The World Bank February 15, 2008 Evidence on cost-sharing in health care: Applications to Hungary 1 Executive Summary 70039 This study has been written at the request of the Government of Hungary, and provides evidence on cost-sharing in the health sector and its application to Hungary. It presents results on the impact of cost-sharing on (i) revenues in health facilities and insurance, (ii) financial sustainability, (iii) informal payments, (iv) overall service use, and (v) equity in access. Five key- findings emerge: • Cost-sharing could lead to a reduction in unnecessary care provided to insured patients who do not have to pay the “full priceâ€? of care (reduce moral hazard). However, findings from the US and Europe also show that the effectiveness of cost sharing in reducing the demand for care depends on several factors including patients’ socio-economic and health status, the type of care, and the financial incentives set to the provider through the provider payment mechanisms. • Cost-sharing with exemption policies are a prerequisite to provide equity in access to care. Exemption mechanisms need to be transparent and easy to apply to reduce administrative costs and ensure they reach their objective. • Cost-sharing could help reduce informal payments and keep patient payments in the system. • Cost-sharing could support cost containment strategies if implemented combined with supply-side measures and financial incentives set through the provider payment system. However, in the absence of any evidence on the marginal effect of cost-sharing over supply-side measures, the impact of cost-sharing on cost containment should not be overestimated. • The experience from OECD countries suggests that successful cost-sharing policies: (i) Are transparent to all, simple and easy to understand; (ii) Result in cost-savings to the insurance or state general revenue funds; (iii) Ensure that patients share in the cost of health care; (iv) Are administratively cost-effective, feasible and practical; (v) Provide positive reinforcement for patients’ cost sharing behavior and encourage patients’ "cost-sensitivity" and "thoughtful" utilization of care Based on these findings the study recommends that Hungary continues to monitor and evaluate the impact of cost-sharing on access to identify possible negative effects on equity in service use. Household survey analysis could be conducted on equity in utilization adjusted by the need for care, and focusing on GPs, specialist and hospital care. In addition, analysis on equity in health financing could monitor the effectiveness of exemption mechanisms and detect any inequity in health financing. Telephone surveys with insured patients or patient exit surveys could be conducted of inpatients after discharge, and outpatients at the completion of treatment, with the 1 This paper was written by Pia Schneider, ECSHD, The World Bank, and peer-reviewed by Jack Langenbrunner (EASHD) and George Schieber (EASHD). It draws heavily from the background paper written by Tunde Szabo. 1 The World Bank February 15, 2008 goal of eliciting feedback on informal and formal payments, service satisfaction, and other factors. The Hungarian Government has recently introduced substantial supply-side measures to reduce overcapacity, including closure of hospitals and beds. Although cost-sharing could influence prices and utilization of care, the experience from several countries shows that reducing hospital overcapacity will have a stronger impact on total healthcare cost than demand-side measures. In Hungary, financial incentives through the provider payment system could eventually support the effect of the Government supply-side strategy. Such financial incentives could be set to providers through capitation payment to reward better quality of care, and Diagnosis Related Group (DRG) payments with expenditure ceilings with strict utilization review and quality assurance control to set an incentive for efficient provision of care. 2 The World Bank February 15, 2008 1. Introduction In February 2007, the Hungarian Government implemented a cost sharing policy requiring direct payments made by patients for outpatient visits, hospitalization and emergency visits. Through cost-sharing the Government aims to support the following objectives: cost containment through moderation of service use; revenue raising; formalizing informal payments and making individuals responsible for their health. However, in many countries worldwide, the introduction of cost-sharing has caused critics, arguing that cost-sharing leads to inequity in access and is are not an effective tool to control costs and suppress informal payments but rather creates an additional administrative burden to healthcare providers. This paper has been written at the request of the Government of Hungary, and provides evidence on cost-sharing in the health sector and its application to Hungary. Cost sharing refers to out-of-pocket payments made by patients to the providers at the time of service use. For patients with health insurance there are three main forms of cost sharing: • deductible: amount that must be paid out-of-pocket before benefits of the insurance become active; • co-payment: flat amount that the insured patient must pay for each service used; • co-insurance: percentage of the total charges for a service that must be paid by the beneficiary. Other policies that are frequently associated with cost sharing mechanisms include benefit maximums, out-of-pocket maximums, extra billings, pharmaceutical reference pricing and coverage exclusions. All these options aim to influence patients’ care seeking behavior. These demand-side policies are often introduced based on the moral hazard argument, which posits that individuals with health insurance will overuse health services if they bear no portion of the financial burden. Overuse implies that the benefits are less than the risks and costs, which then could result in unnecessary cost increases. The purpose of this study is to present information on cost-sharing arrangements and the level of out-of-pocket (OOP) payments in health facilities in Europe and the US2. The analysis focuses on the impact of OOP on (i) revenues in health facilities and insurance, (ii) financial sustainability, (iii) informal payments, (iv) overall service use, and (v) equity in access. Based on findings, the report provides policy advice on the definition, level and management of out-of-pocket payments to ensure health policy objectives including equity in access and financial sustainability. 2. Cost sharing for healthcare in Hungary Patients already finance one-fourth of total health expenditures before “official cost- sharingâ€? was introduced. Total heath expenditures (THE) in Hungary are about 8.5% of GDP, similar to other EU countries. Informal payments (“gratitudeâ€?) by patients are estimated at 3% of THE, and it was hoped that cost-sharing would help in “formalizingâ€? informal payments. Even before official cost-sharing was introduced, Hungary reported higher private spending than 2 Data were collected from Austria, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Slovakia, Slovenia, Spain, Sweden, Switzerland, United Kingdom, Norway, Slovakia, Czech Republic, Poland, Croatia, Bulgaria, Romania, Estonia, Latvia, Lithuania, Albania, Kyrgyzstan and Cambodia. 