René C.J.A. van Vliet

How can we prevent cream skimming in a competitive health insurance market

Recently in many countries market oriented strategies have been implemented or proposed in order to improve efficiency in health care. In several countries (e.g. the Netherlands, Australia, Germany, Israel, Switzerland and the United States of America (USA)) an essential aspect of these strategies is that competing insurers receive a risk-adjusted premium-replacing payment per insured. In return the insurers are responsible for paying for or delivering the services as described in the benefits package. The payment per insured is dependent on the risk category to which the insured belongs and provides the insurer with an incentive for efficiency. However, if the risk groups are heterogeneous, cream skimming may arise, resulting in adverse effects to society in terms of reduced (incentives for) quality and efficiency of care.

Predictability of Individual Health Care Expenditures

It is widely believed in both academic and political circles that direct, uniform capitation payments to health insurers may induce them to contain costs. The premiumreplacing capitation payments should account for predictable variations in individual health care expenditures. This article describes a model that encompasses and rejects several previous models that have been proposed for analyzing these variations. The results indicate that, at most, 20 percent of the variance among individuals in annual, short-term health care expenditures is predictable.

Access to coverage for high-risks in a competitive individual health insurance market: via premium rate restrictions or risk-adjusted premium subsidies?

A competitive market for individual health insurance tends to risk-adjusted premiums. Premium rate restrictions are often considered a tool to increase access to coverage for high-risk individuals in such a market. However, such regulation induces selection which may have several adverse effects. As an alternative approach we consider risk-adjusted premium subsidies. Empirical results of simulated premium models and subsidy formulae are presented. It is shown that sufficiently adjusted subsidies eliminate the need for premium rate restrictions and consequently avoid their adverse effects. Therefore, the subsidy approach is the preferred strategy to increase access to coverage for high-risk individuals.

Risk sharing as a supplement to imperfect capitation: a tradeoff between selection and efficiency

This paper describes forms of risk sharing between insurers and the regulator in a competitive individual health insurance market with imperfectly risk-adjusted capitation payments. Risk sharing implies a reduction of an insurers incentives for selection as well as for efficiency. In a theoretical analysis, we show how the optimal extent of risk sharing may depend on the weights the regulator assigns to these effects. Some countries employ outlier or proportional risk sharing as a supplement to demographic capitation payments. Our empirical results strongly suggest that other forms of risk sharing yield better tradeoffs between selection and efficiency.

Costs in the last year of life in The Netherlands.

The costs of health care in the last year of life are a subject of debate and myth. Expensive interventions at the end of life often are blamed for the rapid increase in health care spending, but evidence about the existence of such exceptionally high expenditures at the end of life is rare and faulty. This investigation examines the development and composition of health care costs at the end of life for all age groups in The Netherlands. In contrast with earlier studies, this research analyzes both acute care (cure) and long-term care (care) costs. As an alternative for the frequently used concept of calendar years, we employed the concept of life years for calculating the costs at the end of life. We found that when life approaches its end, health care expenditures indeed rise sharply, especially in the last months. However, when we compared total cure costs in the last year of life to the total cure costs for the entire population, we concluded that the end-of-life share was only about 10%. Results of this study show that interventions to reduce costs in the last year of life will have only a modest impact compared to the total health care budget.

Deductibles and Health Care Expenditures: Empirical Estimates of Price Sensitivity Based on Administrative Data

Objective: To estimate the price sensitivity of demand for health care by analysing the relation between deductibles and expenditures found in an administrative database. Data: Data are from 100,048 privately insured in the Netherlands. Information is available on expenditures in 1996, demographic variables, deductibles, and both diagnoses from hospitalisations and pharmacy costs for the years 1993–l1995. The data are unique because prior pharmacy costs are good predictors of future expenditures while in the three years concerned these costs were covered by a separate national, mandatory insurance scheme. Therefore, these costs are not affected by deductibles and can be seen as excellent proxies for health status, which this study uses to correct for the existing adverse selection with respect to the choice of deductible. Methods: For the group of insured without deductibles an expenditure model is estimated which is subsequently used to estimate expected expenditures for insurance policies with various levels of deductible. From the difference between actual and expected expenditures the impact of deductibles on expenditures is estimated, and from this price sensitivity. Results: The principal finding is an estimated price elasticity of −0.14. The highest price sensitivity is found for GP care (−0.40) and physiotherapy (−0.32), and the lowest for specialist care (−0.12) and prescription drugs (−0.08); hospital care was hardly affected.AbstractObjective: To estimate the price sensitivity of demand for health care by analysing the relation between deductibles and expenditures found in an administrative database. Data: Data are from 100,048 privately insured in the Netherlands. Information is available on expenditures in 1996, demographic variables, deductibles, and both diagnoses from hospitalisations and pharmacy costs for the years 1993–l1995. The data are unique because prior pharmacy costs are good predictors of future expenditures while in the three years concerned these costs were covered by a separate national, mandatory insurance scheme. Therefore, these costs are not affected by deductibles and can be seen as excellent proxies for health status, which this study uses to correct for the existing adverse selection with respect to the choice of deductible. Methods: For the group of insured without deductibles an expenditure model is estimated which is subsequently used to estimate expected expenditures for insurance policies with various levels of deductible. From the difference between actual and expected expenditures the impact of deductibles on expenditures is estimated, and from this price sensitivity. Results: The principal finding is an estimated price elasticity of −0.14. The highest price sensitivity is found for GP care (−0.40) and physiotherapy (−0.32), and the lowest for specialist care (−0.12) and prescription drugs (−0.08); hospital care was hardly affected.

