivery system.

82. The Act on Capacity Reduction aims at reducing excess capacity in the hospital system, excessive and inappropriate hospitalization rates, and excessive length of stay. It defines inpatient and outpatient capacity by county and for each specialty until 2000.71 The Act also recognizes that bed reductions have to be determined locally, by ad hoc consensus committees, and it spells out the negotiating process between the consensus committee and the HIF. The objective of the government was to reduce the number of hospital beds by 10,000 in 1996. Beyond 1996, the planned reductions were in the range of 1,000–2,000 annually until 1999, when it was hoped that the formula would provide a ratio of 80–85 hospital beds per 10,000 inhabitants and an occupancy rate of about 80 percent.

83. As it turned out, the formula delivered a smaller reduction in capacity than was expected. Both the average number of beds used by hospitals and those contracted by the HIF were reduced by fewer than 5,000 units in 1996 (Table 7). However, the HIF expects that the reductions in 1997 and 1998 will be larger than originally expected and that the act will yield the expected downsizing by the end of the four-year period.72

Table 7.Hungary: Hospital Beds, 1993–2000
Statistical beds 1/104,764104,764
Actual beds in use100,43898,45395,98891,514 2/
Contracted beds96,35596,31493,84289,49184,50081,30079,60078,000
Occupancy rate0.750.740.740.760.790.80.80.8
Source: HIF

Statistical beds coincide with actual beds in use from 1995.


Source: HIF

Statistical beds coincide with actual beds in use from 1995.


84. The Health Care Act was intended to reshape the organization of the sector and replace the 1972 Health Act, which had been amended several times since the beginning of the economic reforms.73 The current draft of the law is basically that of a framework law, outlining the basic principles regarding the role of the government, the rights and responsibilities of individuals, the agents/units included in the health delivery system, and the professional requirements for supplying health services (Appendix III). The draft law also underscores the importance of prevention and primary care. There is little dispute that any health care system should be informed by most of the principles in the draft. For example, the Act clarifies that the role of the State in health care provision is to ensure equal rights to health care by: (i) setting health policy, including control over economic activities with an adverse health impact; (ii) organizing and planning the health care delivery system; and (iii) defining the extent of health care provision. However, since the draft act is a framework law, it remains quite general and does not contain any apparent structural improvement. For example, rather than “organizing and planning the health delivery system” (point ii above), the Act defines in general terms what the health delivery system is comprised of (i.e., “institutions and professionals responsible for health promotion and prevention,…regardless of ownership.”) and the general principles that these institutions and professionals should follow in delivering health services (i.e., emphasis on prevention and primary care).

85. The Act on Health Insurance regulates the benefit package provided by the HIF, the modalities of provision, the contracting procedures between the HIF and the service providers, and assigns ministerial responsibilities for follow-up legislation (Appendix IV). The main objective of the Act is to increase the efficiency of the health system, mainly by diverting excess demand for secondary health care services toward preventive and primary care. The main points of the Act are summarized below.

86. First, the Act classifies health treatments into three broad categories, according to the degree of patients’ participation in the cost of services: i) treatments not covered by insurance (e.g., several types of transplants, sterilization, and acupuncture); (ii) treatments provided free of charge (e.g., screening, preventive, and curative services); and (iii) specific situations when the institution delivering the service is authorized to collect co-payments.74 Second, the Act specifies treatment guidelines that regulate access to services. These include the requirement of a written referral for most out/inpatient treatments and the physicians’ obligation to refer patients to the lowest possible level of care. Third, the Act redefines the HIF’s contracting procedures with, and its supervision over the health institutions and the physicians. Fourth, the Government and the Ministry of Social Welfare (MoW) are mandated to design implementing legislation spanning most aspects of the Act.75

87. The emphasis of the Act is on regulating utilization through access and treatment protocols and information-intensive controls. Although the Act attempts to improve efficiency in health care delivery, some of its regulations may not have the desired impact on this objective. In fact, some regulations either do not bring about the needed improvement in the incentive structure or they establish disincentives for health care suppliers to operate within a strict budget constraint. Moreover, several regulations in the Act either re-present solutions and arrangements that are already operating—unsuccessfully—or that need to be implemented through further legislation. These four issues are discussed below.

