The differential effect of compensation structures on the likelihood that firms accept new patients by insurance type.
Adequate access to primary care is not universally achieved in many countries, including the United States, particularly for vulnerable populations. In this paper we use multiple years of the U.S.-based Community Tracking Survey to examine whether a variety of physician compensation structures chosen by practices influence the likelihood that the practice takes new patients from a variety of different types of insurance. Specifically, we examine the roles of customer satisfaction and quality measures on the one hand, and individual physician productivity measures on the other hand, in determining whether or not firms are more likely to accept patients who have private insurance, Medicare, or Medicaid. In the United States these different types of insurance mechanisms cover populations with different levels of vulnerability. Medicare (elderly and disabled individuals) and Medicaid (low income households) enrollees commonly have lower ability to pay any cost sharing associated with care, are more likely to have multiple comorbidities (and so be more costly to treat), and may be more sensitive to poor access. Further, these two insurers also generally reimburse less generously than private payors. Thus, if lower reimbursements interact with compensation mechanisms to discourage physician practices from accepting new patients, highly vulnerable populations may be at even greater risk than generally appreciated. We control for the potential endogeneity of incentive choice using a multi-level propensity score method. We find that the compensation incentives chosen by practices are statistically and economically significant predictors for the types of new patients that practices accept. These findings have important implications for both policy makers and private health care systems.
Bullock, JB; Bradford, WD
Volume / Issue
Start / End Page
Pubmed Central ID
Electronic International Standard Serial Number (EISSN)
Digital Object Identifier (DOI)