3 The World Bank February 15, 2008 several European countries that have cost-sharing policies in place. In 2004, Hungary reported similar values for private spending as a share of THE as the EU average (25%). Private sources include 90% direct payments by patients (co payments for pharmaceuticals, user fees to private sector providers, and informal payments), and 7% payments by enterprises. Almost 43% of private spending goes towards medicines and medical appliances and 39% for outpatient care (Figure 1). Within outpatient care, the largest care item financed from private sources is outpatient dental care: in 1998 62.8% and in 2001 58.9% of outpatient care was spent on dental care services, including prosthetics. About 11% of private expenditures went to inpatient care. Figure 1: Private health expenditures, 1998 and 2001 (before official cost-sharing) Source: Hungary National Health Accounts, 2001. OECD High service use is related to demand and supply-side incentives. In 2005, prior to the introduction of co-payments, Hungary reported highest rates for hospital admissions in the EU8 countries and high outpatient visit rates (Figure 2 and 3). In the absence of household survey data, it is not known which socio-economic groups are the beneficiaries of this high service use and whether this utilization level contributes to inequity in access. However, high use rates point to moral hazard behavior among insured patients who do not pay the full price at the time of service use. In addition, several supply-side factors in health also induce a higher demand for care. A relatively high number of hospitals and hospital beds combined with Diagnosis Related Group (DRG) payments set an incentive to hospitals to increase the number of inpatient admissions. Outpatient providers who are paid fee-for-service have an incentive to see more patients. Hence, these supplier-induced demand increases require corrective measures on the delivery and provider payment side, while cost-sharing policies focus on the demand-side and influence patients’ care- seeking behavior. 4 The World Bank February 15, 2008 Figure 2: Inpatient admits per 100 population Figure 3: Outpatient visit per capita 30 20 25 20 15 15 10 10 5 5 0 0 Lithua De n m Italy Swed Croat Polan Slove Slova Eston Bulga Czech Latvia Roma Hugar Sw e Port Latv NL Bulg Ru m Slov Hung Slov Cze c Finl a Esto Croa den ugal ia enia aki a a ria a nia nia hR tia nd en ia ary nia kia d ark ia ria nia y nia R Source: WHO health for all. 2005 Hungary introduced cost-sharing in February 2007. Patients co-pay €1.2 per outpatient visit, and €1.2 per hospital day (“vizitdíjâ€?), leaving the remaining part to be paid by insurance. In addition, user fees are charged for services excluded from the national insurance benefit package, such as medicines and medical aids, and medical care provided in private sector. The Health Care Act defines cost-sharing as a financial support to providers from the health insurance fund. Co-payments and user fees are paid by patients when receiving healthcare services unless official exemptions are applied (Table 1). The responsibility for collection is with the provider. Based on Government recommendations, providers who collect OOP may use at least 30% of these funds to increase salaries of health workers or pay salary premiums. Table 1: Fees introduced and effective since 15 February 2007 Type of Care Cost-sharing paid by patient Primary outpatient care 300 HUF (€1.20) per visit Specialized outpatient care 300 or 600 HUF (€1.20 or €2.40) per visit (increased to 600 HUF at specialists if patient has no referral from GP) Hospital (i.e. room and board) 300 HUF (€1,2) per day (max 20 days) Emergency service 1000 HUF (€3,9) per visit Source: Hungary Ministry of Health, 2008 Like other countries, Hungary has explicit exemption criteria for co-payment. Population groups and services exempt from fees include (i) children and adolescents under 18 years of age; (ii) homeless people; (iii) emergency services as defined in a separate law (the lack of attending the patient in emergency status involves threat of death or possibility of permanent damage to health); (iv) preventive services, public health services, pregnancy, obstetric and neonatal care services. Although, it is a bit early to draw conclusions on the impact of cost-sharing, four short-term findings emerge and suggest that the policy could have the expected effect: (i) Cost-sharing may have a moderating effect on utilization. From January until August 2007, ambulatory care encounters decreased by 23% and hospitalization by 29% compared to the previous year. In 2007, the number of prescriptions decreased by1 million compared to 2006. 5 The World Bank February 15, 2008 (ii) Cost-sharing appears to support financial sustainability. Hungary has a long tradition of health financing deficit. In 2007, providers raised HUF 21.7 billion from cost-sharing reflecting about 1.3% of the total health insurance budget. Almost half of this amount was raised by GPs3. For the first time, the Health Insurance Fund closed the year with a surplus in the amount of HUF 28.2 billion. (iii) Cost-sharing may have a positive impact on the MOH cost containment strategies. Together with other cost-containment measures (e.g. hospital restructuring) implemented by the MOH, co-payments led to a reduction in hospital costs by HUF 6 billion compared to the budget. (iv) Cost-sharing seems to have a positive impact on provider income, quality of care and staff satisfaction. Providers used the revenues from user fees to pay for health worker salary increase which resulted in a 25% increase in the average income for family doctors and a 17% increase for pediatricians. At the same time family doctors saw 25% fewer patients resulting in better working conditions for health staff and more time spent per patient. 3. Cost-sharing policies in other countries Western European countries introduced cost-sharing with the objective of reducing unnecessary demand for services. The majority of Western European countries employ some form of cost sharing (Annex Table 1 provides an overview). In the three tax-funded health systems Denmark, Spain and the UK, access to inpatient and outpatient care is free for patients, but patients are charged a fee for pharmaceuticals and for the use of specialist care if used without referral from a GP gatekeeper. The most common forms of cost sharing in countries with health insurance are co-payments and co-insurance. Only insurance companies in Switzerland and the Netherlands charge deductibles before insurance coverage kicks in. Some insurers also charge different levels of prices mainly to set financial incentives to patients and direct them towards preferred healthcare providers (e.g. providers contracted at a lower rate). Generally, patients have to pay a higher fee in case of self-referral directly to the specialist without seeing a GP. All countries with cost sharing also protect vulnerable groups from inequity in access to care, mainly with exemption policies and co-payment ceilings. Central and Eastern European (CEE) countries have introduced cost-sharing policies to raise additional revenues for the health sector and to formalize informal payments. In transition economies providers have been relying on informal payments from patients which often have affected the quality of services and led to inequity in service use. Poor patients did not seek care if they didn’t know whether they have enough money to pay informally. Co-payments are thus seen as a measure to formalize informal payments and keep revenues from patients within the system. Explicit and transparent prices make the level of OOP expenses at the point of service use predictable for the patients. Most CEE countries have some form of cost-sharing. Five countries introduced similar cost sharing mechanisms mostly in form of flat user fees charged to the patient by the provider – per ambulatory visit, per hospitalization day and per emergency visit (Table 2). In all countries, co- 3 Source: MoH website. The publication is from August 2007. Results are based on 10 months of cost- sharing experience. 