88. First, with regard to the incentive structure, at least two areas are germane. First, there is little incentive for health institutions to charge co-payments.76 In a very few specific cases (for higher level services than those defined in the treatment protocols and for services demanded without written referral) health providers may, but are not required to, collect co-payments. There is no national mandate for imposing co-payments on specific treatments; rather, each medical institution may decide to charge co-payments when utilization and referral guidelines are not followed. However, if the hospital charges the patients, it will have to charge the HIF less for reimbursements and physicians will lose their gratuities—which are an important component of their income. Moreover, in the absence of any threat of bankruptcy or sanction for the hospital administrators, the latter may not be motivated to enforce controls on the propriety of referrals—on which the imposition of co-payments hinges—for the following three reasons. First, since hospitals are financed by the HIF on the basis of output indicators, co-payments may have the undesired effect of reducing available financing by lowering demand and output. Second, it may be difficult for an hospital administrator to define which level of care is appropriate for each diagnosis and, unless there are very specific national guidelines in this area, each individual institution may prefer not to impose co-payments, to avoid losing patients. Third, the new procedures for providers in financial difficulty (see below) will also discourage the use of co-payments, as these procedures soften the budget constraint on hospitals.

89. The second factor that may not yield an improved incentive structure is that the recent law fails to assign broader autonomy to the HIF to contract with providers for the level and composition of capacity it deems necessary. Although the contracting procedures between the HIF and the providers of health services have now been clarified, the HIF remains obliged to contract with each county for at least the level of capacity specified in the 1996 Act on Capacity Reduction. This act fixed the level and composition of health care capacity by county until 2000. Therefore, the HIF’s ability to influence the allocation of resources for efficiency and effectiveness gains by negotiating contracts with health providers remains limited.

90. Second, with regard to disincentives for health care suppliers to operate within a strict budget constraint, the Act now mandates the HIF to lend interest-free funds to health providers in financial difficulty, provided the owner of the institution (mostly LGs, and universities and the MoW) also supplies own resources and guarantees of payment. This regulation effectively weakens the budget constraint on health institutions and sanctions the possibility for them to overcommit expenditures, while forcing the HIF to finance, rather than sanctioning, such overspending. There is limited incentive for financial accountability, or against debt accumulation.

91. Third, the Act re-proposes arrangements that are already operating—unsuccessfully—but it does not change the mechanisms that make these arrangements ineffective. For example, written referrals are already required for higher-level treatments, but this requirement has not been observed. The difference in the new act is that institutions can now charge a co-payment to patients without referral and for referrals to services above the “lowest possible level of care.” Although the power to charge co-payments is a clear improvement, it is unclear how each hospital and specialist will be able to decide what price to charge, or what constitutes the “lowest possible level of care” for each diagnosis. The latter is a subjective medical opinion that does not lend itself to regulation. Therefore, it may be difficult for health providers to justify charging a co-payment against a referral to higher-level treatment by the GP. Another example of an arrangement that is currently operating is that the MoW already has the task of preparing treatment protocols, and the HIF has had to monitor their observance. The MoW has finalized five prescription protocols over the last two years. However, the role of these protocols is simply one of guidance, as the HIF cannot impose prescription ethics on GPs. The experience with the existing five protocols has shown that the supervising agency (the HIF) does not have the legal entitlement or the instruments to verify whether doctors/institutions prescribe and treat according to the protocols.77 Moreover, it is unclear why service providers would start using treatment protocols as a yardstick for charging co-payments (see discussion on incentives above).

92. Fourth, several regulations in the Act have to be implemented through further regulatory acts, which may result in further delays in the effective implementation of the Act. A shortlist of the necessary implementing regulation would include treatment protocols, a list of basic treatments to be provided free of charge, and revised remuneration mechanisms for health care providers (Appendix IV). Most of these regulations are of key importance for the effective implementation of the Act. For example, if the list of the basic free-of-charge treatments is not available, it may be more difficult to justify imposing co-payments on specific cases of mis-referral. Also, since only chest X-rays are currently mandatory, it may be difficult to reduce excess demand for higher level services or to collect many co-payments from patients who “miss the mandatory screenings”—unless more screenings become mandatory. However, this is not a simple task. The MoW or the HIF would have to prepare a list of mandatory screenings, and ready the necessary facilities and human resources. Moreover, the MoW and the HIF would have to identify what screenings should become mandatory for each higher-level treatment. All this will take time to plan and implement and, until the process is completed, it may be difficult to justify charging co-payments to patients with no screening record.


