Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

In the Netherlands certain groups in the population are offered preventive health care free of charge, including flu shots (65+ population, every year), cervical cancer tests (women 30-60, every five years), mammograms (women 50+, every two years), and kidney checks. We analyze individuals' decisions to decline or accept these interventions, using a survey that includes information on actual participation, risk factors, and socioeconomic background variables. In particular, we study individuals' perceptions of the effectiveness of the interventions, by eliciting subjective probabilities of sickness and survival, both with and without the interventions. We also analyze the relationship between risk factors and perceived probabilities, and compare the subjective probabilities with epidemiological estimates based on individuals' risk factors. A majority of respondents answer the probability questions in an internally consistent way and appear to be aware of some of the qualitative relationships between risk factors and probabilities. However, we find that individuals have very poor perceptions of the absolute levels of the probabilities of sickness and survival.

The exogenous changes in participation costs at the time of reaching the entitlement age and the multiplicity of decisions by the same individuals (both across preventive interventions and over time) allow for the estimation of the willingness to pay for risk reductions (and the implied value of a statistical life) without the endogeneity problems that characterize much of the earlier literature. The key differences between this work and previous VSL literature are that we are able to calculate multiple upper and lower bounds for the VSL for each individual and we estimate individual's willingness to pay to reduce their own risk of death. Our initial estimates suggest that a wide range of values of a statistical life; lower bounds (for people who choose risk reducing preventive care) range from 0 Euro to 93,000 Euro with a median of 53 Euro and upper bounds (for people who do not choose risk reducing preventive care) range from 0 Euro to 8.6 million Euro with a median of 2,990 Euro. Extremely low values are driven by the fact that preventive care is free to many people. The wide variation is driven primarily by the variation in subjective probabilities. In addition, we compare VSL calculated based on subjective assessments of risk and based on epidemiological predictions of risk. Using epidemiological probabilities gives lower bounds (for people who choose risk reducing preventive care) range from 0 Euro to 4.4 million Euro with a median of 3,336 Euro and upper bounds (for people who do not choose risk reducing preventive care) range from 0 Euro to 36 million Euro with a median of 1.1 million Euro. The differences between these estimates are primarily driven by the large overestimation of subjective risk relative to the underlying epidemiological risks.