Primary Care Physician Supply, Access to Care, and Market Adjustments to Physician Shortages

Abstract

Research ObjectiveTo alleviate the problem of primary care physician (PCP) shortage, policy makers have established new medical schools and expanded medical school class sizes over the last 15 years. However, few studies have examined whether the increased PCP supply improves access to care. We aimed to fill the gap by examining the effect of PCP supply on adults’ access to care and assessing the extent to which market adjustments mitigate PCP shortages.Study DesignWe used primary care service areas (PCSAs) as market areas and merged PCSA‐level PCP supply (PCPs per 1000 population) with individual‐level data from the 2004‐2013 Medical Expenditures Panel Survey (MEPS). The data on physician supply and number of patients that PCPs saw per day came from SK&A while all other measures were from the MEPS.We modeled the probability that adults had a nonemergency room usual source of care (USC) as a function of PCP supply and individual sociodemographic characteristics, health status, and beliefs and attitudes regarding health and health care. We estimated analogous models to examine how PCP supply affects travel time to the USC and the ability to get needed care right away and make appointments as soon as needed. To examine market adjustments, we assessed the relationship between PCP supply and whether PCPs offered night and weekend hours, how many patients PCPs saw per day, and how frequently individual adults had office visits. As sensitivity analyses, we tested alternative PCP definitions (eg, including vs. excluding obstetricians/gynecologists) and estimated the models separately for adults with and without health insurance. All the models controlled for nurse practitioners and physician assistants.Population Studied378 400 adults included in the 2004‐2013 MEPS.Principal FindingsPCP supply had no effect on the probability of having a USC or the ability to get needed care right away or make timely appointments by adults. By contrast, higher PCP supply reduced adults’ travel time to the USC. We also found evidence of market adjustments that enhanced the physicians’ capacity to see different patients in areas with low PCP supply. Specifically, PCPs in these areas were more likely to offer night and weekend hours and saw more patients per day (elasticity at mean = −0.19), while adults in these areas had fewer office visits (elasticity at mean = 0.07). Taken together, the latter two adjustments compensate for more than one‐fourth of the lost capacity to see different patients where there are fewer PCPs. The sensitivity analyses confirmed robustness of the results.ConclusionsSince both having a USC and the ability to get timely care when needed and make timely appointments are likely to be more important indicators of access than travel time to the USC, our findings suggest that more PCPs do not improve the most important access indicators for most U.S. adults. Market adjustments in areas with fewer PCPs are at least partly responsible for the lack of effects of PCP supply on having a USC.Implications for Policy or PracticeProducing more physicians may be a blunt instrument for improving access to care. Targeted policies are likely to be more efficient in improving access for adults who currently lack it.Primary Funding SourceAgency for Healthcare Research and Quality.