6 The World Bank February 15, 2008 payments are charged for diagnostics, pharmaceuticals, and some medical materials. In Bulgaria, fee levels were set at the equivalent of 1 percent of the minimum salary. Slovakia implemented user fees in 2003 which were abolished in 2006, but patients still have to co-pay for pharmaceuticals. The Czech Republic just implemented user fees on 1 January 2008. Some countries apply a stop-loss clause with an annual ceiling (e.g., Czech Republic). All countries use some form of exemption measures for low income persons and other vulnerable groups. Table 2: Current Co-payment practice in Central and Eastern Europe Country Ambulatory physician Hospital Emergency Comments Exchange (per day of services rate (per visit) hospitalization) (per visit) (1 €= ) Slovakia 0 SKK 0 SKK 60 SKK Fees were introduced 34 SKK (June 2003 -Sept 2006 (June 2003 - Sept 2006 the from June 2003 until the fee was 20 SKK) fee was 50 SKK) September 2006. Hungary 300 HUF 300 HUF 1000 HUF Fees introduced on 15 251 HUF (Increased to 600 HUF (max 20 days) February 2007 at specialists if patient has no referral) Czech 30 CZK 60 CZK 90 CZK Fees introduced 1/1/’08. 28 CZK Republic Annual limit of 5000 CZK per person per year. Poland 0 PLN 0 PLN 0 PLN No plans to introduce fees 3,89 PLN Croatia 10 HRK 50 HRK User fees were 7,34 HRK (max 30 HRK per (total fee 150 HRK, but 100 implemented in 2005 month) HRK is paid by the insurance company) Slovenia 0€ 0€ 0 € Payment for prescriptions Bulgaria 1,8 BGN 3,6 BGN 0 BGN User fees implemented in 1,95 BGN (2007: 1% of min (max 10 days) 1999 salary) (2007: 2% of min salary) Romania 0 RON 0 RON No plans to introduce fees 3,38 RON Estonia 50 EEK 25 EEK User fees implemented in 15,65 EEK (max 10 days) 2002 Latvia 0 LTL 0 LTL 0 LTL 3,46 LTL Lithuania 0,5 LVL 1,5 LVL User fees implemented in 0,71 LVL (home visit + 5 LVL per admission 1999 2 LVL) (max 25 LVL per one hospitalization) Source: Health Policy Institute, 2007 based on a survey conducted with each country Methodological comments: 1. In some countries, some population groups are exempt from charging out-of-pocket payments (e.g., children, pensioners, chronically ill, expectant women) “Free careâ€? does not protect patients against high private health expenditures. In 2004, several CEE countries – among them Slovenia, Latvia, Hungary, Czech Republic – did not have cost-sharing for inpatient and outpatient care and their insurance policy provides comprehensive coverage, or supplementary insurance for co-payment as in the case of Slovenia. Despite this “free careâ€? several of these countries report substantial private health expenditures as expressed by the share of OOP in % of total health expenditures (Figure 4), and total private health spending per capita per year in PPP$ (Figure 5). Without cost-sharing, Latvia, Romania, Poland and Hungary report an out-of-pocket share that is above the European average (which includes Western Europe with cost-sharing countries). Similarly, Slovenia, Hungary and Latvia report highest private spending per capita, despite “free careâ€?. This experience shows that patients have 7 The World Bank February 15, 2008 to pay for services and pharmaceuticals and seek care in the private sector where prices are higher and services not necessarily covered by health insurance. Figure 4: Out-of-pocket payments in % of Figure 5: Private health spending PPP$ per total health expenditures, 2004 capita, 2004 42 43 45 500 429 443 40 32 370 371 35 28 400 30 21 24 25 25 300 255 279 285 25 18 19 181 210 20 200 147 153 174 15 10 10 10 100 5 0 0 er y e y ov ia Es kia La ia ia Es atia La ia ro Hu tvia th ia ul i a ze a th ia lg a C hR C hR om d ov d ze a pe n a Po ge Po nia ov e ia av gar av ar Sl rag Bu aki C eni C ani ro Hu ani R lan Sl lan tv Sl at r ar Li ton B an Li ton en a ga a pe n g ua c c ro ro om u ov Sl R Eu Eu Source: WHO. http://data.euro.who.int/hfadb/ Private health expenditures in form of out-of-pocket expenditures may result in inequity in health financing (regressivity). Wagstaff et al (1999) measure inequity in health financing using household survey data from early 1990s. Table 3 presents results based on the Kakwani index for four financing sources and selected European countries. Findings suggest that indirect taxation and out-of-pocket payments are regressive in all countries. Health financing is progressive under direct taxation. Social health insurance (SHI) is regressive with the exception of two countries (France and Switzerland) where SHI is progressive mainly because all income groups are insured and low-income groups are exempt from paying contributions. Social insurance is regressive in Germany since high-income groups are exempt from SHI and enroll in private insurance. In France private insurance is regressive due to a higher enrolment among high-income groups4 who purchase supplementary private insurance to insure against co- payments under SHI. Similarly, in Switzerland higher income groups are more likely to purchase complementary private insurance to cover luxury care. Table 3: Equity in health financing Financing UK ‘93 Spain ‘90 France ‘89 Germany ‘89 Switzerl‘92 Direct taxes progressive progressive progressive progressive progressive Indirect taxes regressive regressive regressive regressive regressive SHI progressive regressive progressive Private progressive regressive regressive progressive regressive insurance OOP regressive regressive regressive regressive regressive Source: Wagstaff A, et al. (1999): Equity in the finance of health care: some further international comparisons. Journal of Health Economics; (18):263-290. 4 Buchmueller, et al (2004) find that private insurance enrolment tends to be higher for managers and highly educated professionals than for semi-skilled and unskilled workers, and is lowest for the unemployed. 8 The World Bank February 15, 2008 4. Arguments and empirical evidence about cost-sharing Cost-sharing tends to be criticized based on the following five arguments. For each argument supporting or refuting evidence is presented. Argument 1: Cost-sharing for health care helps to decrease an overuse of health care by insured patients (moral hazard argument) • Based on US data, Newhouse and Phelps (1976) found no substantial impact on the length of hospital stay in response to changes in prices charged to patients. A 10% price increase reduced the average length of stay by 1.7%. • In Belgium, van de Voorde at al. (2001) exploited the price variation generated by a substantial increase in co-payment rates (nearly 50%) in 1994 to estimate the impact on demand for care. They find that a co-payment change has the strongest impact on the number of home visits. In the general population a 10-percent price increase results in a reduction of GP home visits by 3.4%; GP office visits decreased by 1.4% and specialist visits by 1% only. Price increases appear to have little impact on service use by the elderly and disabled. • In Germany, Winkelmann (2004) evaluated the price sensitivity of demand for physicians’ services following a 200% increase in co-payments for prescription drugs in 1997. Since all prescriptions are issued by physicians, the number of visits and the demand for prescription drugs are linked. Results show that increasing co-payments for pharmaceuticals prescribed during a visit caused a reduction in the number of physician visits by about 10 percent. • In Slovakia, Pazitný and Zajac (2005) found based on insurance data that in the second half of 2003, following the introduction of cost-sharing, there was a 10% reduction in the number of outpatient visits compared to the same period in 2002. Similarly, the number of