A comparison of the determinants of primary care physician participation in Medicare and Medicaid in five states

The viability of both Medicare, providing medical services to the elderly and disabled, and Medicaid, providing medical services to the poor, are both critically dependent on the willingness of physicians, in particular primary care physicians, to accept Medicare and Medicaid patients.

In recent years, there has been an annual ritual in which the Centers for Medicare & Medicaid Services announces a fee cut for physicians under the sustainable growth rate formula; physicians respond that they will be forced to no longer accept Medicare patients or in the extreme that they will cease seeing patients in their entirety; and Congress adopts a short term fix. While the anecdotal stories of physicians restricting or leaving practice abound, the facts with respect to Medicare are somewhat mixed. Reported access issues are even greater among Medicaid beneficiaries; the Center for Studying Health System Change (HSC) found that more physicians are reporting receiving no Medicaid revenue and more are reporting that they accept no new Medicaid patients. HSC also found that Medicaid services are "increasingly concentrated among physicians who practice in large groups, hospitals, academic medical centers and community health centers" and that in 2004-05 "35.3 percent of physicians in solo and two-physician practices were not accepting new Medicaid patients." The issues with respect to Medicaid are potentially more complex since Medicaid is far more varied across the States than is Medicare.

Our dataset is the Physician Perspectives on Managed Care Survey, which was fielded in 2007 and funded by the Agency for Healthcare Research and Quality, the California Endowment and the Commonwealth Fund. The sample frame was primary care physicians (general family practice, general practice, general internal medicine, and general pediatrics) in five states – California, Georgia, Illinois, Pennsylvania, and Texas. The survey covered, among other things, physician characteristics and demographics, practice characteristics, use of health information technology, workload, ownership, practice revenue and physician earnings. Our dataset includes 1,967 completed surveys with a completion rate of just under 70%.

Roughly 57% of our sample of physicians accepts both new Medicare and Medicaid patients while 21% are not accepting either. Five percent accept new Medicaid patients only while 17% are just accepting Medicare patients. Our study, using logistic regression, examines which physician characteristics, practice characteristics (including extensive information on the financial aspects of the practice), and other geographic factors, including patient mix, determine whether a physician accepts Medicare or Medicaid and how these compare schemes.