3. Constraining public health expenditure growth.
83. Cross-country comparisons show that richer countries spend more on health care on a per capita (Figure 3.1) basis, and expenditure on health care by Belgian consumers is in line with its level of per capita income. Because of how health care is financed and various market and government failures, consumption decisions by individuals are not necessarily the only driver of spending levels. Trend growth of health expenditure per capita exceeds that of GDP per capita, as in other OECD countries, resulting in the share of health expenditure in GDP rising over time from 4% in 1970 to 7.8% in 1991 and 9.1% in 2002 according to the OECD Health Data. (1) This trend presents a great challenge to government budgets because health-care expenditure is predominantly publicly funded in most OECD countries (exceptions are Korea, Mexico and the United States), its share reaching almost three quarters of total expenditure on average (2) in all OECD countries as well as in Belgium in 2002. As a result, public spending on health care amounted to 6.6% of GDP in Belgium in 2002.
[FIGURE 3.1 OMITTED]
84. Public health expenditure growth has accelerated recently in Belgium. Whereas the average real growth rate was equal to 3.5% for the past ten years, it reached 4.3% on average during the past five years (1999-2003). In the middle of the nineties, the Belgian government introduced restraints on volumes and prices and caps on health budgets in order to keep expenditure growth under control in the run-up to the euro. As has been the experience in other countries, such measures are only effective in controlling expenditure for a limited period of time, and then give way to a period of buoyant growth as a result of catching up with unsatisfied demands. The current government has agreed to cap real growth of the public health budget at the high rate of 4.5% annually until 2007. It is very important that this period of relative budgetary slack be used to get an agreement on and, if necessary, finance for one-off measures that facilitate the introduction of efficiency-enhancing structural reforms (3), given that the pressure on expenditure growth is likely to remain strong in the future, with population ageing and technological progress both contributing towards strong growth in the demand for health care services. Moreover, analysis by the Ageing Commission has shown that the estimated costs of ageing are very sensitive to health expenditure growth rates, with continued high growth resulting in unsustainable public finances (see Chapter 2).
85. The Belgian government expects to be able to restrain growth in the medium run by continuing its policy of gradual reform aimed at raising the efficiency of the system. To achieve this goal, it relies on raising the responsibility (responsibilisation) of all actors, peer pressure on providers and control measures. Some specific recommendations with respect to these policies are summarised in Box 3.1. Radical reforms are currently not on the political agenda because, in contrast to some other countries, Belgians are satisfied with their public health-care system. It needs to be noted that the system has been successful in achieving universal and broad coverage with a relatively high degree of horizontal equity, providing quality services at a reasonable price, minimising waiting lists and offering patients a free choice of insurer and care providers.
Box 3.1. Constraining public health expenditure growth: policy recommendations Policies affecting the level of aggregate health-care spending * The Belgian government employs an aggregate budgetary cap to determine the global budget for health care expenditure and relies on regular monitoring and automatic corrective mechanisms to enforce budgetary discipline. However, in view of recurring budgetary overruns, the government should step up the frequency and nature of monitoring (1) and / or further improve the corrective mechanisms. * In order to contain spending on pharmaceuticals, the government should not reimburse the price difference between registered and generic drugs as long as they are medically equivalent to the registered drugs they substitute for that are no longer protected by current patents. This can be done by setting the reference price equal to that of the (cheapest) generic. The latter measure should not be too harmful to incentives to develop new medicines given that Belgium is also liberalising licensing policies for new drugs. In addition, the government should reassign drugs which tend to be consumed in excessive amounts to lower reimbursement categories, and periodically review its reimbursement policies in view of the medical evidence, as intended. * The authorities should be vigilant that restrictions on entry to medical school don't result in shortages of medical professionals in the longer run, when demand for their services will increase as a result of ageing, increases in per capita income, new technologies and reforms aimed at improving the performance of the system by making more use of the cheaper general practitioner services. Policies improving cost efficiency at the micro level * Benchmarking against peers is a proven technique to improve performance. The already existing process for general practitioners could be augmented by including benchmarks against evidence-based standards, and should be extended to other medical professions. Stakeholders outside of the medical profession should be included in the development of benchmarks. * With the introduction of case-related payment systems for the 26 most frequent surgical interventions, the government aims to increase the financial responsibility of hospitals. The cost of a treatment will only be fully reimbursed if it does not exceed 1.2 times the nationwide average cost. This policy could be tightened. Moreover, the move to case-related payment systems requires regular monitoring by the government to avoid cost-induced deteriorations in the quality of hospital services. * General practitioners currently open and maintain medical files for patients upon their request. Further efficiency gains could be obtained from the planned development of complete and up-to-date electronic medical files, which will be available for consultation by medical practitioners for all patients. * The government should strongly encourage patients to consult their general practitioner first as a general rule (except for emergencies) by not reimbursing medical expenses for patients not referred by their GP (gatekeeper). * More attention should be paid to the administrative burden imposed by new regulations in the healthcare sector. Policies improving access to health care and its outcomes * The implementation of the maximal medical bill, which introduces an upper-limit on annual out-of-pocket spending on healthcare by households, may offer the government some scope for selective increases in co-payments, reducing the budgetary impact of this measure. * In order to arrive at a more integrated approach towards the treatment of patients with a chronic disease or a need for long term care, the government should not only promote the role of the general practitioner as a gatekeeper, but also provide them with technical support in the form of best practice guidelines for the treatment of specific chronic diseases. (1.) In November 2004, the government decided that measures will be taken to ensure better expenditure monitoring. An audit will be conducted into the mechanisms involved in drawing up the budget, the control of spending and the evaluation of measures taken with respect to compulsory health care insurance. There will be in-depth revisions of spending control and warning light triggering procedures within sickness insurance.