first-aid visits dropped by 13%. However, specialized outpatient-care doctors and hospitals saw only a slight decline (2%, respectively). Only 1.5% of individuals interviewed in a survey stated the fees prompted them to stop seeing their doctors. Cost- sharing mainly caused visits for unnecessary care to drop as indicated by 18% of respondents. This suggests that co-payments did have a moderating impact on service use but were not high enough to deter patients who needed care from service use. • In the Netherlands in 2006 about 5% of the insured indicated that deductibles (individuals pay € 255 OOP before insurance coverage kicks in) had restricted their healthcare consumption. These were mainly younger and healthy persons suggesting that paying deductibles did not have a negative impact on the use of necessary care (Goudriaan et al., 2007; De Jong et al., 2006; Groenewegen & De Jong, 2004). Conclusion: Findings from selected OECD countries suggest that cost-sharing can help reducing the utilization of healthcare services that are less urgent. Results also indicate that cost-sharing with exemption policies does not appear to influence the service use of more seriously ill patients as well as specialist and hospital care. 9 The World Bank February 15, 2008 Argument 2: Cost-sharing can help contain the growth of health care costs and thereby contribute to financial sustainability. • In Switzerland, in January 2006, co-insurance rates for original brand drugs with a generic product were set at 20% whereas the rate for the generic drug remained at 10% of the price. Prescribers are obligated to inform patients about different co-insurance rates. Exemptions exist for patients who due to medical reasons (e.g. side-effects) can not take the generic product. Schuetz (2006) identified two effects as a result of different co- payment price for drugs. First, patients used generics instead of brand drug which caused the market share for generics to increase from 19.6% in 2005 to 32.7% in 2006; resulting in a 46% revenue increase for generic drugs while total revenue for branded drugs with expired patent went down by 49%. Second, as a reaction to the co-payment policy, pharmaceutical companies reduced prices for brand drugs leading to a decrease in the average, weighted price difference between substitutable brand and generic drugs from 51% in 2005 to 28% in 20065. • In Slovakia, introducing cost-sharing and the resulting decline in patient-doctor encounters had two positive effects: first, the quality of provided health care improved because doctors spent more time with each patient; and second, the volume of prescribed drugs reduced. Co-payments for prescriptions prompted 20% of the patients to ask for fewer prescriptions (Pazitný and Zajac, 2005). • In Slovakia, abolishing co-payments for hospital and outpatient care in September 2006 led to a shortfall in revenue for providers, which providers tried to compensate by negotiating higher prices with insurance companies. In January 2008, the largest insurer agreed to a 30%-40% price increase per inpatient admission paid to state-hospitals and a 7% increase in the capitation rate paid to GPs, which will be followed by another 5% increase in April 2008. As a result health expenditures by insurers are expected to increase, as is the number of patients on the waiting list (World Bank, 2008). Conclusion: There is limited evidence suggesting that combined with supply-side and provider payment reforms cost-sharing could support the effectiveness of a cost containment strategy. However, little is known about how significant the marginal impact of cost-sharing actually is when combined with other strategies. Argument 3: Cost-sharing leads to inequity in access to care with low-income groups being excluded or paying a higher share of their income for health care than the rich • A study including nearly 80 countries (WHO 2006) suggests that countries with a higher share of OOP in percent of total health expenditures also have a high percentage of families that face catastrophic health spending (Interpharma, 2007).6. However, catastrophic spending due to OOP was highest in poorer countries such as Vietnam (10.5%) and in Latin American countries, but not in Europe. In 2003, catastrophic payments were at a negligible level in Hungary (0.2%), as well as in Slovakia and the Czech Republic (0.1%). 5 Weighted price reductions are calculated top down. 6 Expenditure is usually defined as being catastrophic if a household’s financial contributions to the health system exceed a determined percentage i.e. 40% of income remaining after subsistence needs are met. 10 The World Bank February 15, 2008 • In Europe, exemption policies have protected low-income groups against the negative impact of cost-sharing on equity in access. Based on 1996 household survey data, and after adjusting for need of care, van Doorslaer et al. (2000) find in all OECD countries a fairly equal distribution of total physician visits across income quintiles when adjusted by the need for care. However, there is evidence that the rich report higher use of specialist care than what would be expected on the basis of their need for care. This finding was confirmed based on data from 2000, when several countries reported a significant pro- poor GP visit distribution, but the rich still have a significantly higher mean number of specialist visits per year, than the poor. The exceptions were the UK and Netherlands, where specialist use is fairly equal. Leu and Schellhorn (2004) find similar results based on Swiss household surveys from 1982-2002. Service use for GPs and hospitals is distributed equally across income groups, whereas the distribution of specialist visits is significantly pro-rich despite co-payments. These findings support evidence suggesting that the wealthier are less responsive to co-payment. Considering that the rich are more likely to report “unnecessaryâ€? specialist care, countries might want to explore income- dependent visit fees for specialists. • In Bulgaria, household survey findings suggest that a majority of respondents accept payment in public health facilities provided that services are of good quality and access was quick. Respondents did not feel worse equity in access than before co-payment were introduced. The majority preferred flat-rate cost-sharing over co-insurance (percentage of actual service price). Nearly all considered a ceiling on payments appropriate, and strongly supported exemption mechanisms from payments. (Pavlova et al., 2002) • Some countries have introduced supplementary insurance to insure co-payment among them France, Slovenia and Croatia. However, supplementary insurance did not have the expected effect on equity and financial sustainability, which is one of the reason why it is forbidden in Switzerland7 Evidence shows that supplementary insurance to cover co- payment is generally purchased by those who use a high amount of health care (Tapay, 2001). The resulting adverse selection endangers the financial sustainability of insurance. In Slovenia and Croatia, supplementary insurance run into financial sustainability problems caused by adverse selection of higher risk-groups into insurance8. In France supplementary insurance negatively affects equity in health, as lower income groups are more likely to be uninsured; and when insured, they use care less often than the rich (Buchmueller et al. 2004). • There is only limited evidence about the effect of cost sharing on drug therapy compliance and, as a consequence, on health outcomes, measured through indicators of hospitalization and mortality. Not complying with medication, possibly because of affordability, could have serious consequences for