Abstract
Importance
The number of both US citizen and noncitizen international medical graduates (IMGs) practicing in the United States has grown substantially over the last two decades, but little is known about how these groups differ in clinical specialty and practice location.
Objective
To evaluate the effects of citizenship status on IMG specialty choice (primary care vs nonprimary care) and practice location (rural and health professional shortage areas [HPSAs]).
Design
To analyze the retrospective, cross sectional data between 2010 and 2019, we estimated a series of multivariate logistic regressions with year fixed effects after assessing trends in new physician specialty choice and practice location.
Setting
Survey of Residents Completing Training in New York, the largest publicly available annual survey of graduating residents and fellows in the US.
Participants
15,133 new physicians who accepted a job offer in our study period, including 8,177 US medical graduates (USMGs), 2,753 US citizen IMGs, 1,057 permanent resident IMGs, and 3,146 noncitizen non-permanent resident (NCNP) IMGs.
Exposure
IMG and Citizenship status.
Main Outcomes
Whether a new physician chose to practice primary care; whether a new primary care physician chose to work in a rural area; whether a new primary care physician chose to work in a HPSA.
Results
Compared to 2010, there were 20% fewer NCNP IMGs entering primary care in 2019. Our logistic analysis found that US citizen IMGs were five times, permanent resident IMGs seven times, and NCNP IMGs nine times more likely to enter primary care compared to USMGs. Among new primary care physicians, NCNP IMGs were more likely to work in a rural area compared to USMGs while US citizen IMGs and permanent resident IMGs acted similarly to USMGs. NCNP IMGs were five times more likely to work in a HPSA while US citizen IMGs and permanent resident IMGs were less likely, compared to USMGs.
Conclusion
There was a declining proportion of NCNP IMGs entering primary care; citizenship status affected IMGs’ specialty and practice location choices with NCNP IMGs more likely to choose primary care and work in rural areas and HPSAs.
Supplementary Information
The online version contains supplementary material available at 10.1007/s11606-024-09123-9.
INTRODUCTION
Over the last two decades, the number of international medical graduates (IMGs)—those who graduated from a medical school located outside the United States and Canada— practicing in the United States have grown by 18% and now comprise a quarter of all licensed physicians.1 The largest share of IMGs are graduates of medical schools in India (23%), the Caribbean (18%), Philippines (6%), and Mexico (5%).2 The number of IMGs who attended Caribbean medical schools have increased by 115% from a decade prior.3 In fact, US citizens have made up a larger proportion of all graduates from Caribbean medical schools, increasing from 48% in 2010 to 67% in 2022.3 However, there has been minimal research to understand the difference in job market behavior between US citizen and noncitizen IMGs.
IMGs play a vital role across community hospitals and academic medical centers. Studies have shown that patients treated by IMGs and US medical graduates (USMGs) have similar outcomes as general internists.4 IMGs are also more likely to work at a critical access hospital5, in underserved and rural areas6, and as primary care physicians compared to USMGs.6 Furthermore, more than 18% of academic physicians are IMGs.7
While previous studies have generally compared IMGs to USMGs, an estimated 58% of IMGs are born in the United States and have US citizenship with the remaining 42% born outside the United States.1 IMGs are not homogenous in terms of citizenship status which differentially affects their ability to enter and remain in the United States after medical school. This divergence could affect where IMGs practice and the type of specialties that they enter.
For US citizen IMGs, they must pass licensing requirements established by the Education Committee on Foreign Medical Graduates (ECFMG) and successfully match to a US residency. Noncitizen IMGs have the same ECFMG and residency requirements but also require visa sponsorship for legal entry and post-graduate medical training in the US. Noncitizen IMGs are eligible for a J1 trainee visa sponsored by ECFMG or a H1B visa sponsored by individual hospitals which require filling fees.8 The majority of noncitizen IMGs enter the country through a J1 visa for their postgraduate clinical training. However, the J1 visa requires a period of time during which the visa holders must return to their country of origin before applying for H1B to practice medicine in the United States. The Conrad 30 waiver program allows noncitizen IMGs to convert their J1 visa to H1B without the return to origin requirement as long as they work in a Health Professional Shortage Area (HPSA) or for a medically underserved area/population defined by the Health Resources and Services Administration (HRSA). Previous research has identified the Conrad program as the largest supplier of physicians to HPSAs and a critical tool in alleviating the domestic physician shortage, since it was established in 1994.9,10
However, the effect of citizenship status on IMGs’ specialty choice and practice location are not well known. Recent policies including state licensure laws (e.g. Tennessee SB 1451), board eligibility policies (e.g. American Board of Dermatology, American Board of Internal Medicine), and the recent increase in US citizens attending medical school abroad have expanded the recruitment of IMGs to the US.11–13 After Tennessee, other states (e.g., Florida and Illinois) also passed similar laws, allowing IMGs who have not completed a US residency to practice medicine with a two-year provisional license under the supervision of a licensed attending physician in these states with the goal of mitigating physician shortages. The IMGs may obtain full licensure after the two-year provisional licensure period. Given these recent expansions of IMG practice eligibility, it is important to have a better understanding of IMG physicians’ choices of specialty and practice location. In this study, we describe trends in new physicians entering primary care and their practice locations and assess the effect of US citizenship status on new IMG physicians’ choice of specialty and practice locations.