Factors driving high growth
Excess supply and supply mismatches
86. There is evidence of excess supply of human resources in some segments of the health care sector. Most striking is the high density of practising physicians in Belgium -3.9 per 1 000 inhabitants-relative to the OECD average of 2.9 (Table 3.1). This can be attributed to an exceptionally high density of general practitioners according to international norms, which has continued to grow during the last decade. There is also an abundant supply of dentists and pharmacists in Belgium. To the extent that health care expenditures are shaped by the amount of installed capacity as supply may induce demand, the existing dense networks of physicians, dentists and pharmacists are a source of cost pressure.
87. On the other hand, nurses are in short supply, with their number of 5.6 per 1 000 inhabitants well below the average of 8.0 for all OECD countries (Table 3.1). The current shortage of nurses is costly because several studies have shown that there is a positive relationship between nurse staffing ratios on the one hand and the reduction in patients' mortality and medical complications on the other hand (OECD, 2004). There are different policies through which the shortage of nurses could be addressed: wage increases, improved working conditions for nurses, better education and access to continued education, campaigns to attract young people into the profession, or selective immigration. During the previous government, wages of nurses have been allowed to go up by 10% in real terms over 4 years and wage conditions for different nurses have been harmonised, as the density of nurses in institutional care for the elderly was particularly low and even declining because nurses in these institutions were lower paid than nurses in hospitals. However, as long as shortages persist, wage pressures are likely to remain strong and it may take some time for the supply of nurses to increase.
88. The Belgian authorities also see an excess supply of hospital beds in acute care and a shortage of beds in elderly care homes. There are no recent data available to support this assessment but an earlier OECD survey (OECD, 1999) found that the number of hospital beds per inhabitant and the average length of hospital stay were broadly in line with international norms, whereas the number of long-term care beds (4) in Belgium was among the lowest in the OECD. Such supply mismatches reduce the ability of the health care system always to provide treatments at the lowest cost, as elderly persons who suddenly lose their ability to live independently are admitted into hospitals for care they could receive more cheaply in an institution for elderly people. As a consequence, there are waiting lists for admission into institutions for long-term elderly care, although the extent of the problem is possibly exaggerated by the fact that some elderly people sign up early as a precaution.
Excessive consumption on healthcare services
89. Excess consumption of health services, pharmaceutical products and medical aids is a classical distortion in a public health insurance scheme, arising from asymmetry of information concerning the exact medical necessities between service providers on the one hand, and patients and their insurers on the other hand. In addition, demand is not very sensitive to price developments because consumers only pay a fraction of the total cost of a medical service, with the public insurer covering most of the costs.
90. Compared to the averages in the OECD and the EU15, Belgium is characterised by a high number of doctors' consultations (7.9 consultations per person per year in 2002, Figure 3.2). Institutional arrangements have contributed towards excessive consumption of doctors' services. First of all, physicians in Belgium are paid on a fee-for-service basis. (5) This payment system is known to give physicians an incentive to inflate volumes by providing unnecessary services and prescriptions and to make insufficient use of secondary providers (Docteur and Oxley, 2003). The incentives are aggravated by the high density of practicing physicians in the population. Secondly, general practitioners don't play a gate-keeping role and patients can visit specialists and take tests as frequently as they want at no penalty, resulting in an inappropriate use of medical resources because patients are not systematically directed towards the cheapest treatment options.
[FIGURE 3.2 OMITTED]
91. An earlier OECD Survey (OECD, 1999) showed that total expenditure on pharmaceuticals per capita (6) was also well above the average in the EU and OECD. Expenditure on pharmaceuticals has also grown more rapidly than public expenditure on health care (7) during the period 1993-2003, and its share reached 17.6% of total health care spending in 2003 (National Bank of Belgium, 2004a) The relatively high consumption of pharmaceutical products, and in particular of antibiotics, can be attributed to several factors. First of all, doctors have a general tendency to prescribe large amounts of medicine. Second, there are large differences in the prescription practices by hospitals and physicians, suggesting that some of them prescribe irrational amounts. Third, there is generally insufficient information about the health risks associated with over-consumption of medicine. Fourth, the share of generics in total pharmaceutical prescriptions is very low in Belgium, only 1.9% in 2001 whereas it reached 40% or more in the Netherlands, Germany, the UK and the US (European Generics Medicine Association). However, since then considerable effort has been made to encourage the use of generic medicine. According to a recently published study (8), the share of generic medicine rose from 1.9% of all ambulant medicines in 2001 to 10.3% at the beginning of 2004. It is expected that the share of generics in prescriptions, which is currently still lower in Belgium than in other European countries, will increase to make up 30% of the total volume of prescriptions when the new reference payment principle will be applied.
Insufficient attention to prevention
92. Lifestyle and poor dietary habits are responsible for the rapid spread of certain diseases, such as diabetes, heart diseases, cancers, and other ageing-related diseases. The dietary habits of Belgians, characterised by a rather high calories intake, a rather low protein intake, very high butter and sugar consumption, and an about average intake of fruits and vegetables relative to OECD averages, could be improved (Table 3.2). At the moment, about one third of the population is considered to be overweight and 12% obese. These numbers are slightly below the averages for the OECD and EU15 countries. If trends in other countries are anything to go by, the proportions of overweight and obese persons in the Belgian population are likely to rise further as obesity is more widespread among younger cohorts. Alcohol and tobacco consumption are around the averages for the EU15 and OECD, though considerably higher than in the US (Table 3.2).
