Carine Van de Voorde

Consumer mobility in social health insurance markets : a five-country comparison.

During the 1990s, the social health insurance schemes of Germany, the Netherlands, Switzerland, Belgium and Israel were significantly reformed by the introduction of freedom of choice (open enrolment) of health insurer. This was introduced alongside a system of risk adjustment to compensate health insurers for enrolees with predictable high medical expenses. Despite the similarity in the health insurance reforms in these countries, we find that both the rationale behind these reforms and their impact on consumer choice vary widely.In this article we seek to explain the observed variation in switching rates by cross-country comparison of the potential determinants of health insurer choice. We conclude that differences in choice setting, and in the net benefits of switching, offer a plausible explanation for the large differences in consumer mobility.Finally, we discuss the policy implications of our cross-country comparison. We argue that the optimal switching rate crucially depends on the goals of the reforms and the quality of the risk-adjustment system. In view of this, we conclude that switching rates are currently too low in the Netherlands, and an active government policy to encourage consumer mobility seems warranted. In Germany and Switzerland, high switching rates call for an improvement of the rather poor risk-adjustment systems. Given low switching rates in Israel and Belgium, improving risk adjustment is less urgent, but still required in the long run.

Supplementary health insurance as a tool for risk-selection in mandatory basic health insurance markets

As the share of supplementary health insurance (SI) in health care finance is likely to grow, SI may become an increasingly attractive tool for risk-selection in basic health insurance (BI). In this paper, we develop a conceptual framework to assess the probability that insurers will use SI for favourable risk-selection in BI. We apply our framework to five countries in which risk-selection via SI is feasible: Belgium, Germany, Israel, the Netherlands, and Switzerland. For each country, we review the available evidence of SI being used as selection device. We find that the probability that SI is and will be used for risk-selection substantially varies across countries. Finally, we discuss several strategies for policy makers to reduce the chance that SI will be used for risk-selection in BI markets.

Belgium: risk adjustment and financial responsibility in a centralised system

Since 1995 Belgian sickness funds are partially financed through a risk adjustment system and are held partially financially responsible for the difference between their actual and their risk-adjusted expenditures. However, they did not get the necessary instruments for exerting a real influence on expenditures and the health insurance market has not been opened for new entrants. At the same time the sickness funds have powerful tools for risk selection, because they also dominate the market for supplementary health insurance. The present risk-adjustment system is based on the results of a regression analysis with aggregate data. The main proclaimed purpose of this system is to guarantee a fair treatment to all the sickness funds. Until now the danger of risk selection has not been taken seriously. Consumer mobility has remained rather low. However, since the degree of financial responsibility is programmed to increase in the near future, the potential profits from cream skimming will increase.

Supplemental health insurance and equality of access in Belgium

The effects of supplemental health insurance on health-care consumption crucially depend on specific institutional features of the health-care system. We analyse the situation in Belgium, a country with a very broad coverage in compulsory social health insurance and where supplemental insurance mainly refers to extra-billing in hospitals. Within this institutional background, we find only weak evidence of adverse selection in the coverage of supplemental health insurance. We find much stronger effects of socio-economic background. We estimate a bivariate probit model and cannot reject the assumption of exogeneity of insurance availability for the explanation of health-care use. A count model for hospital care shows that supplemental insurance has no significant effect on the number of spells, but a negative effect on the number of nights per spell. We comment on the implications of our findings for equality of access to health care in Belgium.

Direct versus indirect standardization in risk adjustment

Direct and indirect standardization procedures aim at comparing differences in health or in health care expenditures between subgroups of the population after controlling for observable morbidity differences. There is a close analogy between this problem and the issue of risk adjustment in health insurance. Traditional methods of risk adjustment are analogous to indirect standardization. They are equivalent to the so-called conditional egalitarian mechanism in social choice. In general, they do not remove incentives for risk selection, even if the effect of non-morbidity variables is correctly taken into account. A method of risk adjustment based on direct standardization does remove the incentives for risk selection, but at the cost of violating a neutrality condition, stating that insurers should receive the same premium subsidy for all members of the same risk group. Direct standardization is equivalent to the egalitarian-equivalent (or proportional) mechanism in social choice. The conflict between removing incentives for risk selection and neutrality is unavoidable if the health expenditure function is not additively separable in the morbidity and efficiency variables.

Equity in Health And Equivalent Incomes

We compare two approaches to measuring inequity in the health distribution. The first is the concentration index. The second is the calculation of the inequality in an overall measure of individual well-being, capturing both the income and health dimensions. We introduce the concept of equivalent income as a measure of well-being that respects preferences with respect to the trade-off between income and health, but is not subjectively welfarist since it does not rely on the direct measurement of happiness. Using data from a representative survey in France, we show that equivalent incomes can be measured using a contingent valuation method. We present counterfactual simulations to illustrate the different perspectives of the approaches with respect to distributive justice.

The Belgian Health System Performance Report 2012: Snapshot of results and recommendations to policy makers

Following the commitments of the Tallinn Charter, Belgium publishes the second report on the performance of its health system. A set of 74 measurable indicators is analysed, and results are interpreted following the five dimensions of the conceptual framework: accessibility, quality of care, efficiency, sustainability and equity. All domains of care are covered (preventive, curative, long-term and end-of-life care), as well as health status and health promotion. For all indicators, national/regional values are presented with their evolution over time. Benchmarking to results of other EU-15 countries is also systematic. The policy recommendations represent the most important output of the report.

Hospital supplements in Belgium: Price variation and regulation

OBJECTIVES Although there is a comprehensive public health insurance system in Belgium, out-of-pocket expenditures can be very high, mainly for inpatients. While a large part of the official price is reimbursed, patients are confronted with increased extra billing (supplements). Therefore, the government imposed various restrictions on the amount of supplements to be charged, related to the type of room and the patients insurance status. We investigate how prices are set and whether the restrictions have been effective. METHODS We use an administrative dataset of the Belgian sickness funds for the year 2003 with billing data per hospitalisation and hospital characteristics. Boxplots describe the distribution of several categories of supplements. OLS is used to explore the relationship between hospital characteristics and extra billing. RESULTS There is a large and intransparent variation in extra billing practices among different hospitals. Given the room type, supplements per day are smaller for patients qualifying for protection, confirming that the regulation is applied quite well. However, because of their longer length of stay this does not result in lower supplements per stay for these patients. CONCLUSIONS Currently the price setting behavior of providers lacks transparency. Protective regulation could be refined by taking into account the longer length of stay of vulnerable groups.

Incentives for risk selection and omitted variables in the risk adjustment formula.

Risk adjustment in health insurance raises the question of how to treat variables which influence health care expenditures but do not capture acceptable costs differences. We argue that these variables should be included in the explanatory model and neutralized afterwards for the computation of the premium subsidies. This explicit approach is better than the conventional approach in removing the incentives for cream-skimming. We illustrate the empirical relevancy of the problem with data for Belgium.

Best affine unbiased response decomposition

Given two linear regression models y1 = X1β1 + u1 and y2 = X2β2 + u2 where the response vectors y1 and y2 are unobservable but the sum y = y1 + y2 is observable, we study the problem of decomposing y into components ŷ1 and ŷ2, intended to be close to y1 and y2, respectively. We develop a theory of best affine unbiased decomposition in this setting. A necessary and sufficient condition for the existence of an affine unbiased decomposition is given. Under this condition, we establish the existence and uniqueness of the best affine unbiased decomposition and provide an expression for it.