health. This is suggested by surveys from the US and Italy where higher co-payments for drugs had a negative effect for low- income groups on drug compliance, and could have caused higher morbidity, emergency care admissions and mortality (Atella et al. 2005). Such adverse effects of cost-sharing could be prevented by defining clear and transparent exemption policies that protect vulnerable groups from being excluded from care. 7 See Art 64 in the Swiss health insurance law (KVG) 8 See: Stefan Skledar, Institute of Macroeconomic Analysis and Development. European industrial observations observatory. http://www.eiro.eurofound.eu.int/2004/12/feature/si0412302f.html; Miroslav Mastilica, Sanja Kusec: Croatian health care system in transition from a user point of view. In: BMJ VOLUME 331 23 JULY 2005 11 The World Bank February 15, 2008 Conclusion: Several OECD surveys indicate that cost-sharing with exemptions policies can provide equity in access for physician visits and prevent adverse effects on health outcomes. However, these surveys also suggest that despite cost-sharing the better-off are more likely to seek unnecessary specialist care, suggesting as income increases, co-payments have less effect on limiting overuse of care. Higher income groups appear to be more likely to “shortcutâ€? the GP gatekeeper and see a specialist directly resulting in higher fees. Therefore, countries could eventually explore income-dependent co-payment levels for specialist care and ensure exemption policies are implemented (sliding scale co-payments). Evidence from France, Slovenia and Croatia suggests that supplementary insurance to cover co-payments does not improve equity in access to care; rather it poses a financial risk for the system by generating an incentive for adverse selection and neutralizing the moderating effect of co-payments on service use. Argument 4: Cost-sharing is a means to formalize informal payments charged by providers and combat corruption. • In Slovakia, cost sharing has successfully reduced under-the-table payments. According to a study by Pazitný and Zajac (2005), in December 2002 (before cost-sharing), 32% of respondents considered corruption the most serious problem for the country’s health-care system. By January 2004 (after cost-sharing), only 10% of respondents still viewed corruption as the most serious problem. People do not see a need to give gifts or bribes to health personnel after they have already paid a legal user fee. • In Albania the co-payment policy lacks clarity and transparency. There is insufficient information about official price and exemption policies. As a result, both insured and uninsured patients continue to make informal payments (Vian et al., 2004). • Several low-income countries report that introducing a formal co-payment policy has helped combating corruption and retaining health funds in the system, and even improved access to care9. Conclusion: As findings from Slovakia and Albania indicate, cost-sharing can help reduce informal payments and corruption and retain health funds within the healthcare system, as long as cost-sharing policies are transparent and widely disseminated to all stakeholders. 9 In the Kyrgz Republic the government lowered informal fees by introducing a formal co-payment policy, which includes defining and publishing official price lists for drugs and services and implementing sound financial management in all health facilities (Kutzin, et al. 2002). In Cambodia (Barber et al., 2004) the Takeo Referral Hospital formalized informal payments through explicit user fees and implemented a resource management system with a transparent official fee system designed to generate revenue to cover a proportion of operational costs. Once fees were official, utilization levels increased by more than 50% for inpatient and surgical services, and cost recovery from user fees averaged 33%. 12 The World Bank February 15, 2008 Argument 5: Cost-sharing creates an additional administrative burden for providers. • A small, clearly delineated, across-the-board co-payment may not be perceived as a significant administrative burden for providers and their staff. However, some co- payment systems could increase administrative burden. For example, targeted and tiered co-pays could be confusing to patients and providers (Hopkins et al, 1975). • Co-pays could reduce the administrative costs of claims handling (invoices to insurer) for providers and insurers because fewer claims would be generated (due to lower utilization). Conclusion: Transparent and simple co-payments easy to understand for all will not increase providers’ administrative burden. Rather, a reduction in service can lead to fewer invoices sent to insurers for processing; thus, co-payments could actually reduce administrative costs for providers and insurers. 5. Conclusion and Recommendations This overview on cost-sharing for insured patients provides five key-findings. • Cost-sharing could lead to a reduction in unnecessary care provided to insured patients who do not have to pay the “full priceâ€? of care (reduce moral hazard). However, findings from the US and Europe also show that the effectiveness of cost sharing in reducing the demand for care depends on several factors including patients’ socio-economic and health status, the type of care, and the financial incentives set to the provider through the provider payment mechanisms. • Findings from OECD household surveys suggest that cost-sharing with exemption policies are a prerequisite to provide equity in access to care. As the experience from Albania shows, exemption mechanisms need to be transparent and easy to apply to reduce administrative costs and ensure they reach their objective. • Slovakia and several low-income countries indicate that cost-sharing helps reduce informal payments and keeps patient payments in the system. In poorer countries, formalizing informal payments can even contribute to the financial sustainability of hospitals. • Cost-sharing could support cost containment strategies if implemented combined with supply-side measures and financial incentives set through the provider payment system. However, in the absence of any evidence on the marginal effect of cost-sharing over supply-side measures, the impact of cost-sharing on cost containment should not be overestimated. • The experience from OECD countries suggests that successful cost-sharing policies: (vi) Are transparent to all, simple and easy to understand; (vii) Result in cost-savings to the insurance or state general revenue funds; (viii) Ensure that patients share in the cost of health care; (ix) Are administratively cost-effective, feasible and practical; 13 The World Bank February 15, 2008 (x) Provide positive reinforcement for patients’ cost sharing behavior and encourage patients’ "cost-sensitivity" and "thoughtful" utilization of care These findings have implications for Hungary’s cost-sharing policy. Preliminary results from Hungary based on data from the first months suggest that cost-sharing could have an effect in the expected directions, including reduction of overuse of care, better revenue management, and cost containment. While these results are encouraging, it is recommended that Hungary continues to monitor and evaluate the impact of cost-sharing on access to identify possible negative effects on equity in service use. Following the OECD example, household survey analysis could be conducted on equity in utilization adjusted by the need for care, and focusing on GPs, specialist and hospital care. In addition, analysis on equity in health financing could monitor the effectiveness of exemption mechanisms and detect any regressivity in health financing. As this is the case in OECD