93. In Belgium, only 0.5% of the public health care budget is currently devoted to public health care prevention, although this figure may be the result of inadequate data on resources allocated. This suggests that there is scope to improve the performance of the health care system by re-balancing expenditure more in favour of prevention. However, the division of responsibilities between the different levels of government has resulted in a misalignment of incentives. Health education and disease prevention are mainly competencies of the Communities, implying that it is financed from their budgets, whereas the federal social security system reaps most benefits from effective health campaigns and disease prevention resulting in a better health status of the population on average. There also appears to be some concern that the efforts undertaken by the different Communities are not sufficiently coordinated. It is important that the federal government and the Communities work out some arrangements for cost-sharing and a better coordination of prevention campaigns (as was successfully done for breast cancer).
Productivity growth and technological progress
94. Price increases in the medical sector tend to exceed the inflation rate for the total economy, and this difference has contributed on average almost 1 percentage point to the real growth of health care expenditure during the period 1991-2000 (National Bank of Belgium, 2001). The main factor explaining the increases in the relative prices of medical goods and services is the relatively low productivity growth in this labour-intensive sector. Technological progress and automation provide more scope for productivity gains in capital-intensive industries than in the medical sector. However, wage developments in the medical sector follow those in the total economy, implying that they outpace productivity growth in this sector and raise the cost of service provision (Baumol effect).
95. Technological progress is very important in the medical sector though. Rather than taking the form of labour-saving technological progress, it often takes the form of quality improvements and new treatments, made possible through expensive new medical procedures, products and drugs. Over time, the cost of medical technologies declines, but this is often more than offset by the rise in the demand for these treatments, which are paid for by the public insurer. Hence, the interplay between technological progress and the principle of "free, equal and high quality health care for all" exerts a strong upward pressure on the health care budget.
Consequences of ageing
96. Although spending on long-term care in nursing homes and nursing care received at home currently makes up a modest share of total spending on health care--about 9% of public spending on health (9) in 2003 (National Bank of Belgium, 2004a) and in line with international comparisons (10)--it has been growing at higher rates than public spending on health care during the period 1993-2003.
97. In addition, elderly people demand more health care: they visit their general practitioner more regularly, they are more likely to be admitted into hospital for longer stays and they consume more drugs. This trend is expected to continue in the view of population ageing, implying that its impact on health expenditure growth, which has been estimated to contribute 0.5 percentage points to average growth between 1993 and 2003, will get stronger (see estimates in Chapter 2). The projections are based on the currently observed consumption levels of the elderly relative to a person of age 35. Expenditure growth could be higher than projected if factors, such as higher unit-cost increases or reduced family care, create higher demand than currently from these age groups. The projected increase could also be lower if very elderly people in the future enjoy better health and lower levels of disability.
Policy responses: already initiated and desirable reforms
Policies affecting the level of aggregate health-care spending
98. The Belgian government employs an aggregate budgetary cap or a "growth norm" to determine the global budget for health care expenditure, complemented by budgetary targets for sub-sectors. The real growth norm was initially set at 1.5% in 1994, and revised upwards to 2.5% by the previous government in 1999 and once more to 4.5% by the current government in 2003. With the exceptions of 1994, 1997 and 2002, (11) the norm has not been respected and the budgetary overruns were usually substantial, in particular for pharmaceutical products and to a lesser extent ambulatory care. This outcome is not unusual: growth norms are only an instrument to temporarily control health care cost and need to be complemented (or replaced) by structural reforms that impact on the efficiency and effectiveness with which services are provided. Another drawback of growth norms is that providers have an incentive to spend to the limit, so generous norms are likely to generate more spending.
99. However, the introduction of a growth norm has also contributed towards a better monitoring of spending and has prompted initiatives to claw back some of the excess spending. Monitoring takes place on a quarterly basis in all sub-sectors, and corrective measures--such as the adjustment of fees and reimbursement rates--are taken when there appears to be a risk of a target overrun in one or more sub-sectors. In the sub-sectors of clinical biology and medical imaging, a system has been put in place that allows for the exact realisation of the allocated budgets, with a triggering of consolidation measures if spending exceeds its limit and measures to allocate the margins if spending remains below its limit. However, the norm allows for a certain degree of flexibility in total spending, for instance, it excludes some exceptional or specific expenditure, such as a part of the increases in salaries granted to health care personnel, and epidemics. In addition, some of the corrective measures taken do not necessarily result in a respect of the norm within the year itself because of time lags in the decision-making and implementation. For instance, an agreement with the pharmaceutical sector was concluded in 2001 which stipulates that the pharmaceutical industry will share in the burden of any budgetary overrun by paying back at most 65% of the excess spending through a turnover tax in the following year. Nevertheless, given the persistence of large budgetary overruns--the estimates for the year 2004 are around 600 million [euro]--it may be worthwhile to step up the frequency of monitoring and / or to improve the mechanism further.