Health-based risk adjustment: Improving the pharmacy-based cost group model to reduce gaming possibilities

The pharmacy-based cost group (PCG) model uses medication prescribed to individuals in a base-year as marker for chronic conditions which are employed to adjust capitation payments to their health plans in the subsequent year. Although the PCG model enhances predictive performance, possibilities for gaming may arise as it is based on prior utilization. This study investigates several strategies to mitigate this problem. The best strategies appear to be: use a (high) number of prescribed daily doses to assign persons to PCGs, do not allow for comorbidity, and remove PCGs with low future costs. This PCG model accounts for almost twice as much variance as do demographic models. In 2002 the Dutch government implemented this model in the sickness fund sector (twothirds of the population).

Risk adjusted premium subsidies and risk sharing: key elements of the competitive sickness fund market in the Netherlands

As part of a market-oriented health care reform, in 1991 risk adjusted premium subsidies were introduced in the Dutch social health insurance sector. Currently the premium subsidies are primarily based on demographic variables. To mitigate the obvious inadequacy of these risk adjusters, the system of risk adjustment is supplemented with a system of risk sharing. This paper describes the main characteristics of the Dutch health care system and the development of risk adjustment and risk sharing in the last decade. The effects of introducing financial risk for Dutch sickness funds on risk selection and consumer mobility are analysed. The paper concludes with a description of expected future developments.

Diagnostic, Pharmacy-Based, and Self-Reported Health Measures in Risk Equalization Models

Background:Current research on the added value of self-reported health measures for risk equalization modeling does not include all types of self-reported health measures; and/or is compared with a limited set of medically diagnosed or pharmacy-based diseases; and/or is limited to specific populations of high-risk individuals. Objective:The objective of our study is to determine the predictive power of all types of self-reported health measures for prospective modeling of health care expenditures in a general population of adult Dutch sickness fund enrollees, given that pharmacy and diagnostic data from administrative records are already included in the risk equalization formula. Research Design:We used 4 models of 2002 total, inpatient and outpatient expenditures to evaluate the separate and combined predictive ability of 2 kinds of data: (1) Pharmacy-based (PCGs) and Diagnosis-based (DCGs) Cost Groups and (2) summarized self-reported health information. Model performance is measured at the total population level using R2 and mean absolute prediction error; also, by examining mean discrepancies between model-predicted and actual expenditures (ie, expected over- or undercompensation) for members of potentially “mispriced” subgroups. These subgroups are identified by self-reports from prior-year health surveys and utilization and expenditure data from 5 preceding years. Subjects:Subjects were 18,617 respondents to a health survey, held among a stratified sample of adult members of the largest Dutch sickness fund in 2002, with an overrepresentation of people in poor health. Data:The data were extracted from a claims database and a health survey. The claims-based data are the outcomes of total, inpatient, and outpatient annualized expenditures in 2002; age, gender, PCGs, DCGs in 2001; and health care expenditures and hospitalizations during the years 1997 to 2001. The SF-36, Organization for Economic Cooperation and Development items, and long-term diseases and conditions were collected by a special purpose health survey conducted in the last quarter of 2001. Results:Out-of-sample R2 equals 17.2%, 2.6%, and 32.4% for the models of total, inpatient and outpatient expenditures including PCGs, DCGs, and self-reported health measures. Self-reported health measures contribute less to predictive power than PCGs and DCGs. PCGs and DCGs also predict better than self-reported health measures for people with top 25% total expenditures or hospitalizations in each year during a 5-year period. On the other hand, self-reported health measures are better predictors than PCGs and DCGs for people without any top 25% expenditures during the 5-year period, for switchers, and for most subgroups of relatively unhealthy people defined by self-reported health measures. Among the set of self-reported health measures, the SF-36 adds most to predictive power in terms of R2, mean absolute prediction error, and for almost all studied subgroups. Conclusion:It is concluded that the self-reported health measures make an independent contribution to forecasting health care expenditures, even if the prediction model already includes diagnostic and pharmacy-based information currently used in Dutch risk equalization models.

Health-based risk adjustment: Improving the pharmacy-based cost group model by adding diagnostic cost groups

Since 1991, risk-adjusted premium subsidies have existed in the Dutch social health insurance sector, which covered about two-thirds of the population until 2006. In 2002, pharmacy-based cost groups (PCGs) were included in the demographic risk adjustment model, which improved the goodness-of-fit, as measured by the R2, to 11.5%. The models R2 reached 22.8% in 2004, when inpatient diagnostic information was added in the form of diagnostic cost groups (DCGs). PCGs and DCGs appear to be complementary in their ability to predict future costs. PCGs particularly improve the R2 for outpatient expenses, whereas DCGs improve the R2 for inpatient expenses. In 2006, this system of risk-adjusted premium subsidies was extended to cover the entire population.