Does physicians' compensation affect the probability of their vetoing generic substitution?

Physicians' decisions whether or not to veto generic substitution were analyzed using a sample of 350,000 pharmaceutical prescriptions. Point estimates show that - compared to county-employed physicians on salary - physicians working at private practices were 50-80% more likely to veto substitution. The results indicate that this difference is explained by the difference in direct cost associated with substitution, rather than by private physicians' possibly stronger incentives to please their patients. Also, the probability of a veto was found to increase as patients' copayments decreased. This might indicate moral hazard in insurance, though other explanations are plausible.