Controls on wages, prices and health-care productive resources
100. Several initiatives have recently been taken with respect to pricing and reimbursement rates of pharmaceutical drugs in an effort to generate savings. In Belgium, the price of all drugs introduced on the market is set by the Ministry of Economic Affairs after consulting the Price Commission. In addition, for drugs approved for reimbursement, the price is also reviewed by a technical committee of the INAMI (Institute National d'Assurance Maladie Invalidite) and made known to the pharmaceutical firm. For reimbursement purposes, drugs are divided into six categories, with reimbursement rates of 100, 75, 50, 40, 20 and 0% respectively. Until 2001, reimbursement was based on the administrative price of the drug. For drugs with a high reimbursement rate, the system provided no incentive to buy a cheaper alternative to a brand drug. Hence, a system of reference pricing was developed in 2001, and entered into effect at the start of 2003 as the basis for reimbursement. In a number of well-defined cases, the reference price is lower than the administrative price:
--If a generic drug or a copy for the original specialty drug exists and qualifies for substitution, (12) the reference price of the latter will be 26% lower than its actual price. This difference will be raised to 30% as of 1 July 2005, after a study of the microeconomic impact.
--A drug whose active component has been reimbursed for 15 years, receives a reference price that is 14% below its current price. The reference price drops by another 2% after 17 years of reimbursement. However, the exception system for the price reduction of medicines in an innovative galenic form that are older than 15 years will become redundant when the new system of reference pricing enters into effect in July 2005.
101. In this new set up, the patient pays the fraction of the reference price that is not reimbursed as well as the full difference between the reference price and the actual price paid. This raises the out-of-pocket payment for the consumer buying the original, inducing him to prefer the generic instead. The introduction of reference pricing has also stimulated the supply of generics, which used to be very limited before 2001. As a consequence, the market share of generics increased to 10% by the beginning of 2004. International comparisons suggest that a further expansion of the market for generic drugs could be expected if inertia in prescription practices of physicians can be overcome. The reluctance to change practices can partly be explained by a lack of objective information about the quality and effectiveness of generic substitutes as a counterbalance to the information distributed by the pharmaceutical industry which has no strong interest in the promotion of generics. The government could promote the use of generics by information campaigns and by setting the reference price equal to the price of the (cheapest) generic, implying that the difference in price between registered and generic drugs will no longer be reimbursed. The latter will be achieved by a recent decision to extend the reference payment system (13) by stipulating that the presence of a generic for the active substance of a registered drug implies that all the pharmaceutical forms, dosages and packagings are included in the reference refund. To compensate the producers of registered drugs and new medicine, the government has also taken measures to liberalise licensing policies for new drugs.
102. In addition, the government has periodically reassigned drugs to a lower (and occasionally higher) reimbursement category in view of the medical evidence and in an attempt to reduce excessive consumption. It intends to review its reimbursement policies for a broad range of antibiotics, which are currently reimbursed generously and without limits because Belgians are heavy consumers of antibiotics whereas the medical evidence points to some dangers associated with over-consumption of antibiotics, such as reduced resistance.
103. Financial instruments to control the supply of hospital beds in acute care have been used for over a decade now. Public funding of hospital budgets is based on a standard per diem rate combined with overall budgets based on capacity levels, as measured by the number of recognised hospital beds. (14) As a result, the government could induce hospitals to get rid of excess beds and transform them into long-term care beds by limiting the number of recognised beds. In addition, the government--in cooperation with Regional and Local authorities--controls the capacity of hospitals through the financing of investment in this sector. Centralised investment budgets, combined with a policy which encourages cooperation and specialisation may be helpful in avoiding a wasteful duplication of expensive medical technologies and services, which is common in a system with more open competition between hospitals (Docteur and Oxley, 2003). The government intends to further strengthen the specialisation and cooperation between hospitals by programming the supply of hospital services per healthcare zone (bassin de soins).
104. Limits on entry to medical schools (humerus clausus) and on the annual accreditation of physiotherapists by INAMI are used as an instrument to reduce the density of medical professionals in Belgium. (15) The limits certainly provide savings for the education system, but its impact on the number of practising physicians may be limited in the face of larger migration flows of professionals within an enlarged European Union. In the longer run, the authorities should also be vigilant that entry restrictions don't result in future shortages when demand increases as a result of ageing and other reforms aimed at improving the performance of the system by, inter alia initiatives giving general practitioners greater responsibilities.
Policies improving cost efficiency at the micro level
Efficiency-enhancing changes in payment systems
105. Countries' experiences have revealed that simple payment methods for physicians can often provide adverse incentives (OECD Health Project, 2004). A fee-for-service payment system for providers, combined with no controls on the services actually delivered, may result in high rates of unnecessary service utilisation and rising expenditures. A capitation payment system may induce physicians to under-serve their patients, to increase the rates of referrals or to select people with fewer health risks ("cream-skimming"). In response to the shortcomings of both payment systems, some countries have moved towards more complicated payment systems that combine a fixed component (either capitation payments or a salary) with fee-for-service payments for specific interventions, possibly supplemented by caps on expenditure, control of fee levels and health-care utilisation reviews. In Belgium, the traditional fee-for-service payments for general practitioners have been complemented by fixed payments per maintained medical record on a patient's request (global medical file) and per night during which the GP is on duty to respond to medical emergencies.
