Medicaid HMO penetration and its mix: did increased penetration affect physician participation in urban markets?

To use changes in Medicaid health maintenance organization (HMO) penetration across markets over time to test for effects on the extent of Medicaid participation among physicians and to test for differences in the effects of increased use of commercial versus Medicaid-dominant plans within the market.

The nationally representative Community Tracking Study's Physician Survey for three periods (1996-1997, 1998-1999, and 2000-2001) on 29,866 physicians combined with Centers for Medicare and Medicaid Services (CMS) and InterStudy data.

Market-level estimates of Medicaid HMO penetration are used to test for (1) any participation in Medicaid and (2) the degree to which physicians have an "open" (i.e., nonlimited) practice accepting new Medicaid patients. Models account for physician, firm, and local characteristics, Medicaid relative payment levels adjusted for geographic variation in practice costs, and market-level fixed effects.

There is a positive effect of increases in commercial Medicaid HMO penetration on the odds of accepting new Medicaid patients among all physicians, and in particular, among office-based physicians. In contrast, there is no effect, positive or negative, from expanding the penetration of Medicaid-dominant HMO plans within the market. Increases in cost-adjusted Medicaid fees, relative to Medicare levels, were associated with increases in the odds of participation and of physicians having an "open" Medicaid practice. Provider characteristics that consistently lower participation among all physicians include being older, board certified, a U.S. graduate and a solo practitioner.

The effects of Medicaid HMO penetration on physician participation vary by the type of plan. If states are able to attract and retain commercial plans, participation by office-based physicians is likely to increase in a way that opens existing practices to more new Medicaid patients. Other policy variables that affect participation include the presence of a federally qualified health center (FQHC) in the county and cost-adjusted Medicaid fees relative to Medicare.