56708 Setting Incentives for Health Care Providers in Serbia Cheryl Cashin, Johannes Koettl and Pia Schneider How Will the Reforms Improve Health Care? 1 Key Messages The Serbian Government wants to Serbian primary health centers (PHCs), known as Dom change the way health care providers Zdravlja or DZ, are organized either as separate entities or are paid by setting incentives for better as part of secondary care hospitals - Zdravsteni Centri. DZs provision of care. provide basic primary services while patients needing specialized primary care are referred to one of the 19 Currently, health care funds are specialized centers (Zavodi) or to one of the 120 hospitals. allocated on the basis of number of staff As part of a new decentralization strategy, all DZs are and beds at health facilities; this system becoming independent from hospitals. does not reward improvements in productivity, quality of care and health Under the planned reforms for provider payments, fund outcomes. allocations to DZs will not be linked to the inputs used The World Bank, on the request of the (beds and staff) or to the volume of services provided. Serbian Government, conducted a cost Rather, the DZs will be paid, in advance, pre-determined and efficiency study of 147 Serbian fixed per capita payments. These payments will be based on Primary Health Care facilities (DZs) the number of individuals enrolled with a DZ and the latter before the implementation of the will be expected to provide a defined set of services for the payment reforms to determine how enrolled individuals for a fixed period of time. This method capitation would affect the provision of would help shift some financial risk from the Health care. Insurance Fund (HIF) to the DZs. If a DZs expenditures The survey showed that financial are greater than the budget provided, it will be liable for the incentives may not be enough to bring difference. If the DZ shows efficiency gains and its costs about a behavioral change among DZs; are lower than the capitation budget, it will be allowed to other supporting policy changes are retain and reinvest the surplus for providing better health needed. care. The reforms are expected to improve efficiency by reducing Introduction inputs such as staff and infrastructure and increasing outputs such as the number of patients receiving preventive care. The Serbian Government plans to reform its provider This method will also help combine inputs more effectively payment system for health care by setting incentives for (for example, shift some expenditures from staff and providers to improve the quality and efficiency of care. utilities to medicines and supplies), increase preventive Funds for health care are currently allocated on the basis of services, and decrease diagnostic services. The change in the number of staff and beds at health facilities. This treatment behavior is expected to lead to increased encourages health care providers to use more staff and beds productivity, better quality of care, and better health. to define their budgets but does not reward improvements in productivity, quality of care or health outcomes. The Government also plans to make reforms in its hospital payment system and introduce case-based payments. Under 1 This Knowledge Brief summarizes the findings of the report, "Serbia: case-based payments such as Diagnosis Related Groups Baseline Survey on Cost and Efficiency in Primary Health Care Centers (DRGs), hospitals are paid the average cost of producing a before Provider Payment Reforms", The World Bank, 2009. ECA Knowledge Brief ,,case in an average hospital, which may be adjusted to Survey Findings account for regional economic conditions and include indirect costs such as teaching and capital costs. A shift The main finding from this baseline survey was that DZs from line-item budgets to case-based payment in hospitals is differ substantially in their efficiency. Although DZs expected to lead to more inpatient admissions, shorter generally work with the same level of staff, medical average length of stay, and higher patient turnover per bed, equipment and space--which are largely dictated by the which may also lead to higher hospital expenditures for the system--they produce different levels of output such as HIF. However, as capitation is expected to improve access numbers of consultations, etc. To some extent, the level of to DZ services, it may be expected also that hospitalization productivity in DZs may be affected by the age/gender rates, particularly for conditions that can be prevented or structure of the populations, particularly by the number of managed at the DZs, will decline. children in the DZ catchment area. Per capita payments may also create unintended Figure 1 shows the level of relative production efficiency consequences. For example, there may be an incentive for for DZs, which is the ratio of the total number of DZs under the capitation payment system to under-provide consultations to the maximum possible output. An services and keep costs low, or refer patients to hospitals or efficiency score of 1 indicates that the maximum possible