countries, telephone surveys with insured patients or patient exit surveys could be conducted of inpatients after discharge, and outpatients at the completion of treatment, with the goal of eliciting feedback on informal and formal payments, service satisfaction, and other factors. In the longer-run, Hungary could eventually follow the example of Switzerland and the Netherlands, and consider implementing deductibles for insured care. Hungary might also want to consider exploring with different co-insurance rates for generic and brand-name drugs to stir the demand to cost-effective service delivery. Similarly as Canada, Australia and Switzerland, supplementary health insurance for co-payments could be prohibited by law to prevent inequity in access and financial sustainability problems caused by adverse selection. However, Hungary may eventually wish to explore with complementary health insurance to cover services currently excluded from the basic benefit package or “luxuryâ€? services (e.g. 1-bedroom hospital room). Given the absence of research on the marginal effect of different cost-containment strategies, the impact of cost-sharing on healthcare cost in Hungary should probably not be overestimated. The Hungarian Government has recently introduced substantial supply-side measures to reduce overcapacity, including closure of hospitals and beds. Although cost-sharing could influence prices and utilization of care, the experience from several countries shows that reducing hospital overcapacity will have a stronger impact on total healthcare cost. In Hungary, financial incentives through the provider payment system could eventually support the effect of the Government supply-side strategy. Such financial incentives could be set to providers through capitation payment to reward better quality of care, and DRG payments with expenditure ceilings with strict utilization review and quality assurance control to set an incentive for efficient provision of care. 14 The World Bank February 15, 2008 Annex Table 1: Current cost sharing practices in Western Europe Country Ambulatory Hospital Emergency services Other services Exemption groups Comments physician (per day of (per visit) (per visit) hospitalization) Austria 0 to 20% of tariff Appr. €10 per day up to same as ambulatory Prescription drugs: Exemptions due to socio- rate, dependent on maximum 28 days per visit and hospitalization €4,80 per pack (from economic conditions are social insurance year, variations across day resp. 2008: up to a applied by all social agency social insurance agency maximum of 2% of insurance agencies with and Federal State; yearly net income) variations; exemptions additional co-payments for maternity and of members of certain delivery social insurance agencies Belgium €0.99 to €14.86 co- Co-payment by patient Co-payment of €18.3 or Co-payments are paid Low-income households Very complex cost payment and 10-40% status, length of stay and €10.17 (poor) except for almost all medical pay reduced co-payments sharing mechanism.. of co-insurance type of hospital. when patient was services. The most (“preferential schemeâ€?). depending on the General: First day: referred by a GP or important example are provider and location €40.33 or € 4.94 (poor) when patient arrived in drugs, for which the Annual ceiling for (copayment set by From second day on: an ambulance; only in co-payment is individual cost-sharing income categories + €13.06 or €4.94 (poor) these 2 cases the co- differentiated for low-income maximum billing Psychiatric: €13.06 that payment is reduced to according to the households (€450, 650, applied). increases to €21.78 from €4.7 or €1.52 (poor). medical necessity; 0- 100, 1400, 1800 by net Supplementary the 6th year onwards 80% coinsurance. taxable income) private insurance permitted Denmark No No No Variable co-insurance n/a If longer than 1 rate (0-50%) applied to month on waiting list reference price of patient has the right drugs and to dental to be referred from a care. public to a private hospital free of charge. 15 The World Bank February 15, 2008 Finland €11 per doctor's - €26 in somatic No In addition to inpatient Patients up to 18 years, in OOP in public health outpatient visit, or inpatient care, care charge, special health care centers. They care will be revised alternatively €22 - €12 in psychiatric payment category may be charged in upwards by MOH in annually. inpatient care patients may have to pay: hospitals for a maximum 2008. The last time In addition to those - €12 for day and - €68 per Radiography of 7 treatment days this was done in mentioned above €15 night care therapy (€26/€12) or for 2002, when every duty payment on -long term care - €673 per Other ambulatory hospital care payment was weekends between based on patient's examinations and (€22 per visit). increased 6 % on 8:00-20:00 o'clock income treatment average. may be charged. - €68 per conservative Health care charge No charges for nurse inpatient treatment, plus ceiling adjusts the co- visits, preventive care days for payments in public health care or ancillary examinations and care. The annual ceiling services such as treatment procedures is €590 and after this laboratory visits. - Consultations €68 payment limit the In hospitals: €22 per - €236 per anesthetic outpatient services ambulatory visit. procedure become free of charge Also, there are home and the short-term nursing copayments, hospital inpatient care their magnitude depends charge drops from €26 to on the income of the €12 per day. family France10 Flat user fee of 1 € is 16€ per day, The same rules apply in Poorest people benefit due in addition to the reimbursable by ambulatory care and in from free supplementary co payment rate, in supplementary private hospitals, extra insurance. Exempt from the limit of 4€ a day insurance (except tariff for emergency care: cost-sharing are patients and 50€ a year; it is for psychiatric care: 22.60€, which is submitted suffering from a long and not reimbursable by 12€). to co payment rates as well costly disease (in a list of supplementary Special technical as the basic fee. For 30), pregnant women insurances care: 18 € per emergency visits in public over 5 months, victims of Special technical service hospitals, there is no user work accident or care: 18 € per fee. professional disease. service. 10 The French system is much more complex than described in this table. 16 The World Bank February 15, 2008 Germany €10 for the first €10per day No Higher cost sharing for n/a physician visit per prescription drugs and quarter. Second, third dental care. per quarter is free. Greece HOSPITALS No €3 or nothing (For most For pharmaceuticals or For people without Hospitalization per 1.) €3 in outpatient hospitals there is no some lab tests not income, who are under diem fees need to be clinics charge, but some providers covered by insurance, Welfare Benefits revisited in general in 2.) €75 for charge the same as when OPP payments may be Scheme, the state covers Greece, due to lack of consultant, €60 for attending an outpatient required. all costs. valid costing systems. senior registrar, €45 clinic) for registrar in evening (private hospital clinic of patient’s choice) HEALTH CENTRES No OPP payment Ireland Free for hospital €66 up to a €66 if not referred by GP Individual/family must All people with full outpatient visit maximum of €660 pay first €90 per month eligibility11 receive in any consecutive for prescribed drugs. public health services 12 months This is statutory limit entirely free of charge. 