Abstract
Pharmaceutical and medical device companies make direct payments to physicians as speaking fees, gifts, travel expenses, and other payments while marketing their products. Numerous studies have investigated the impact of these payments on physicians’ treatment choices, however, they are limited in their ability to draw causal inference. This study contributes to the question by providing insight into what drives physicians to accept payments. I employ a “movers” panel design to estimate the relative contribution of individual effects versus environmental effects on the amount of payments received. I examine the impact of environmental factors on the annual amount of small payments received by US physicians before and after a cross-state move. The key finding is that geographical variation in these payments is driven primarily by individual effects (~80%) and only modestly by environmental effects (~20%). This finding is robust to using different subsets of the movers population. The implication is that policies focused on individual effects are more likely to be effective than those addressing environmental factors. For example, a policy applied during training in medical school (individual effect) is more likely to be effective than a policy applied to workplace norms (environmental effect).
