Presented By: Department of Economics
Applied Microeconomics/IO Seminar: Quantifying sources of persistent prescription behavior: Evidence from Belgium
Pieter De Vlieger, University of Michigan
Abstract
Health care policymakers often find it challenging to change physician behavior, as it is highly persistent. Drivers of this persistence are not well understood, but physicians and patients both play a role. I study the importance of both in the prescribing behavior of primary care physicians (PCPs) in Belgium. I exploit a mandate introduced in 2006 that required PCPs to prescribe a minimum percentage of cheap or generic drugs, and analyze the change in PCP prescription habits using administrative data linking 26 million dispensed prescription drugs to 150,000 patients and 45,000 physicians. I find that PCPs exhibit a bias towards prescribing a brand name drug when an equally effective generic is available, and adjust this behavior in response to the mandate without compromising the quality of drugs they prescribe. I show that the type of patient to whom a drug is prescribed matters as well. Compared to patients prescribed medication for the first time, PCPs switch long-time users from branded to generic versions of the same drug at much lower rates, especially when these patients are older or use multiple prescription drugs. This suggests that there is a cost to switching between drugs and that it varies by patient characteristics. Using an instrumental variables framework, I estimate that switching a patient from a brand name to generic version of the same drug indeed comes at a cost, measured with decreased medication adherence. I develop a structural model of prescription behavior to quantify the relative importance of physician bias and patient considerations in PCP prescribing behavior, and find they are about equally important. Using this model, I show that the introduction of a Mandatory Generic Substitution policy may decrease overall welfare as a result of patient considerations. This suggests that policy efforts aimed at changing physician behavior should also consider potentially negative health impacts on patients.
Health care policymakers often find it challenging to change physician behavior, as it is highly persistent. Drivers of this persistence are not well understood, but physicians and patients both play a role. I study the importance of both in the prescribing behavior of primary care physicians (PCPs) in Belgium. I exploit a mandate introduced in 2006 that required PCPs to prescribe a minimum percentage of cheap or generic drugs, and analyze the change in PCP prescription habits using administrative data linking 26 million dispensed prescription drugs to 150,000 patients and 45,000 physicians. I find that PCPs exhibit a bias towards prescribing a brand name drug when an equally effective generic is available, and adjust this behavior in response to the mandate without compromising the quality of drugs they prescribe. I show that the type of patient to whom a drug is prescribed matters as well. Compared to patients prescribed medication for the first time, PCPs switch long-time users from branded to generic versions of the same drug at much lower rates, especially when these patients are older or use multiple prescription drugs. This suggests that there is a cost to switching between drugs and that it varies by patient characteristics. Using an instrumental variables framework, I estimate that switching a patient from a brand name to generic version of the same drug indeed comes at a cost, measured with decreased medication adherence. I develop a structural model of prescription behavior to quantify the relative importance of physician bias and patient considerations in PCP prescribing behavior, and find they are about equally important. Using this model, I show that the introduction of a Mandatory Generic Substitution policy may decrease overall welfare as a result of patient considerations. This suggests that policy efforts aimed at changing physician behavior should also consider potentially negative health impacts on patients.
Co-Sponsored By
Explore Similar Events
-
Loading Similar Events...