What drives out-of pocket health expenditures of private households? : empirical evidence from the Austrian household budget survey

Out-of-pocket health expenditures (OOPHE) are a substantial source of health care financing even in health care systems with an established role of prepaid financing. The empirical analysis of OOPHE is challenging, because they are fixed in an interaction with other sources of health care financing. We analyze to what extent a set of socio-economic and socio-demographic covariates of private households in influences the OOPHE-patterns in Austria. Our empirical research strategy is guided by the approach Propper (2000) used to study the demand for private health care in the NHS. We use cross-sectional information provided by the Austrian household budget survey 2009/10. We apply a Two-Part Model (Logit/OLS with log-transformed dependent variable or Logit/GLM). We present results for total OOPHE and for selected OOPHE-subcategories. Overall, we find mixed results for the different expenditure categories and for the two decision stages. Probability and level of OOPHE increase with the household size and the level of education, while household income shows mixed results on both stages. Private health insurance and OOPHE seem to be complements, at least for total OOPHE and for OOPHE for physician services, while this relationship is insignificant for pharmaceuticals. Different forms of public insurance have an effect on the total OOPHE-level, for physician services and pharmaceuticals on both stages. To some extent the participation decision is influenced in a different way compared to the intensity decision. This is especially true for age, sex, household structure and the status of retirement. It turns out, that the explanatory power of the used variables is low for OOPHE for pharmaceuticals. A splitting up of pharmaceuticals into prescription fees and direct payments gives better insights into the determinants. We conclude: It is necessary to investigate subcategories of OOPHE. It also turns out, that systematic covariates explain only a very small part of the variation in the OOPHE-patterns. Finally, we also conclude that information on OOPHE from general household budget surveys are of limited value when studying the determinants.