    Central Statistical Office (CSO)1995Main Features of the Hungarian Demographic Situation in the Early Nineties (Budapest).

    Central Statistical Office (CSO)1993Statistical Yearbook of Hungary 1992 (Budapest).

    GoldsteinEllen and others1996“Trends in Health Status, Services, and Finance,”World Bank Technical Paper No. 341 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    HoTeresa J. and ÉvaOrosz1994“Health Reform In Hungary: Taking Stock,”mimeo.

    JózanPéter E.1996“Health Crisis in a Society in Transition: The Hungarian Case,”mimeo.

    LosoncziÁ.1986A kiszolgáltatottság anatomiája az egészségügyben (Anatomy of defencelessnessBudapest).

    MakaraPeter1994“Four Variations on One Topic: Changes in Hungarian Health Policy (1980–1994),”case-study for the European Conference on Health PolicyCopenhagenDecember5–9.

    • Search Google Scholar
    • Export Citation

    MakaraPeter1994“Health for All in CCEE/NIS - the Policy Environment,”mimeo.

    MusgrovePhilip1996“Public and Private Roles in Health: Theory and Financing Patterns,”World Bank Discussion Paper No. 339 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    National Health Insurance Fund Administration1996Statistical Yearbook 1995 (Budapest).

    Organization for Economic Coordination and Development (OECD)1997Health Database (Paris).

    Organization for Economic Coordination and Development (OECD)1994The Reform of Health Care Systems: A Review of Seventeen OECD Countries (Paris).

    • Search Google Scholar
    • Export Citation

    OroszÉva1996“Decentralization and Health Care System Change: Hungary,”mimeo.

    OroszÉva1996“Main Challenges for Health Policy in a Period of Socio-economic Transformation in Hungary,”paper presented to the Third International Research Seminar on “Issues in Social Security”Sigtuna, SwedenMay11–14.

    • Search Google Scholar
    • Export Citation

    OroszÉva1993“Two Possible Scenarios of Health Care,”draft paper prepared for the “Hungary in 2005” EC project.

    OxleyHoward and MaitlandMacFarlan1995“Health Care Reform: Controlling Spending and Increasing EfficiencyOECD Economic Studies No. 24 pp. 7-55.

    • Search Google Scholar
    • Export Citation

    RussellL.1986Is Prevention Better Than Cure? (Washington: The Brookings Institution).

    SchneiderMarkus and KlausJacobs1996“The Impact of Policy Options for Restructuring Sources of Health Care Financing,”draft prepared for the National Health Insurance Fund Administration of HungaryAugust23.

    • Search Google Scholar
    • Export Citation

    SZDSZ1993Szabad Demokraták Szövetsége: Egészségügyi programtervezet (Free Democrats: Health Program) draft (Budapest).

    United NationsDemographic Yearbook (New Yorkvarious issues).

    World Bank1997aWorld Development Indicators (Washington: World Bank).

    World Bank1997b“PSAL - Initiating Memorandum, Restructuring of Health Financing,”mimeo.

APPENDIX I Provider Payment Mechanisms
Type of ServiceRemuneration Method
1.Family DoctorFixed allowance per doctor plus weighted capitation
2.Inpatient Care:
a.Acute hospital care
-General ruleDRGs 1/
-Special services (kidney Dialysis, CT, MRI, cardiac surgery, etc.)Fee-for-service
b.Long-term carePer diem (with hospital-specific weights)
3. Outpatient Specialist CareBasic fee (70 percent of former year’s budget) plus capped fee-for-service (German point system)
4. Special Tasks 2/Fixed budget (historical basis)

Diagnostic Related Groups is a payment-per-case system that pays prospectively for services, on the basis of the diagnosed medical condition and its standardized treatment cost.

School health, public health nurse services, dental care, special dispensaries, etc.

Diagnostic Related Groups is a payment-per-case system that pays prospectively for services, on the basis of the diagnosed medical condition and its standardized treatment cost.

School health, public health nurse services, dental care, special dispensaries, etc.

