The 131st Annual Meeting (November 15-19, 2003) of APHA

Abstract #69376

Janet M. Coffman, MPP¹, Timothy T. Brown, PhD¹, Brian C. Quinn, BA¹, Douglas D. Schwalm, PhD², and Richard Scheffler, PhD¹. (1) Petris Center - School of Public Health, University of California, Berkeley, 2150 Shattuck Avenue, Suite 525, Berkeley, CA 94720-7380, 510-643-4100, janetmc@uclink.berkeley.edu, (2) Economics, Louisiana State University, 2111 CEBA, Baton Rouge, LA 70803

Physician stock is an important dimension of access to medical care, which, in turn, can have significant impact on health and well-being. Concerns have been raised regarding the impact of health maintenance organizations (HMOs) on the relative stock of physicians in various geographical markets. In markets with high rates of HMO penetration, HMOs have reduced physician compensation and restricted physician autonomy. As a consequence, physicians in these markets are increasingly dissatisfied with the practice of medicine. This dissatisfaction has prompted concern that many physicians in markets with high HMO penetration will leave the medical profession or relocate to markets with lower HMO penetration. Previous studies suggest that the stock of specialists has grown less rapidly in metropolitan areas with high rates of HMO penetration. This trend appears to be due largely to the choices of young physicians, who have become less likely to locate in areas with high HMO penetration. The current study analyzes a unique set of panel data on California physicians extracted from the American Medical Association’s Masterfile. This file contains information regarding the practice location of all physicians who practiced in California from 1988 and 1998. These data are aggregated to assess whether changes in HMO penetration are associated with changes in the number of active, patient care physicians (excluding medical residents) at the county level. We estimate a reduced form model using a linear dynamic panel estimator. This specification includes county level data on HMO penetration, hospital market structure, and socio-economic and demographic characteristics. This model represents a refinement over previous studies, which have measured changes in physician supply over a fixed, multi-year interval. Endogenous variables, including HMO penetration, are instrumented using lagged values. We hypothesize that growth in HMO penetration will be associated with a decrease in the number and percentage of physicians who are specialists, because reductions in income have been more pronounced among specialists than generalists. We further hypothesize that the response to HMO penetration will be most pronounced among young physicians since relocation costs are likely to be lower for young physicians. Finally, we hypothesize that the relationship between HMO penetration and physician stock will be moderated by the characteristics of the counties in which they practice. This project will help policymakers assess how physician stock adjusts to HMO penetration and determine whether policy interventions are necessary to maintain adequate physician stock in areas with high HMO penetration.

Learning Objectives:

At the conclusion of the session, participants will be able to:
• discuss methods for estimating the relationship between changes in physician stock and changes in HMO penetration,
• describe the impact of changes in HMO penetration on physician stock in California counties,
• identify other covariates associated with changes in physician stock,
• assess whether the impact of HMO penetration varies by physician age or specialty, and
• discuss the study's implications for policy.

Keywords: Physicians, Managed Care

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: not applicable
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.