Working Paper Series, University of Zurich, Department of Economics 204
From 2004 to 2012, the German social health insurance levied a co-payment for the first doctor visit in a calendar quarter. We develop a new model for estimating the effect of such a co-payment on the individual number of visits per quarter. The model allows for a one time increase in the otherwise constant hazard rate determining the timing of doctor visits, and uses a difference-in-differences strategy to identify the reform effect. The model can be adapted to a situation where the reporting period and the calendar quarter differ. Using data from the German Socio-Economic Panel, we do not find an effect of the co-payment on demand for doctor visits.
count data Poisson process co-payment hurdle model