APPENDIX II A Chronology of Reform Measures in Health Care, 1989–97
YearReform Measures
1988-Creation of the Social Insurance Fund.
1989-Private medical practice is authorized.
1990-Switch from tax-based funding to funding through compulsory insurance.
-Ownership of health facilities transferred to local governments.
-Ministry of Social Affairs and Health renamed Ministry of Social Welfare.
-Introduction of consensus management (a 3-member management team comprising of a medical director, an economic director, and a nursing director) in hospitals.
1991-Establishment of National Public Health Service (responsibility for local hygiene offices transferred from LGs to Central Government as part of this Service).
-Approval of Act on the Self-Government of Social Insurance.
1992-Social Insurance Fund separated into a Pension Fund and a Health Insurance Fund (23.5 percent health contributions—19.5 for employer and 4 percent for employee).
-Parliament defines eligibility conditions for health insurance.
-Family Physician Service is created and capitation-based payment introduced.
1993-Authorization of Voluntary Mutual Health Insurance.
-Self-Governments of Social Insurance are set up with employer and employee representation.
-Out-patient care remuneration based partly on a fee-for-service scheme, and hospital care remuneration on DRG-type scheme (July 1993).
-Share of Family Physician remuneration based on capitation increased to around 80 percent of total.
1994-The Act on the Hungarian Medical Chamber establishes ethical norms and procedures for doctors; general rules of contracts between health insurance and physicians; and the right for the Chambers to participate in the definition of health policy and legislation.
-Government adopts the National Health Promotion Strategy.
1995-New pharmaceutical regulations introduce a Basic List of essential drugs available at high HIF reimbursement rate. Reimbursement rates and the number of drugs on the Basic List are reduced twice in 1995.
-Some closure of hospital beds is undertaken.
1996-Employer contribution rate is reduced by 1.5 percent.
-Employer responsibility in sick pay is increased from 10 to 15 days, and employers are to contribute one third of sick pay expenditures of their employees.
-Reductions in reimbursement rates and changes in the drugs available in the Basic List (May).
-Adoption of Act on Hospital Capacity Reduction (July). One of its objectives is to reduce hospital beds by 10,000.
-Hospital specific weights in the DRG financing scheme are eliminated.
-Eligibility for drug card is extended.
1997-Social security base is expanded to coincide with the Personal Income Tax base. Several non-wage remunerations and incomes are included.
-Reduction of employer contribution by 3 percent; adoption of a monthly payment of Ft 1,800 for every citizen—to be paid by the employer for the employees; increase of the monthly contribution ceiling from Ft 75,000 (as set in 1992) to Ft 99,000.
-Elimination of cross-financing between the HIF and the Pension Fund.
-All public funding for Investments in the Public Health Care Sector to be allocated through the Ministry of Welfare (no investment funds will be budgeted through the LGs and the HIF).
-Reductions in reimbursement rates and changes in the drugs available in the Basic List (January).
-Adoption of Institutional Law (June) and Health Insurance Law (July) by Parliament.
APPENDIX III Areas Regulated by the draft health care Act

1. The role of the state in health care provision is to ensure equal rights to health by

  • providing health policy, including control over economic activities with adverse health effects
  • creating and enforcing legal regulations
  • organizing and planning the health care delivery system
  • defining the extent of health care provision

2. The rights and responsibilities of individuals:

  • patients have the right to equal access to health services regardless of income status, gender, race, language, religion, etc.
  • patients have the right to obtain adequate information of their health condition
  • patients have the right to self-determination, etc.

3. The health care delivery system comprises of institutions and professionals, either public or private entities, who are responsible for health promotion and disease prevention, curative care, nursing, and rehabilitation. The basic principles of the operation of the delivery system are the following:

