|
|
|
|
|
Y Richard Wang, MD, PhD, Public Policy, AstraZeneca Pharmaceuticals, 1800 Concord Pike, FOC 3 CE 317, Wilmington, DE 19803, 302 886 5394, y.richard.wang@astrazeneca.com and Mark V. Pauly, PhD, The Wharton School, Health Care Systems, University of Pennsylvania, 3641 Locust Walk, Philadelphia, PA 19104-6218.
Several prior studies examined the impact of health maintenance organizations (HMOs) and their management techniques on physician practice styles for non-HMO sectors or patients. In this paper, we study how a physician reacts to the presence of restrictive drug formularies and that of cash-paying patients in his practice and makes prescribing decisions for other, unaffiliated patients. A simple measure of drug formulary restrictiveness is the inverse of the total number of preferred products listed for a therapeutic class. It is difficult for a physician to keep track of all drug formularies and he may remember best those restrictive ones associated with pharmacist phone calls or prior authorization. We hypothesize that a restrictive drug formulary leads to more prescribing of its preferred product(s) for other patients, so called “spillover effects”. We also hypothesize that cost-conscious cash-paying patients generate similar spillover effects or lead to more prescribing of lower-priced drugs for other patients. The open drug formulary, uniform cost sharing, and national presence of conventional Medicaid make it an ideal sample to explore differences in physician practice and study such spillover effects. Combining two national databases of health plan drug formularies and physician prescription patterns for the proton pump inhibitor (PPI) class, we examine physician prescribing of Protonix for conventional Medicaid in Quarter 3, 2001, approximately one year after its launch in May 2000. Protonix is similar in efficacy and safety to three existing PPIs but is marketed at an approximately 20 percent lower list price. We find that both 1-PPI and 2-PPI formularies have a significantly positive effect on physician prescribing of Protonix for conventional Medicaid, with a larger effect from the most restrictive, 1-PPI formularies. In addition, we find that spillover effects of cash-paying patients are similar in scale to those of 2-PPI formularies. Each 10 percent increase of 1-PPI formularies in a physician practice leads to a 4.4 percent increase in Protonix share of his Medicaid PPI prescriptions (p<0.05), and each 10 percent increase of 2-PPI formularies or cash-paying patients in a physician practice leads to a 0.6 percent increase in Protonix share of his Medicaid PPI prescriptions (both p<0.01). Results from the other PPI products in the same period further verify the spillover effects of restrictive drug formularies.
Learning Objectives:
Keywords: Prescription Drug Use Patterns, Medicaid
Presenting author's disclosure statement:
I have a significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.
Relationship: I am an AstraZeneca Pharmaceuticals employee.