Abstract (FR)

BACKGROUND: In France, only 75% of health expenditures are covered by the French public health insurance. Health expenditures which are not publicly covered can be covered by complementary health insurance contract. The weighty out-of-pocket resulting from the lack of CHI and the poor quality of some contracts are, therefore, a public health concern since this may hinder the access to health care, especially for poor households. This concerns can be illustrated by the implementation in 2000 of a free complementary health insurance for the 7% of the poorest French population, called the “Couverture maladie universelle complémentaire” (CMUC), and the implementation of a voucher, called the “Chèque Santé”, in 2005 in order to reduce financial barriers to access to health insurance for the households whose resources are just above the CMUC eligibility threshold. However, almost 8% of the population remains not covered by any complementary health insurance. Therefore, more researches are needed to understand the determinants of the non demand of complementary health insurance in France. OBJECTIVE: This study aims to explore the determinants of the demand for Complementary Health Insurance (CHI) in France, through a two-stage decision process: (1) to opt for take-up an individual contract of CHI, and (2) to choose the level of coverage. STUDY SETTING: The study is based on the French National Expenditure Survey, carried out by INSEE in 2006. This survey provides for a representative sample of 10240 households accurate information on household resources and on all household expenditures, including complementary health insurance expenditures. The study sample excludes individuals who are covered “freely” through the CMUC or who are covered through a contract provided by their employer. METHODS: We develop a two stages Heckman Sample Selection model in order to analyse the determinants of the probability of being insured and the amount of health insurance expenditures. CHI expenditures are interpreted as commensurate with the level of coverage offered by the CHI; and represent therefore a choice of quality with fixed risk. The model takes the correlation between the decision to take-up a CHI and CHI expenditures engaged into account. Independent variables include age, gender, educational level, employment status, income level (specified as piecewise linear function), composition of household, and location of residence. RESULTS: The finding indicate that the level of income has a strong impact on the decision of being insured, especially for poor households, but it is not a main determinant of the level of coverage chosen. CONCLUSION: Income is the main determinant of the decision to take-up a CHI. These finding raise a concern for the access of quality health services, especially if the French system evolves to rely more on private insurance.