Primary Care Clinicians Available for New Patient Visits

Introduction

Amidst widespread concern that finding a new primary care physician (PCP) in the U.S. is increasingly difficult,¹ U.S. adults without a usual source of care increased from 21% in 2013 to 30% in 2022.² While the declining number of practicing PCPs in the U.S. is well-described, a recent trend in PCPs reducing their panel sizes calls for new methods to accurately measure the supply of primary care services.³ Additionally, while there is hope that primary care advanced practice providers (APPs) might mitigate the PCP shortage,¹ it is unclear whether this is happening. We examined trends in new patient primary care visits as a pragmatic measure of the absolute and relative supply of PCPs and primary care APPs.

Methods

This repeated cross-sectional study used data from Medicare fee-for-service (FFS) 100% administrative claims and Parts B and D public use files. The study population was primary care clinicians submitting FFS Medicare claims between 2013 and 2021.

We defined active primary care clinicians as family practice, general practice, geriatrics, internal medicine or preventive medicine physicians, and primary care APPs who submitted at least 50 outpatient visit evaluation and management (E/M) codes for FFS Medicare beneficiaries annually (eTable 1). Because APP specialty codes do not capture primary care specialization, we used a method validated in prior work⁵ to identify primary care APPs using Medicare Part D claims (eMethods).

At the national level by clinician type (PCP vs APP) and year, we captured per beneficiary counts of new visits (defined by E/M codes associated with new outpatient visits, eTable 1), counts of primary care clinicians available for new visits (billing >11 new visit E/M codes for FFS Medicare beneficiaries annually (eMethods, eTables 1-2)), percentage of new visits completed, and percentage of clinicians available for new visits. Because the percentage of Medicare beneficiaries enrolled in Medicare Advantage (MA) increased from 29% to 46% from 2013-2021,⁶ we performed a sensitivity analysis stratifying clinicians by their percentage of Part D claims attributed to MA (eMethods).

This study was determined exempt from review by the Institutional Review Board at Harvard Medical School and followed the STROBE reporting guideline. Analyses were done in R (v.4.4.1).

Results

In 2013, there were 75,140 PCPs and 12,768 primary care APPs who were available for new visits for Medicare FFS beneficiaries. By 2021, the number of PCPs available for new visits decreased 24.9% to 56,438, and APPs increased 91.4% to 24,432.

Per 10,000 FFS beneficiaries between 2013 and 2021, total clinicians available for new visits decreased from 23.4 to 22.2, new visits fell from 1,162 to 1,102, PCPs available for new visits decreased from 20.0 to 15.5 and APPs available for new visits increased from 3.4 to 6.7 (Figure 1).

Figure 1.

New Primary Care Visits and Primary Care Clinicians Available for New Visits per 10,000 Medicare Fee-For-Service Beneficiaries, 2013-2021

In 2013, 86.1% of new patient FFS Medicare visits were completed by PCPs, decreasing to 69.9% by 2021. The percentage of PCPs available for new visits decreased from 57.0% to 47.5% from 2013 to 2021 (Figure 2). The percentage of PCPs available for new visits from 2013 to 2021 declined across all quintiles of PCPs stratified by percentage of their Part D claims attributed to MA (eTable 3).

Figure 2.

Relative Trends in Primary Care New Visits Performed by Physicians vs APPs, 2013-2021

Discussion

In this analysis, we found a persistent decline in the number of PCPs available for new patient FFS Medicare visits between 2013 and 2021. The growth in APPs available to see new patients may only partially mitigate the decline in physicians, possibly leading to barriers ranging from longer wait times for new patient visits to limited ability to access a new provider altogether.

Study limitations include its restriction to the Medicare FFS population, which may have changed in unmeasured ways that affected patient demand during the study period and may experience greater morbidity and different barriers to care than other patient populations.

In conclusion, our findings are consistent with national data showing high rates of PCP turnover⁷ and burnout, and suggest that without intervention, the supply of new visits for FFS Medicare beneficiaries may continue to decline.