Direct primary care: Family physician perceptions of a growing model

INTRODUCTION: Direct Primary Care (DPC) is a relatively new primary care practice model in which patients receive unlimited access to a defined set of primary care services in exchange for a monthly practice-specific membership fee. DPC is a bottom-up physician-driven approach in contrast to typical top-down insurer-centric healthcare delivery reform efforts. The degree to which physicians are aware of this practice model and whether they believe it addresses two key challenges facing primary care, access and administrative burden, are unclear. METHODS: An online survey was distributed in July 2017 to 672 members of a research marketing sample of the American Academy of Family Physicians (n=225; response rate 33%). Based on AAFP definitions, the survey consisted of both open- and close-ended questions that gauged family physicians’ awareness of DPC, as well as their perceptions about the model. RESULTS: Most respondents (85%) had heard of DPC and 8% practised in a DPC model at the time of the survey. In general, respondents reported that DPC can offer positive outcomes through lower administrative burden for physicians, improved doctor–patient relationships, and better access. Respondents also suggested DPC may result in improved patient health outcomes and lower overall healthcare spending. Respondents’ concerns included inappropriateness of the model for vulnerable populations and physician shortages. Survey responses differed depending on whether the respondent practised in a DPC model. DPC physicians had a more favorable view of the model and were focused on benefits to patients rather than benefits to physicians. CONCLUSIONS: Any changes to practice models will require a better understanding of clear definitions of practice models. Further education is required about the specific benefits to vulnerable patients and practice standards. As the perceptions of DPC vary by practice experience, DPC physicians appear to be strong potential advocates of the model.