Improving Prevention Compliance Through Appropriate Incentives:Theoretical Modelling and Empirical Evidence

Abstract

Rationale: Primary prevention concerns the avoidance of undesirable outcomes. Each prevention activity shed the light on the relationship between costs of an investment in health in the present opposed to gains in health in the future. Most theoretical work deals with secondary prevention concentrating on physician’s effort and the influence of reimbursement schemes. In contrast to this, we take a closer look at the patient’s primary preventive efforts and the corresponding monetary incentives. Connected to this problem are questions of cost-sharing rules to enhance patient’s responsibility for his own health status and improvements of the insurance system to centre more on preventive effort than just relying on therapeutic care.

Objectives: We are interested to investigate the relation between primary prevention, the effects of asymmetric information and uncertainty, and the possibility for governments and insurers to encourage prevention. In our setting, we show that contract arrangements, as well as the distribution of information are important for preventive activities. Furthermore, we analyse the determinants of the patient’s decision for preventive efforts and its relation to health insurance conditions using panel data.

Methodology. In our setting, we make use of a principal-agent setting in which the patient is acting as an agent in deciding about his preventive effort. For reasons of simplifying the analysis the principal in our model is an HMO that offers health insurance and decides about the necessary treatment in the case of an illness. Our baseline scenario is the full-information case. That is, in the case of asymmetric information, the design of the contract parameters, the level of medical services, and additional monitoring activities influence the patient’s decision about his prevention activity. Hence, we can concentrate on the analysis of a co-payment effect, a coverage effect, and a monitoring effect. Based on these results we formulate hypotheses for the empirical investigation. Data basis is the German Socio-Economic Panel (GSOEP), a representative longitudinal study of private households in Germany. We use panel data on smoking behaviour as an indicator of primary prevention and estimate the influence of contract parameters on the probability to quit smoking using a discrete time hazard model.

Results: The theoretical model shows in the full information scenario the well-known result that the patient is charged no co-payment for use of medical services. In contrast to this, the insurer uses co-payment schemes to enhance patient’s preventive behaviour if preventive activities cannot be monitored perfectly. The estimation results suggest that the decision for changes in preventive efforts depends on age effects, education, working time and health status. With respect ot the tangible effects, e.g. the effects of contract parameters, we obsere no clear effects due to the low number of private insured individuals. Moreover, it can be reasoned that in our sample an increase in the unobservable monitoring activities can partly compensate charging higher co-payments.

Conclusions: Our research stresses the relationship between monetary incentive schemes and individual behaviour as well as the influence of additional non-transparent information schemes. In addition to our theoretical results, we find evidence that changes in health behaviour depend on education and individual health assessment, too. For health policy, the question has to be raised whether it is possible to control the preventive behaviour using monetary incentives or whether we need more information on this behaviour to cope with problems of compliance. Moreover, additional problems like time inconsistency in individual behaviour have to be investigated in further research.

Keywords: prevention, asymmetric information, prevention monitoring, panel data hazard model

JEL Classification: I12, D82, C33