11 Entitlement to health services in Ireland is primarily based on residency and means. Any person who is accepted by the Health Service Executive (HSE) as being ordinarily resident in Ireland is entitled to either full eligibility (Category 1, i.e. medical card holders) or limited eligibility (Category 2) for health services. Persons in Category 1 are medical card holders and they are entitled to a full range of services including general practitioner services, prescribed drugs and medicines, all in-patient public hospital services in public wards including consultants services, all out-patient public hospital services including consultants services, dental, ophthalmic and aural services and appliances and a maternity and infant care service. Determination of eligibility for medical cards is the responsibility of the Health Service Executive. Persons not entitled to a medical card, but with an income below a certain threshold (50% above the medical card income guidelines) may be entitled to a GP visit card. A GP visit card entitles the holder to free GP services. For those who do not qualify for a medical card, a number of schemes exist which provide assistance towards the cost of medication. Under the Drug Payment Scheme a person and his/her dependants do not have to pay more than €90 in any calendar month for approved prescribed drugs, medicines and appliances. Persons in Category 2 (non medical card holders) are entitled, subject to certain charges, to all in-patient public hospital services in public wards including consultant services and to out-patient public hospital services including consultant services. The current public hospital statutory in-patient charge is €66 per night, up to a maximum of €660 in any twelve consecutive months. There is no charge for outpatient services, other than in respect of attendance at accident and emergency departments which is subject to a charge of €66 where the patient does not have a referral note from his/her doctor. 17 The World Bank February 15, 2008 reviewed annually. Italy No fee in primary No No payment in case of real Pharmaceutical cost Patients exempted are: care. need or emergency, sharing and co-payments who suffers from specific A specialist visit otherwise patients have to in laboratory tests. or chronic diseases; up to fee12 is €20,66 pay €25. 6 years old; the elderly over 65 years old with low income (€36,000); retired, unemployed, low income groups. Luxembourg Co-insurance of €12,60 per hospital €3,41 = 10% of the price 10% co-insurance is for No exemptions exist Fees must be applied 10%=€3.18. day (medical care is generalists and specialist, by the doctors. There For home visits by reimbursed by the 20% for dental care is no distinction doctor 20% =€6.46. funds at 100%) Co-payment for drugs between public and private sector. Netherlands Deductible policy13: Care provided by general There was a major For all services in basic insurance package, except for general practice services and obstetrical practitioners, maternity reform14 of the services, patients have to pay deductibles of €150 per year (introduced in January 2008). care and children (< 18 insurance system in Patients can chose a higher deductible (maximum of €500 per adult) in order to pay a lower years) are exempt from 2006 premium for their healthcare insurance. paying deductibles. Chronic patients receive €47 to compensate for the higher OPP paid by them (in comparison with non-chronic patients). 12 Regional Laws can introduce less expensive charges and fees for specialist visit and treatments. Specialist visits and treatments are generally prescribed by the GP, who acts as a gatekeeper to specialist services. The GP can prescribe on a prescription form until 8 specialist treatments in the same specialist branch. In this case the payment is limited to €36,15. 13 This cost sharing mechanism has just been implemented. Brief descriptions of the new systems can be found in Maarse & Bartholomée (2007) and Groenewegen & de Jong (2007). 14 After a major insurance reform in 2006, the Netherlands has now a uniform system of health insurance. The current insurance system is based on public regulation in the Health Insurance Law, enacted by private organizations (insurance carriers). Insurance carriers are obliged to offer the basic package of coverage (which is the same as the former public insurance) and to accept everybody without risk selection or premium differentiation. People are free to take out additional insurance. Additional insurance is private and insurance carriers define the conditions in terms of acceptance, coverage, premiums and cost- 18 The World Bank February 15, 2008 Norway Primary care: No No Co-payments for drugs Children under 12 years NOK 120 (€15) are exempt. Annual cost- Primary care sharing ceiling is NOK (evening): NOK 220 1600 (€27,5) Special ambulatory care: NOK 280 (€35) Portugal Central Hospitals - €5,10 per Central Hospitals - €9,20 Co-payments for drugs Almost 50% of €4,40 hospitalization day District Hospitals - €8,20 population, including District Hospitals - in the first 10 days Primary care - €3,60 pregnant, children until €2,90 €10,20 for 12, above 65, people Primary care - €2,15 ambulatory surgery without salaries above €300 a month, some diseases, etc Spain No No No Co-payments for drugs Pensioners (40%) and special medical material and equipments (e.g., orthoprothesis, variable co-payment) Sweden SEK 100-150 SEK 40 (€4) SEK 150-300 Cervix cytology n/a (€11 – 16) (€16-32) SEK 120-150 Primary care (€13 – 16) SEK 120-400 Mammography (€13 – 43) SEK80-200 Hospital care (€9 – 21) Switzerland Deductible policy: Children and adolescents A proportion of treatment costs is paid by the policy-holder. This proportion consists of: up to 18 years of age do • a standard deductible of CHF 300 (€187)per year not pay a standard • a retention fee of 10 percent of the remaining invoiced amount up to a maximum of CHF 700 deductible; (€436) per year (CHF 350 for children and adolescents). sharing. Services covered include dental care for adults, physiotherapy and alternative and complementary therapies. See: Maarse & Bartholomée (2007) and Groenewegen & de Jong (2007). 19 The World Bank February 15, 2008 Mammography - no The standard direct contribution to costs is therefore a maximum of CHF 1,000 (€623) per year deductible is payable for for adults and CHF 350 (€218) for children and adolescents15. mammography carried out for the early detection of breast cancer as part of a cantonal program United No No No Patient co-payment for • under 16 Kingdom drugs16: flat rate of £6.65 • between 16-18 and in (per item in England, full time education Scotland and Northern • Students in full-time Ireland) £3.00 (per item education in Wales) • 60 or over 80% co-payment for • pregnant, or mother of a dental care up to a baby in the last 12 ceiling of £354.00. £4.76 months entitled to for a dental check-up. medical exemptions • a war or Ministry of Defense pensioner entitled to prepayment certificate • People entitled to be included in the NHS Low Income Scheme17 . Approximately 85% of prescriptions are exempt from the charge Source: Szabó, Tünde 2008 15 Example: The cost of the treatment the patient receives in the course of a year (doctor, hospital, medication etc.) is CHF 2,000 in total. He pays a deductible of CHF 300 and 10 percent of the remaining amount, i.e. a total of CHF 470 (CHF 300 + 10% of CHF 1,700 = CHF 470). His health insurance covers the remaining CHF 1,530. 16 People who frequently need prescriptions may apply for 4-month (£31.90) or annual (£87.60) prescription certificates. 