specialists who are paid fee-for-service. These adverse output has been achieved in a DZ, so scores closer to 1 effects may be prevented to some extent with additional indicate more efficient DZs. Survey findings suggest that checks and balances in the system, such as open enrollment efficiency scores for DZs range from 0.136 to 0.866, with a in DZs, a quality monitoring system, outcome-based mean of 0.641 (median=0.640). Six DZs have efficiency bonuses for better quality compliance, or even penalties if scores below 0.40, raising concerns about their levels of DZs skimp on care for patients. Also, capitation rates are inefficiency, whereas 14 DZs report rather high levels of generally adjusted for age and gender of the population efficiency with scores above 0.80. registered with the DZs, and other geographic criteria. Some capitation formulas would include quality-based Figure 1: Ranking of Serbian DZs by Production components (for example, cancer screening rates, immunization rates) to set incentives to providers to Efficiency Score improve quality of care and prevent adverse effects such as 1 0.9 Production efficiency (% of under-provision of care. 0.8 maximum possible) 0.7 0.6 World Bank Survey on Cost and Efficiency in 0.5 0.4 Serbian DZs 0.3 0.2 0.1 In order to determine whether capitation payments will 0 affect the provision of care, the Government asked the 0 20 40 60 80 100 120 140 DZ World Bank to conduct a cost and efficiency study of 147 DZs before the implementation of the payment reforms, Source: "Serbia Baseline Survey on Cost and Efficiency in Primary with a planned follow-up survey about two years after Health Care Centers before Provider Payment Reforms". The World capitation is introduced. The purpose of this survey was to Bank, January 2009. inform the payment reforms and to establish a baseline on health sector performance--including utilization, quality, Some inefficiency is related to overstaffing in the DZs as and cost and efficiency against which the impact of the indicated by the very low population-to-physician ratio of reforms can be assessed in a follow-up survey. The only 782 people per physician in Serbia, a number that is methodology used in the baseline survey included much lower than in WHOs Europe region which has 3,500 descriptive analysis of key performance measures in PHC people per primary care physician. DZs can become more centers, as well as an econometric analysis of the current efficient by reducing their staff numbers and space used, production efficiency and cost efficiency functions in PHC without reducing the total number of visits by patients. centers. The analysis aimed to provide insights into the There are large areas of unused space in DZs that contribute current level of efficiency, as well as the determinants of the towards overall DZ inefficiency. About 50% of the DZs factors that influence efficiency. have at least some equipment that is used for relatively few diagnostic tests. The DZs mainly produce curative visits, provide relatively few preventive services and use excessive ECA Knowledge Brief laboratory tests, injections, etc. Referral rates for DZs are adjustments in the capitation payments to adjust generally low but it is not possible to determine from the for higher utility costs in DZs in mountainous data whether patients are being referred more or less than areas. necessary. The Ministry of Health (MOH) has started a There is very little variation in the cost-efficiency of DZs, review of clinical guidelines to ensure that they because DZ expenditures are largely pre-determined as are compatible with the scope of services that prices of input factors (for example, wages) and are defined will be financed by capitation, and provide on a national level. appropriate guidance to staff on laboratory tests, injections, other procedures, and referrals. Additional findings show that expenditures in DZs are dominated by personnel costs (which make up 70% of total The MOH and HIF, in collaboration with cost). This is at the expense of medicines and supplies partners, have provided extensive management which are also needed to improve the scope and quality of support to DZ managers so they will be able to DZ services. DZs are currently very constrained by their successfully respond to the incentives set by the fixed costs and thus in their ability to improve cost new per capita payment system. This efficiency, as their personnel costs are determined collaboration is ongoing and includes externally by the system. If personnel costs are excluded management training and new accounting from capitation and the HIF continues to pay for staff based systems in all DZs. on a line-item budget, then only about 30% of total DZ costs can be managed by DZs under capitation. The Government is undertaking major investment in the data systems in DZs and HIF There is currently unequal allocation of public resources for in collaboration with the World Bank and other primary care across DZs. This is likely the result of the way partners. The HIF has started to strengthen funds are currently allocated based on the number of