Determinants of Generic vs. Brand Drug Choice: Evidence from Population-wide Danish Data

When prescription medications go off patent, vastly cheaper generic drugs usually enters the market. However, the original brand medication often maintains non-negligible market shares. This paper investigates whether demand for branded medications in post-patent markets is patient- or doctor driven. We use population-wide Danish register data including all prescriptions for seven blockbuster drugs from 1998-2008. At the outset, descriptive statistics suggest large variation in drug choice over doctors. Nonetheless, using a two-way fixed effects model we find that the primary determinants of brand drug use are unobserved patient characteristics and price effects, while observed and unobserved doctor characteristics in general explain only 0.7 % of the variation in drug choice. This is suggestive evidence that the doctors in the Danish setting with no incentives to push expensive brand drugs do indeed not do so. Our results also suggest that one should be careful when applying fixed effects in small samples.