Trends in the Proportion of US and International Medical Graduates at the County Level

Abstract

This study assesses temporal trends in US medical graduates’ and international medical graduates’ relative contributions to the total physician workforce from 2010 to 2022, by rurality and across medically underserved areas.

The US faces a projected shortage of 90 000 physicians by the end of 2025.¹ Historically, international medical graduates (IMGs) partially filled these gaps, but over the last 15 years, little has been known about the role of IMGs compared with US medical graduates (USMGs).¹ From 2010 to 2022, the number of US medical schools including schools granting MD and DO degrees increased from 155 to 203, and the number of MD and DO graduates grew by 40.5%.² We investigated the temporal trends of USMGs’ and IMGs’ relative contributions to the total physician workforce, by rurality and across medically underserved areas (MUAs).

Methods

This retrospective serial cross-sectional study analyzed county-level physician (ie, MD and DO) count data by country of graduation (USMG vs IMG) from 2010 to 2022 using publicly available data from the Health Resources and Services Administration Area Health Resources Files data warehouse.³ During the study period, data were annually recorded for 3067 counties and county equivalents.

We tabulated the county-level unadjusted mean physician count and proportion of USMGs and IMGs each year and their 95% confidence intervals. We focused on overall physician counts per county, rather than physician-to-population ratios, as our focus was on aggregate trends and not on regional variance in physician sufficiency. To investigate the effect of rurality and MUA designation on the proportion of IMGs and USMGs, we fit a linear mixed regression model with independent variables: time, MUA, rurality, MUA × time, rurality × time, and MUA × rurality × time. We plotted the average marginal effects for IMG proportions. Data analysis was completed using SAS version 9.4 (SAS Institute Inc).

This study used publicly available county-level data and was not considered human subjects research by the Ohio State University Institutional Review Board.

Results

Among the 3067 counties, the unadjusted mean count of USMGs increased 24%, from 183 (95% CI, 155-211) in 2010 to 228 (95% CI, 200-256) in 2022. The mean IMG count increased 27%, from 58 (95% CI, 48-68) in 2010 to 74 (95% CI, 64-84) in 2022 (Figure 1). Over the follow-up period, the average proportion of IMGs remained consistently around 18% (P = .24 for time trend).

Figure 1. Unadjusted County-Level Mean Physician Count for USMGs and IMGs From 2010 to 2022 Nationally, in Urban Counties, and in Counties Designated to Have MUAs.

IMG indicates international medical graduate; MUA, medically underserved area; and USMG, US medical graduate.

During the study period, the mean population in rural counties (62% of counties) was approximately 25 000 (SD, 22 000) and physician density was approximately 10 (SD, 8) per 10 000 population. Urban counties (38% of counties) had a mean population of approximately 241 000 (SD, 513 000) and physician density of approximately 19 (SD, 21) per 10 000 population. Eighty-eight percent of rural counties and 86% of urban counties had MUAs. The highest proportion of IMGs (approximately 21%) was in urban counties that have MUAs (Figure 2). USMGs were more represented in rural counties, regardless of MUA status (approximately 85%). Compared with counties without MUAs, counties with MUAs have a small and consistently higher proportion of IMGs of about 0.5% (P = .003). There was no change in the proportions of USMGs or IMGs over time according to the county rural and MUA designation (P = .44).

Figure 2. Marginal Effects of County Designation as Rural or Medically Underserved on Proportion of IMGs.

IMG indicates international medical graduate; MUA, medically underserved area; and USMG, US medical graduate.

Percentage USMG = 1 − percentage IMG. Linear regression mixed model: percentage IMG = 0.8442 + −0.00110 × time + −0.04619 × MUA + 0.002986 × rurality + 0.01997 × MUA × rurality + 0.000796 × MUA × time + −0.000942 × rurality × time + 0.001124 × MUA × rurality × time. MUA was the only significant term (P = .003).

Discussion

In this nationwide study of county-level physician data, we found that the proportion of IMGs in the physician workforce was relatively consistent from 2010 to 2022. Furthermore, we found significant county-level variation in IMG vs USMG representation, with IMGs having a slightly greater role than USMGs in counties with MUAs, particularly urban counties with MUAs, whereas USMGs have a larger role in rural counties. Our analysis updates the relative contribution of IMGs to the medical workforce in rural counties and MUAs over the last 15 years.⁴ In prior work, IMGs were known to have greater primary care representation compared with USMGs⁵; however, our data have insufficient granularity to update these trends. Proposed national policies restricting work visas for IMGs may worsen projected physician shortages and have greater impact within MUAs.⁶ In conclusion, while the number of USMGs over the last decade has increased, IMGs still play a substantial overall role, and a consistently greater role in counties with MUAs.

Supplement.

Data Sharing Statement