106. In addition, the previous government encouraged GPs to compare their prescription behaviour with that of other colleagues on a voluntary basis. LOKs (Local Quality Groups) of GPs were created for this purpose. Each GP must seek accreditation with one LOK, and each LOK is required to hold a limited number of meetings each year to review and discuss prescription behaviour. The government could consider strengthening this peer review process by referring to external benchmarks rather than average prescription practices by also benchmarking Belgian physicians against evidence-based standards and practices. This initiative has been taken one step further by a new law (16) which allows for the possibility that a national agreement between doctors and health insurers makes the payment of certain pre-specified budgetary amounts conditional upon the achievement of well-defined performance objectives, such as restrictions on the volumes of medical services provided or on medical prescriptions. The dates on which the conditional amounts will be paid out, as well as the date by which a specific objective must be achieved, are set by the Insurance Commission and made known beforehand. The current agreement for 2004 and 2005 between GPs and insurers reflects this new law by stipulating that the release of 40 million [euro] will be conditional on the realisation of a clear break with past prescription behaviour with respect to antibiotics and medication against hypertension. A first evaluation of the extent to which a significant break with past prescription behaviour has been achieved has unfortunately been postponed from September 2004 until April 2005 because it was felt that more time was needed to give this experiment a real chance. If successful, the government should consider raising the share of performance-linked payments by extending the number of performance objectives. With this in mind, the National Commission has been asked to propose other measurable indicators of quality and performance that can be used in peer comparisons. Similar procedures are expected to be elaborated for the other medical professions. One weakness of this approach is that the National Commission is not independent from the doctors' association and the resulting indicators are likely to reflect their self-interest and may therefore not represent the best choices from the perspective of health gains, patient values and potential efficiency improvements. Inclusion of other stakeholders in the development and selection of performance indicators is needed to improve the outcomes of the process.
107. For over a decade, policy-makers have taken measures to limit the supply of hospital beds in acute care by reducing the average length of an acute-care hospital stay with some success. This process is likely to be reinforced by the introduction of case-related payment systems for the 26 most frequent surgical interventions. To be fully effective, economic theory suggests that the case-related payment needs to be set at levels that just cover the average cost of treatment by an efficient provider. This is extremely difficult to achieve in practice, and the national average (17) cost of treatment has been used as a norm rather than a fixed payment. In fact, hospitals are reimbursed for the full cost of treatment as long as it does not exceed the norm by more than 20%, but they are financially responsible for any over-run of the upper-limit. The government could consider raising the financial responsibility of hospitals by reducing the margins and by referring to external, evidence-based benchmarks. As international evidence has shown that activity-based payments provide hospitals with an incentive to increase volumes, the Belgian government (like the Austrian government) has attempted to resolve this problem by imposing an overall budget, with the budget envelope allocated to individual hospitals based on their activity levels over the budgetary period. At the same time, the evidence points to a risk that hospitals may have an incentive to lower the quality of service provision under an activity-based financing system, implying that the Belgian government should focus more on quality control mechanisms. This new financing system was introduced in 2002, and the results of a first evaluation are due in 2005. The government intends to include the prescription of medication in the cost calculation, which should help in reducing the existing large variations in prescription behaviour between hospitals. It also plans to extend this system of financing to one-day hospitalisations and physiotherapeutic treatments in long-term care institutions.
Incentives to consumers to use health-care resources more rationally
108. Belgian patients can freely choose any doctor, including any specialist they want to see at no penalty. The co-payment (ticket moderateur) equals 25% of the fee, but low-income groups benefit from a reduced co-payment. The financial incentives to see a general practitioner first are not very strong and, as a consequence, some patients consult a specialist or an emergency ward in a hospital for ailments that could be treated more cheaply by a GP or for conditions that are not very urgent. In order to combat this problem, the previous government introduced the global medical file, which gives the patient an opportunity to keep a medical record with a general practitioner of his choice at no cost to him. By doing so, the patient engages himself morally to consult his doctor first about his medical problems before seeing a specialist and benefits from a 30% reduction in the co-payment. As a compensation for keeping the file, the general practitioner receives an annual flat fee of 18 [euro] per patient from the government, as the measure is meant to be free to the patient. The next steps will be first to transform the global medical file into an electronic medical file and second, as soon as the supporting IT technologies are sufficiently developed and privacy concerns adequately dealt with, to come to an electronic exchange of information by giving all providers access to the electronic records (BE-Health project). The development of such IT projects should be a high priority as centralised access to patient's medical records would reduce the number of wasteful tests and medical complications resulting from inappropriate prescriptions, as well as facilitating the move towards a more integrated medical service.
109. The current government also intends to strengthen the financial incentives to consult the general practitioner first by changing the co-payment structure. In particular, the combination of a visit to the general practitioner in combination with a referral to a specialist or an emergency ward should cost no more to the patient than a single visit to a specialist without referral. This implies that the co-payment on a visit to a specialist should be differentiated depending on whether the visit was made after referral or not. In addition, this policy will be accompanied by measures to raise the accessibility to general practitioners "on duty". The problem with this policy, as with any policy that raises co-payments, is that the penalty in the form of a higher co-payment in some cases must be significant in order to bring about a change in behaviour. Ideally, the government should not reimburse medical expenses for patients not referred by their gatekeeper. However, the group of low-income patients, which have on average also higher health risks, is protected against excessive spending on health care by the maximum medical bill (see further) and reduced co-payments. Equity concerns would suggest that these people pay a reduced penalty, reflecting their lower purchasing power, in the event that they make inappropriate use of scarce health care resources, assuming that all barriers of access to a general practitioner have been removed for this group. At the same time, the general practitioner needs to be financially rewarded for the greater responsibilities he assumes under a gate keeping system.
Efficient use of technology
110. The rapid pace of innovation and publication of new studies result in decision makers having difficulties in keeping up with the large volume of evidence on the impact of new technologies. The federal government has created the Federal Centre of Expertise within the Ministry of Public Health in order to provide technical and scientific analysis in support of decision-making. It became operational in 2004. The centre will facilitate a more systematic use of health technology assessments, a tool which also allows for a better planning of supply. In addition, it is hoped that the accumulation of knowledge by this centre will prepare the way for evidence-based medicine, evidence-based clinical practice guidelines and the development of performance benchmarks, all of which can be used to raise efficiency in health-care delivery. However, a greater focus on quality and outcomes will create new demands for the generation of additional information by providers.