17 Persons receiving any of the following: a) income support, b) family credit d) Disability Working Allowance (DWA), e) job seeker’s allowance (income based) 20 The World Bank February 15, 2008 References Allin, S., Davaki, K., Mossialos, E., Paying for ‘free’ health care: the conundrum of informal payments in post-communist Europe. Global Corruption Report 2006: Special focus on corruption and health. Transparency International, London Pluto Press pp. 62–71. Atella, Vincenzo, Peracchi, Franco, Depalo, Domenico and Rossetti, Claudio, "Drug Compliance, Co-Payment and Health Outcomes: Evidence from a Panel of Italian Patients" (October 2005). CEIS Working Paper No. 76. SSRN: http://ssrn.com/abstract=871395 Barber, S., Bonnet, F., Bekedam, H. Formalizing under-the-table payments to control out-of- pocket hospital expenditures in Cambodia. Health Policy and Planning, 19(4): 199–208, Oxford University Press, 2004 Buchmueller Thomas and Agnes Couffinhal: (2004) Private Health Insurance in France. OECD health working paper No. 12 Cases, Chantal. "Sanitary shield". Health Policy Monitor, October 2007. Available at http://www.hpm.org/survey/fr/a10/3 Chernew, M. Optimal Health Insurance: A Benefit Based Co-Pay. Available at: www.ruf.rice.edu/~econ/conference/papers/chernew.doc, 2006 Couffinhal, A. and Paris, V. “Utilization fees imposed to Public Health Care System Users in Franceâ€? CREDES, 2001. De Graeve D, Lecluyse A, Schokkaert E, Van Ourti T, Van de Voorde C. Eigen betalingen in de Belgische gezondheidszorg. De impact van supplementen. Equity and Patient Behaviour (EPB). Brussel: Federaal Kenniscentrum voor de gezondheidszorg (KCE); 2006. The main text of the report can be downloaded from the KCE-site in English: http://www.kce.fgov.be De Jong, J. D., Verheij, R. A., and Groenewegen, P. P.. No-claimregeling werkt niet; vooral anderen zouden onnodige zorg gebruiken. Medisch Contact 61:1204-1205, 2006. Delnoij, D. M. J., Groenewegen, P. P., Ros, C. C. and Friele, R. D. Die Zuzahlungsregelung im Niederländischen Krankenkassengesetz: eine Evaluation der Effekte. Das Gesundheitswesen 62:39-44, 2000. Gaál, P. and McKee, M. ‘Informal Payment for Health Care and the Theory of “INXITâ€?’, International Journal of Health Planning and Management 19, 2004. Gaál, P. Hála-pénz? Hála-kényszer. Egészségügyi Gazdasági Szemle, 42. évfolyam különszám, 2004 Gaál, P., McKee, M., Evetovits, T. Informal payment for health care: Evidence from Hungary, Health Policy 77 pg. 86–102, 2006 Gaál P. Informal payments for health care in Hungary. London School of Hygiene and Tropical Medicine, Department of Public Health and Policy, Health Services Research Unit, University of London, 2004 Hopkins, CE, MI Roemer, DM Procter, Gartside, F Lubitz, J Gardner, GA Moser, M Moser, "Cost-Sharing and Prior Authorization Effects on Medicaid Services in California: Part I: The Beneficiaries’ Reactions," Medical Care 13, no. 7 (July, 1975). Interpharma, Jan. 2007; IMS HEALTH GMBH: Pharmamarkt Switzerland 2006, 22. January 2007. 21 The World Bank February 15, 2008 Joób, S., Idén elÅ‘ször nyereséges az Egészségbiztosítási Alap, de nagy ára volt. Index, 14 august 2007. Available at: http://index.hu/politika/belfold/oep4113/ Kutzin, J. “The Appropriate Role for Patient Cost-Sharing.â€? In Saltman, Richard B., Josep Figueras, et al., Eds. Critical Challenges for Health Care Reform in Europe. State of Health. Buckingham: Open University Press, 1998. Kutzin, J. A descriptive framework for country-level analysis of health care financing arrangements. Health Policy 56: 171-204, 2001 Kutzin, J. Note on patient awareness of copayment levels in Kyrgyzstan. WHO, Manas Health Policy Analysis Project, Policy research paper no, 17, 2002 Leu R, Schellhorn M. (2004): The evolution of income-related inequalities in health care utilization in Switzerland over time. Diskussionsschriften. Universitaet Bern. www.vwi.unibe.ch Groenewegen, P. P. and De Jong, J. D.. Dutch health insurance reform: the new role of collectives. Eurohealth 13 (no.2):10-13, 2007. Maarse, H. and Bartholomée, Y. A public-private analysis of the new Dutch health insurance system. European Journal of Health Economics 8:77-82, 2007. Newhouse, J.P. and the Insurance Experiment Group, Free for All? Lessons from the Health Insurance Experiment, Harvard University Press, Cambridge, 1993. OECD, SHA-Based Health Accounts In Thirteen OECD Countries Country Studies: Hungary National Health Accounts, 2001 Pavlova, M., Groot, W. and Van Merode, G. Public attitudes towards patient payments in Bulgarian public health care sector: results of a household survey. Health Policy, vol. 59, p. 1-24, 2002 Pazitný, P., Zajac, R., Health care reform in the Slovak Republic. The William Davidson Institute at the University of Michigan, 2005. Available at: http://wdi.umich.edu/files/Publications/PolicyBriefs/2005/Pazitny_Peter_09w.pdf Phelps, C.E. and J.P. Newhouse, Coinsurance, the Price of Time, and the Demand for Medical Services, Review of Economics and Statistics 56, 334-342, 1974. Phelps, C.E. Health Economics. Harper Collins Publishers Inc., 1992. Ringel, J.S., Hosek, S.D., Vollaard, B.A., Mahnovski, S. The Elasticity of Demand for Health Care: A Review of the Literature and its Application to the Military Health System. RAND Corporation, 2002 Ros, C. C., Groenewegen, P. P. and Delnoij, D. M. J.. All rights reserved or can we just copy? Cost sharing arrangments and characteristics of health care systems. Health Policy 52:1-13, 2000. Rosett, R.N. and L.F. Huang, 1973, The Effect of Health Insurance on the Demand for Medical Care, Journal of Political Economy 81, 281-305. Schokkaert, E., Van de Voorde, C., Health Care Reform in Belgium. Centre for Economic Studies, Leuven, 2004 Schüth, T. Perceptions of the co-payment policy among patients and health personnel: raid appraisal study in the pilot area of Chui and Issyk-Kul oblasts in Kyrgizstan, 2001 22 The World Bank February 15, 2008 Schuetz, Stefan: Case study on cost-sharing regulations in Switzerland. Background paper written for the World Bank. ECSHD. July, 2007. Szabó, Tünde: International experience in cost sharing in health care. Background paper written for the World Bank. ECSHD. January 31, 2008. Tapay, N., Private insurance, public health, OECD Observer, December 2001, http://www.oecdobserver.org/news/printpage.php/aid/566/Private_insurance,_public_health.html Van Doorslaer E, Masseria C. (2004): Income-Related Inequality in the Use of Medical Care in 21 OECD Countries. OECD Health Working Papers No. 14 Van de Voorde C, Van Doorslaer E, Schokkaert E. Effects of cost sharing on physician utilization under favourable conditions for supplier-induced demand. Health Economics, 2001; 10:457- 471. Vian, T., Gryboski, K., Sinoimeri, Z., Hall Clifford, R. Informal Payments in the Public Health Sector in Albania: A Qualitative Study, Partners for Health ReformPlus, 2004 Wagstaff A, et al. (1999): Equity in the finance of health care: some further international comparisons. Journal of Health Economics; (18):263-290 Wedig, G.J., Health Status and the Demand for Health – Results on Price Elasticities, Journal of Health Economics 7, 151-163, 1988. Westerhout, E. and Folmer, K. The Co-payment systems in health care: between moral hazard and risk reduction. CPB Discussion Paper, 2007 Winkelmann, R., Co-payments for prescription drugs and the demand for doctor visits: evidence from a natural experiment. Health Economics 13: 1081–1089, 2004 World Bank: Slovakia Public Expenditure Review. 2008. WHO. Approaching health financing policy in the WHO European Region, Regional Committee for Europe, 56 session, Copenhagen, 11-14 September 2006 Xu, K., Evans D. B, Kawabata, K., Zeramdini, R., Klavus, J., Murray, C. J. L., Household catastrophic health expenditure: a multicountry analysis. The Lancet, Vol 362 ,July 12, 2003 Zajac, R., Pažitný, P. & MarcinÄ?in, A. Slovak Reform of Health Care: From Fees to Systemic Changes. Finance a úvÄ›r - Czech Journal of Economics and Finance, 54, no.9-10, pp. 405- 419 (2004). 23