Associations Between the Patient Protection and Affordable Care Act Medicaid Primary Care Payment Increase and Physician Participation in Medicaid

Importance: The Patient Protection and Affordable Care Act (ACA) increased 2013 to 2014 Medicaid payment rates for qualifying primary care physicians (PCPs) and services to higher Medicare payment levels, with the goal of improving primary care access for Medicaid enrollees. Objectives: To evaluate the payment increase policy and to assess whether it was associated with changes in Medicaid participation rates or Medicaid service volume among PCPs. Design, Setting, and Participants: This study used 2012 to 2015 IMS Health aggregated medical claims and encounter data from PCPs eligible for the payment increase practicing in all states except Alaska and Hawaii and included 20 723 PCPs with observations in each month from January 1, 2012, to December 31, 2015. Data are for professional services performed in ambulatory settings, including office, hospital outpatient department, and emergency department. Regression models were used to test whether outcomes differed in months subject to higher payment rates relative to months before the increase and after the expiration of the increase in some states. The models controlled for time-invariant physician characteristics and time-varying characteristics, such as Medicaid enrollment. Interaction terms were included to estimate differential associations in subgroups of states (eg, by Medicaid managed care penetration) and physicians (eg, by specialty). Main Outcomes and Measures: Physician-month records subject to higher Medicaid payment rates were flagged using state-specific implementation and end dates for the payment increase. Five outcomes were measured for each physician-month observation, including (1) an indicator for seeing any patients enrolled in Medicaid, (2) an indicator for seeing more than 5 patients enrolled in Medicaid, (3) the Medicaid share of total patients, (4) a count of new patient evaluation and management visits furnished to patients enrolled in Medicaid, and (5) a count of existing patient evaluation and management visits furnished to patients enrolled in Medicaid. Results: Among 20 723 PCPs, the payment increase had no association with PCP participation in Medicaid or Medicaid service volume. The estimated average marginal effects for all 5 outcomes were not statistically distinguishable from 0. This null result was robust to sensitivity analyses, including different time trend specifications and analyses focusing on the payment increase implementation and expiration time frames. Descriptively, the Medicaid share of patients increased by about 25% from 2012 to 2015, although the share did not increase differentially in states and months subject to higher payment rates. Conclusions and Relevance: The limited duration and design of the payment increase may have dampened its effectiveness. Future efforts to improve access through payment changes or other means can benefit from better understanding of the outcomes of this policy.