111. There is some potential for cost savings through administrative simplification. The multiplicity and rapid introduction of new standards currently imposes a heavy administrative burden on all medical professions. Moreover, the expected cost to providers of compliance with newly proposed norms and standards is rarely estimated before their implementation, a problem which has contributed to the current public funding crisis in hospitals. The current government considers the problem important enough to ask two working groups to look into possibilities for administrative simplification and to investigate the coherence and the pertinence of the whole set of existing norms, standards, procedures and legislation with respect to health care. In addition, all organisations, services and consultative bodies falling under the supervision of the federal health ministry will be asked to draw up an inventory with proposals for simplifications in administrative procedures, regulations and legal aspects. This approach has been used in other countries, for instance the Netherlands, to tackle the administrative burden on industry. In order to be effective, there needs to be some empirical estimate of the administrative burden and its distribution between the different actors, combined with a quantitative target for reductions and regular evaluations of the extent to which the set objectives are being reached. Experience in other countries as well as in Belgium has shown that lasting results can only be obtained if new legislative proposals are systematically subjected to an analysis of their expected impact on the administrative burden and if these cost estimates are included in a broader cost-benefit analysis.
Policies improving access to health care and its outcomes
Preservation of horizontal equity in access to health care
112. Horizontal equity in access to all health care services is a legitimate concern because it fosters a better health status for the weaker groups in society, improving their possibilities for participation and inclusion. The maximal medical bill, introduced by the previous government, is a measure that improves the financial accessibility to health care for all by capping annual family expenditure on health care services (18) as a function of net annual family income before taxation. Once spending on health care services has exceeded its income-dependent ceiling, the difference between the actual co-payments and the ceiling is fully reimbursed within the same year by the insurance company for low-income groups and with a delay of 2 years by the personal income tax administration for higher income groups. The introduction of the maximal medical bill may offer the government some scope for selective increases in co-payments--an option that the current government has stated that it will not pursue- because people are protected against unexpectedly large medical expenditure (large risks), but the impact of such measures will only be felt by the occasional consumer of health care services (small risks).
113. Some corrective measures appear necessary to ensure that all hospitals can continue to provide high quality services, regardless of the socio-economic characteristics of their patients. The structural under-funding of hospitals (19) has led them to increase the supplements on the per diem rate paid by patients (or their private insurance) wishing to receive care in a double room or a single room--a triple room being the standard. In addition, the Belgian law allows physicians to charge such patients supplements on the services they provide, which are subject to a splitting arrangement with the hospital. Physicians have been under pressure from the hospital management to lower their personal share in the supplements. However, not all hospitals have been equally successful in attracting additional private funding this way. Due to their location, some hospitals receive mainly patients from a less favourable socio-economic background who cannot pay for more comfort, so they resolve their budgetary problems either by contracting loans (some hospitals have high debts) or by cutting services. The government feels that this trend jeopardises the quality of service provision in some hospitals, and intends to study how it can change the distribution of public funding between hospitals such that the social environment in which a hospital operates is taken into consideration. This should be achieved at the lowest overall cost to the budget.
Integration of medical services
114. The health care system in Belgium is characterised by a high segmentation of health care markets, implying that different professionals operate as separate entities without the benefit of complete information about the patient's conditions, medical history, services provided in other settings or medication prescribed by other providers. This mode of operation is not only costly and wasteful, but it also raises the probability of medical errors. In addition, it is very unsatisfactory for elderly persons needing long-term care at home and patients with chronic conditions, who will find that the delivery of health care is often not well coordinated and communication between care professionals in different sectors insufficient. Hence, there is a need to move towards a more integrated system of health care provision. In its current plan, the government intends to give the general practitioner a pivotal role in the management and coordination of the long-term care process. The GP will also be the prime source of contact for the patient and his family. The government should also provide technical support to general practitioners in the form of best practice guidelines for the treatment of specific chronic diseases. Closer integration could in addition be facilitated by the transformation of the global medical file into an online available electronic file and IT software which allows for the secure electronic exchange of information.
115. Some of the measures that the current government is considering--larger budgets for prevention and for hospitals, higher wages for nurses to induce a supply response, the investment in IT projects and the development of expertise necessary to support decision-making and to monitor performance--do justify a temporary increase in the annual growth of health care budgets. However, the current high growth norm of 4.5% per year can only be maintained over a short period of time because it also fuels demands for wage increases in the entire sector and for more public coverage of new treatments and medications. Such measures increase spending growth in the short run and raise the costs associated with ageing, but hold no promise for savings in the medium run. As a consequence of general wage pressures and excessive consumption of pharmaceutical products, spending has grown rapidly in 2004 and another budgetary overrun of an estimated 640 million [euro] is likely to occur. This is a sign that the pace of structural reforms needs to be stepped up.
116. The structural reforms that are in the process of being implemented offer significant potential to contain expenditure growth temporarily (as they have done in other countries, for example Germany). There is still plenty of scope to increase cost savings by extending some reforms to other sub-sectors of health care and by intensifying them when initial evaluations of new policies are positive. The containment of health spending through structural reforms should be a top priority because it will help to keep the costs of ageing at moderate levels.