  • emphasis on prevention
  • strengthening primary care
  • sector-neutrality

4. Professional requirements/standards will be defined in the Act in the following areas:

  • requirements/standards for starting new practices/institutions
  • personnel requirements for providing health services including education requirements and ability assessment
  • physical requirements for providing services (equipment, diagnostic services)
  • requirements related to quality assurance including supervision of operation of facilities.
APPENDIX IV Main Areas Regulated by the Health Insurance Act
Health benefits are grouped in 3 categoriesBenefits not covered by health insurance: 36 treatments, including: liver, lung, and heart transplantations, sterilization, acupuncture, etc.
Benefits provided free of charge to insurees. These include: screening, preventive, and curative services provided that utilization of the latter follows the general and specific treatment protocols (see below).
Cases when the health institution is authorized to collect co-payment from the patient. These include:
utilization without written referral
utilization of services at a higher level institution than indicated on the referral slip
utilization of services without proof of participation in screening sessions that are mandatory by ministerial decree
majority of dental care services
higher quality hotel services during inpatient treatment.
General treatment protocols for physicians regulate the utilization of services covered by the HIFWritten referrals are required for utilization of outpatient and inpatient care services, with the exception of emergency.
The referring physician is obliged to refer the patient to the lowest possible level of care where his/her condition can be treated.
Only physicians contracted by the HIF are eligible to refer patients.
Different utilization patterns are possible but the institution is then authorized to charge a co-payment.
Specification of HIF contracting procedures with health institutionsThe HIF is mandated to contract with providers for the capacity set in the 1996 Act on Capacity Reduction. Contracts exceeding that capacity require the approval of the Minister of Welfare.
The content of the contracts with providers is specified. These have to include: services to be provided by specialty, the geographical area to be served by the provider, identification data of physicians eligible to prescribe drugs, appliances utilized by contracted institution, registration number of physicians, data provision requirements, etc.
The HIF supervises the provision of services through its network of supervising physicians and financial experts. The Act specifies the documents that are to be made available for supervision (patient records, DOG accounts/charges to HIF, etc.), and the cases when the contract with providers may be terminated.
Should providers have financial difficulty threatening the safe provision of services contracted, the HIF is mandated to provide a lump sum, interest-free advance payment (i.e., loan) provided that the owner of the institution also provides own resources and guarantee of repayment.
Physicians in institutions not financed by the HIF may submit a request for authorization to prescribe drugs and medical appliances.
Follow-up regulation to be prepared by the government and the MOWGovernment
definition/revision of provider payment mechanisms and the details of related contracts
setting of pharmaceutical reimbursement rates
definition of specialist treatment accessible without written referral.
Minister of Welfare
definition of preventive and screening services to be provided
development of certification mechanism to verify attendance at mandatory screening sessions
design of protocols for the main diagnosis groups and drug prescriptions
revision of DRG tables and weights
revision of outpatient point values
waiting list regulations and ceilings on interventions provided to limited number of patients.

Prepared by Edgardo Ruggiero.


The expenditures of the HIF also include under-retirement-age disability pensions and other health-related outlays. Issues related to disability pensions were discussed in SM/96/207.


Since the late 1960s, the official state health care system has been intertwined with a semi-legal sector based on gratuity payments, effectively establishing a dual structure within the public health care system. Based mainly on specialty and on access to technology and infrastructures, this system closely reflects power structures within the medical profession and has effectively determined income structures within the health system The phenomenon of gratuity payments has been a significant factor contributing to the patients’ dissatisfaction with health services (Losonczi, 1986). Much of the resistance to reform in the sector can be traced to efforts to protect privileges gained under this “dual structure” (Ho and Orosz, 1994). The fact that patients already pay gratuities is also one of the main reasons advanced by those opposing the introduction of co-payments.


Hospital services are purchased by the HIF on the basis of payments per case, fees for service, and per diem, if the service provided is general admittance for acute care, for special treatments, and for long-term care, respectively (Appendix I). Payments per case set fees prospectively according to diagnosed medical conditions and standardized medical costs. The system used in Hungary is the Diagnostic Related Groups approach (DRGs)—which is also the most commonly used remuneration system for hospital services in the OECD countries. Fees for service pay the health service supplier according to the individual services provided. Per diem payments provide hospitals with a flat daily rate per occupied bed.


Among these, co-payments on drugs increased by 0.2 percent of GDP to 0.6 percent of GDP over the same period.


Total expenditure on health is not available on a comparable basis for central European countries. The figures on health expenditure used by the OECD (1997) and by Goldstein and others (1996) are not exactly comparable with those reported in Table 4. This is due to differences in definition, sources of data, and time of reporting.


Private expenditure on health should also include gratuity payments to health professionals. However, no official estimate of their amount exists. Ho and Orosz (1994) estimate that gratuity payments amounted to around 0.6 percent of GDP in 1993.


Co-payments on prescription drugs increased to 33 percent during the first four months of 1997.


As recently as 1993, outpatient and inpatient institutions were financed through global budgets set on an historical basis (the previous year’s budget was increased by a certain percent). General doctors and specialists were salaried until December 1991 and June 1993, respectively.