staff reporting and analysis of key data related to DZ and other factors. performance, including population size and Once capitation has been introduced, it may be expected demographic structure, services provided, and that DZs will provide more preventive care visits to patients resource use. to reduce the need for more expensive diagnostic and curative visits; in addition, unnecessary laboratory tests and injections may also be reduced. Also, capitation may lead Further Recommendations to higher referral rates to hospitals as DZs will have the incentive to reduce their costs and hospitals paid by DRGs Financial incentives set by the capitation payment system will have an incentive to hospitalize more patients. may not be enough to trigger a behavioral change among Therefore, additional measures may be needed to make sure DZs that leads to more efficient care. Based on the that capitation does not lead to adverse effects on quality conclusion from this baseline analysis, several additional and access to care and on hospital spending. steps could support the development of capitation payment in DZs, prevent adverse effects in reaction to the financial incentives set by the payment reforms, and improve the Implementation of Provider Payment Reforms efficiency of the sector. These measures could be implemented in a phased approach with Phase 1 focusing The Government, in collaboration with health sector on the following seven steps: partners, has already started implementing several measures to prepare the sector for the planned provider payment Pool PHC funds from the HIF and other public reforms: sources, and pay all DZs a unified capitation The Government is currently reviewing rate with appropriate adjustments for cost proposals to adjust the capitation rate according variations caused by age, gender and to the age and gender of the enrolled geographic differences. population, and to include additional incentives for preventive services, such as bonus payment Include salaries in the capitation amount and to DZs that achieve an agreed level for adjust related human resources policies and childhood immunization or maternal care. laws to give more flexibility to DZs to improve Also, the Government is considering their cost efficiency. ECA Knowledge Brief Specify in referral guidelines the scope of During Phase II, additional attention could be given to the services at the PHC level and appropriate following measures: referral pattern to prevent unnecessary referrals to higher-cost hospitals. In addition, the Consider reorganizing space in DZs, moving capitation system should include measures such them to smaller buildings, or redirecting excess as a quality monitoring system, outcome-based space to other purposes. DZs could rent out bonuses and penalties for unjustified referrals. non-clinical space to private doctors, dentists, or day-care centers for individuals who need While the Serbian health insurance law already supervision such as the elderly and disabled. sets the legal frame for co-payments, the Government may consider developing a revised Hospital payment reforms such as DRGs cost-sharing policy as part of overall provider stimulate changes in hospital care such as payment reforms. shorter hospital stays. As a result, DZs and community care, as well as long-term care Assess the regulations and constraints that departments, will have to be ready to provide a affect the ability of DZs to manage their greater degree of follow-up to patients who resources more efficiently, including have been discharged from hospitals. regulations affecting the scope of service at different levels of care, public procurement Collect information on the disease profiles of laws, and the public labor laws according to the populations served, outcomes, and overall which public sector employees in DZs are still access to essential medicines in the country. on the HIF payroll. Develop quality and outcome measures that can be monitored by DZs on their own for internal Conduct an assessment of essential medicines management purposes, and at the system level that are necessary for effective PHCs, and by the HIF and MOH. consider the potential for limited financing of essential medicines within the context of capitation. Develop a cost-effective package of medical equipment that should be available at the PHC level. Consider reducing the numbers of equipment in DZs that are in the vicinity of hospitals to which patients could be referred. Examine whether more basic equipment is available that may contribute to DZ productivity (for example, blood pressure cuffs and scales) About the Authors Cheryl Cashin is a Health Economist Consultant in the Health, Nutrition and Population Network; Johannes Koettl is an Economist; and Pia Schneider is a Senior Health Economist in the Human Development Sector Unit of the Europe and Central Asia Region of the World Bank. "ECA Knowledge Brief" is a regular series of notes highlighting recent analyses, good practices and lessons learned from the development work program of the World Banks Europe and Central Asia Region http://www.worldbank.org/eca