117. Greater use of IT should be strongly encouraged. IT can play an important role in the development of new information and methods geared at cutting waste, reducing medical errors, improving the cost-effectiveness of service provision and raising transparency. Comprehensive electronic records on the prescription behaviour and medical practices of different groups of health care providers would allow the government to track best and worst practices and make peer comparison significantly more effective. Plans to monitor health-care quality, such as developing tools like clinical practice guidelines and performance benchmarks that promote the practice of evidence-based medicine should be given a high priority because accurate and reliable information is badly needed to make payment systems more responsive to performance. Electronic medical records for patients facilitate a closer integration of medical services, which in turn improves the quality of treatment of patients with chronic conditions or the need for long-term care.
118. It is impossible to assess whether the implementation of the structural reforms described earlier will be sufficient to bring down the real growth rate in public healthcare outlays to 2.8% on average over 2008-30. In addition, the impact technological progress, which is very important in the medical sector, on public outlays is also difficult to predict, and there is a risk that it may be stronger than expected. Should public healthcare spending continue to grow at rates well above 2.8%, driving up age-related costs and jeopardising the sustainability of public finances, future governments will be faced with difficult choices. Assuming that they honour the commitment to reducing the tax burden on labour, one option would be to re-consider the public share of healthcare expenditure. Another option would consist of shifting resources from other public spending programs to public healthcare.
Table 3.1. Indicators of supply in the health sectors Per 1 000 inhabitants, 2002 (1) Practicing General Practicing physicians practitioners specialists Belgium 3.9 2.1 1.8 France 3.3 1.6 1.7 Germany 3.3 1.1 2.3 Netherlands 3.1 0.5 1.0 United Kingdom 2.1 0.6 1.5 United States 2.4 0.8 1.6 Average EU15 (2) 3.2 1.1 1.9 Average OECD (2) 2.9 0.8 1.7 Practicing Practicing Practicing dentists pharmacists nurses Belgium 0.8 1.1 5.6 France 0.7 1.1 7.2 Germany 0.8 0.6 9.9 Netherlands 0.5 0.2 12.8 United Kingdom 0.4 0.5 9.2 United States ... ... 7.9 Average EU15 (2) 0.7 0.8 8.9 Average OECD (2) 0.6 0.8 8.0 (1.) 2001 for the United States, for the Netherlands: 2001 for. Practicing specialists, pharmacists and nurses; for United Kingdom 2001 for dentists. (2.) Unweighted average of data for the last year with numbers. Source: OECD Health Data 2004, 2nd edition. Table 3.2. Indicators of lifestyle habits Dietary intake (1) Protein Butter Calories (Grammes/ (capita (capita/day) capita/day) kilos) Belgium 3 682 105.3 6.1 France 3 629 118.4 8.7 Germany 3 567 98.2 6.5 Netherlands 3 282 109.5 2.0 United Kingdom 3 368 100.7 3.3 United States 3 766 114.5 2.1 Average EU15 (5) 3 530.1 110.1 3.5 Average OECD (5) 3 378.9 103.8 3.1 Dietary intake (1) Sugar Fruits and (capita vegetables Tobacco kilos) (capita kilos) consumption (2) Belgium 44.8 224.4 29.0 France 33.5 227.2 28.6 Germany 36.0 212.3 24.3 Netherlands 42.1 219.0 34.0 United Kingdom 34.2 180.9 27.0 United States 30.0 237.9 18.4 Average EU15 (5) 34.7 238.2 27.7 Average OECD (5) 34.5 220.1 26.6 Alcohol Overweight Obese consumption (3) population (4) population (4) Belgium 9.6 32.7 11.7 France 10.5 28.1 9.4 Germany 10.4 36.3 12.9 Netherlands 9.8 35.0 10.0 United Kingdom 11.1 38.0 22.0 United States 8.3 35.1 30.6 Average EU15 (5) 10.6 34.4 12.2 Average OECD (5) 9.4 33.8 13.6 (1.) 2001. (2.) Percentage of population daily smokers 2002. 2001 for the Netherlands and United Kingdom, 2003 for Germany. (3.) Liters per capita (15+) 2002; 2001 for France and United States. (4.) Percentage of total population 2002. (5.) Unweighted average of data for the last year with numbers. Source: OECD Health Data 2004, 2nd and 3rd edition. Figure 3.2. Average number of doctors' consultations Number per capita, 2001 (1) United States 9.0 Belgium 7.8 Germany 7.3 France 6.9 Netherlands 5.8 United Kingdom 4.9 Average EU15 (2) 6.0 Average OECD (2) 6.9 Note: Table made from bar graph. (1.) 2000 for Germany and United Kingdom. (2.) Unweighted average of the last year available. Source: OECD, Health database 2004.
(1.) The Belgian Ministry of Social Affairs is currently running a pilot implementation of the System of Health Accounts (SHA). Preliminary results suggest that the data on total and public expenditure on health as currently presented in OECD Health Data might be underestimated at least by 10%. This is mainly due to methodological differences between SHA and SNA.
(2.) Unweighted average, includes all available countries.
(3.) For example, the introduction of financing on the basis of Diagnosis Related Groups would raise the efficiency in the hospital sector, but this sector currently feels that it is under-funded. Therefore, a one-off increase in public funding to hospitals may help in overcoming their resistance to the introduction of a new financing system. The government has decided to improve the funding of hospital care, more particularly by increasing the budget of hospitals by a total amount of 100 million [euro] over a period of three years. Refunding focuses on items which represent a significant objective cause of underfunding.