Scale considerations may make the purchaser better able to pay for the information needed to set up appropriate contracts and monitor them. Enhanced informational and monitoring capabilities may put the single purchaser into a stronger position to limit unnecessary care and enforce better quality of care. The single payer can take a broader view of health care, including an integrated strategy covering preventive and curative care. Finally, a monopsonist can put pressure on providers through contestability and yardstick competition.


A third source of cost escalation, sick pay, was brought under control in 1996 (Section C).


For example, data on waiting times and length of waiting lists for surgery and specialty visits are not available.


These were partially cleared at the end of 1996 through interest-free loans by the HIF to hospitals.


In the age group 45 to 64 years, the 1995 male mortality rate surpasses the levels of 1920–21.


Only 18 percent of deaths in Hungary could be avoided through better medical treatment. Although the death rate before one year of age is not particularly high, it is strongly related to low birthweight and premature birth (Jozan, 1994). Both of these factors can be corrected through targeted prenatal medical care and assistance.


This is partly the legacy of the pre-reform system, where health sector performance was measured principally in terms of the number of doctors and hospital beds in the system.


It is also possible that the poor health status of the Hungarian population necessitates such high turnover rates.


Chronic, or long-term, hospital care is instead remunerated on the basis of the per diem system. Although this system effectively caps hospital spending on long-term care with the monetary value of bed capacity, it also provides the supplier with an incentive to lower patient turnover and prolong the length of stay so that the more expensive early days are offset by lower-cost days later on during recuperation.


Since the establishment of the Treasury in 1996, the HIF (and the Pension Fund) pay a market interest rate for financing above the approved budget deficit.


These responsibilities have ultimately justified the HIF’s partial financing of hospital debt at end-1996.


The Government also adopted measures to control some areas of overspending in the budget of the HIF (sick pay and pharmaceutical subsidies—see Appendix II for details). The thrust of these measures was to reduce demand for sick pay by increasing the employer’s responsibility for its cost, and to reduce pharmaceutical subsidies by raising the effective co-payment rate of households.


At the time of the 1997 budget, and as part of the broader social security reform, the social security contribution base had been considerably expanded. See Appendix II and the discussion in EBS/97/61 for details. See Appendix I to the Staff Report for a comparison of the content of these health laws and the benchmarks in the program’s third review.


The Act on Social Insurance Eligibility and Financing and the Act on Health Insurance modify the basic law on Health Insurance (1975). The Act on Social Insurance Eligibility and Financing regulates the social insurance revenue of both the HIF and the Pension Fund, including the insured persons, the eligible individuals, and the contribution base and rates. This Act does not introduce any changes in the definition of insured persons, eligibility conditions, and the contribution base for the HIF, with respect to those already approved with the 1997 budget. The Health Insurance Act regulates the health benefit package that is offered to the insured.


The Act defines the maximum number of outpatient (polyclinics) consultation hours per 10,000 inhabitants by specialty, and establishes a formula for calculating the maximum number of hospital beds per county, including the share for each specialty. The number of beds for each county is determined by the sum of two numbers. The first is an ad hoc schedule of reduction, over the period 1996–2000, of the minimum number of beds that each county should have. The second is a function of bed endowments in the previous year, demographic factors, factors indicating the level of health care and social services, and the intercounty movement of patients.


The utilization of output indicators as a payment method implies that a reduction in hospital beds will not have an immediate impact on the level of expenditures, unless hospitals’ output is reduced through closures of wards and hospitals.


The latest amendments had been made in the early 1990s, when the structure of the health sector was undergoing deep changes (ownership of local hospitals to municipalities, private GPs, etc.—see above).


These are mostly cases when the established referral and utilization procedures are not followed (Appendix IV).


The follow-up legislation includes the revision of: (i) contracts and payment mechanisms for providers; (ii) pharmaceutical reimbursement rates; (iii) the DRG tables; (iv) the outpatients’ point values; and the definition of: (i) specialist treatments accessible without written referral; (ii) regulations on what preventive and screening services will be provided; (iii) a certification mechanism to verify attendance at mandatory screening sessions; (iv) protocols, including on drug prescriptions, for the main diagnosis groups; and (v) treatments provided through waiting lists.


The mention of co-payments in the law is a positive development. If introduced, they should help reduce waste in the provision of health services.


It is also debatable to what extent the HIF should have the authority to sanction over- and mis-prescription and treatment.

Other Resources Citing This Publication