(4.) Belgium had 1.1 long-term care beds per 1 000 inhabitants relative to an OECD average of 4.4 in 1997, the most recent year for which there are Belgian data.
(5.) The fee schedule--the so-called nomenclature--is negotiated each year in a system of medical consultations (concertations) between representatives from the health-care purchasers (mutualites) and the medical profession. All agreements need to be approved by the government, which also has the power to set the fees in case no agreement between the bargaining parties can be reached. Fees are the basis for calculating the reimbursements, generally 75% implying that the patient pays 25% out of his pocket. The fees constitute a price floor, but general practitioners and specialists are free to charge higher prices and top-ups (paid entirely by the patient) are commonly applied by specialists.
(6.) Converted into US dollars at purchasing power parity exchange rates, per capita expenditure on pharmaceuticals equalled $306 in Belgium in 1996, compared to an average of $227 for the OECD and $230 for the EU.
(7.) Public expenditure on healthcare is defined more narrowly in this study as all expenditure incurred by the national social security office INAMI (Institut National d'Assurance Maladie Invalidite). Public spending on health care according to the national accounts will be higher as it includes spending by the federal government (co-financing of the standard cost per day of hospitalisation), regions and communities (health prevention and financing of health services benefiting persons with a disability) and the local government (providing access to healthcare to social welfare recipients). Spending by INAMI makes up 88% of total public expenditure on healthcare.
(8.) See Landsbond der Christelijke Mutualiteiten, "Naar een prijsbewuster geneesmiddelenvoorschrift? Generische geneesmiddelen in een ruimer kader", 15 juli 2004.
(9.) Again defined as total spending on healthcare by INAMI.
(10.) For example, long term care accounted for 10.7 and 11.5% of total health expenditure in respectively Germany and the Netherlands in 2002.
(11.) Expenditure was exceptionally low in 2002 because providers had accelerated their billing procedures at the end of 2001 to avoid the conversion into euro of a large number of bills in 2002.
(12.) In order to qualify for substitution and as a consequence to be eligible for reimbursement, a generic must be at least 26% (soon to become 30%) cheaper than the original.
(13.) Until now, the Belgian government has applied the reference payment principle in a strict sense: a registered drug has been classified for reimbursement according to the reference payment principle only if there exists a generic which is identical in terms of active substances, dosage, pharmaceutical form and mode of intake. Under the new system, which will enter into effect in July 2005, the presence of a generic for the same active substance will be a sufficient condition for a registered drug to become classified for reimbursement under the reference payment principle. This implies that the list of registered drugs for which a generic equivalent exists will be significantly longer and include almost all registered drugs, making the exception system redundant.
(14.) The number of recognised beds is based on the number of admissions, the mix of treatments and the length of stay for each treatment according to best practices.
(15.) It should be noted that the numerus clausus is only applied in Flanders. In the Walloon Region, another quota scheme is in operation which evaluates students after three years (the bachelor years according to the new Bologna criteria) to decide who and how many may continue their studies to become a doctor (Master in medical science).
(16.) Loi-programme du 22 decembre 2003, published in le Moniteur belge of 31 December 2003.
(17.) For each APR-DRG (All Patient Refined Diagnosis Related Group), data are taken from the national MKG-MFG databank (minimal clinical data and minimal financial data) to calculate the average cost. The costs considered are mostly related to clinical biology, imaging, and other tests performed in preparation of a surgical intervention.
(18.) Not all medical expenditure is eligible for inclusion in the maximal medical bill. Included are all co-payments and supplements paid for a consultation with a doctor, physiotherapist, nurse, para-medical specialist, the costs of technical provisions, the co-payments on drugs that qualify for (partial or total) reimbursement, the personal contribution in the per-diem rate during hospitalisation, and the personal share in endoscopic materials and materials for viscerosynthesis.
(19.) Funding shortages in the hospital sector have been estimated at around 400 million [euro] by some independent studies. However, the Federal Centre for Expertise came up with an estimate between 112 million [euro] and 295 million [euro]. Hence, the government will raise public funding by 100 million [euro] during the next 3 years.
Docteur, E. and H. Oxley (2003), "Health-Care Systems: Lessons from the Reform Experience", OECD Economics Department Working Papers, No. 374, OECD, Paris.
Economist (2004), "Survey Health Care: the Health of Nations", The Economist, July 17th 2004.
High Finance Council (2003), Studiecommissie van de vergrijzing: jaarlijks verslag, Brussels.
Ministry of Social Affairs and Health (2004a), Begroting Ziekteverzekering 2005: Een beleid in samenhang met de volksgezondheid, Brussels.
Ministry of Social Affairs and Health (2004b), De gezondheidsdialogen, Brussels.
National Bank of Belgium (2002), Rapport 2001: Evolution economique et financiere, Brussels.
National Bank of Belgium (2004), Rapport 2003: Evolution economique et financiere, Brussels.
OECD (1999), OECD Economic Surveys. Belgium/Luxembourg, OECD, Paris.
OECD (2004a), "Towards High-Performing Health Systems: Final Report on the OECD Health Project", Ad Hoc Group on the OECD Health Project, OECD, Paris.
Van Doorslaer, E. and C. Masseria (2004), "Income-related Inequality in the Use of Medical Care in 21 OECD Countries", OECD Health Working Paper, No. 14, OECD, Paris.
Wurzel, E. (2003), "Consolidating Germany's Finances: Issues in Public Sector Spending Reform", OECD Economics Department Working Papers, No. 380, OECD, Paris.