METHODS
This retrospective analysis employed repeated cross-sectional data from the 2010–2019 Survey of Residents Completing Training in New York (New York Resident Exit Survey). This survey is conducted annually by the Center for Workforce Studies at the University at Albany, State University of New York, which fields the survey to all graduating residents and fellows in the state. Over our study period, the survey had a response rate of 60%. As the largest publicly available survey that provides consistent information over the past 25 years about new physicians’ characteristics and first practices, the New York Resident Exit Survey has been used by previous studies to assess physician workforce characteristics and impacts of health policy interventions.14–17
Like prior research13, we defined new physicians as those graduating residents and fellows who were searching for and accepted a job offer and primary care physicians as physicians of general internal medicine, family medicine, and general pediatrics. We used IMG-citizenship status to classify all new physicians into four groups— US medical graduates (USMGs), US citizen IMGs, US permanent resident IMGs, and non-citizen non-permanent resident IMGs (NCNP IMG). NCNP IMGs were defined as IMGs who were either J-1 or H-1B visa holders. Rural practice was self-identified by answering the following question: “Which best describes the demographics of the area in which you will be practicing? Rural, suburban, inner city, small city, major city.” HPSA was self-identified by answering the following question: “Is this principal practice address located in a federally designed Health Professional Shortage Area?”.
Data Analysis
We assessed overall trends over the last decade in new physicians’ specialty choice and practice location, and whether the trends differ by citizenship status. We then performed multivariate logistic regressions on the effects of citizenship status on whether a new physician chose a primary care specialty, whether a new primary care physician chose to work in a rural area, and whether a new primary care physician chose to work in a HPSA area. In our supplementary tables, we also examine whether a new primary care physician chose to work in other geographic settings including inner city, major city, suburban, or small city. Each logistic regression included year fixed effects and the following covariates: gender, age, race (Non-Hispanic White, Non-Hispanic Asian, Non-Hispanic Black, Non-Hispanic Other, Hispanic), level of debt (no debt, < $50,000, $50,000-$350,000, > $350,000), IMG-citizenship status, and year-IMG-citizenship interaction term.
The data were publicly available without personal identifiers; this study was determined as not human subjects research by the corresponding author’s institutional review board. Our study followed the STROBE reporting guideline for cross-sectional studies. All tests were 2 sided, used an alpha level of 0.05, and were conducted in Stata version 18.0 (StataCorp).
RESULTS
Our study sample included 15,133 new physicians entering the workforce over the last decade. Of these, 8,177 were USMGs, 2,753 were US citizen IMGs, 1,057 were permanent resident IMGs, and 3,146 were NCNP IMGs. Table 1 describes the characteristics of the study sample. Approximately, 47.91% of USMGs were male compared to 56.45% for US citizen IMGs, 51.47% for permanent resident IMGs, and 45.84% for NCNP IMGs. USMGs are slightly younger (mean = 32.20 years) than US citizen IMGs (mean = 34.28 years), permanent resident IMGs (mean = 35.76 years) or NCNP IMGs (mean = 33.13 years). The majority of USMGs are non-Hispanic Whites (55.71%) while non-Hispanic Asians account for the largest group among US citizen IMGs (37.52%), permanent resident IMGs (46.22%) and NCNP IMGs (44.65%). USMGs had higher educational debt with 49.29% of them having debt between $50,000-$350,000. In comparison, 47.99% of US citizen IMGs, 77.18% of permanent resident IMGs, and 75.41%% of NCNP IMGs had no debt.
Table 1.
Characteristics of the Study Sample by International Medical Graduate status
| US Medical Graduates | US Citizen IMGs | Permanent Resident IMGs | Noncitizen Non-Permanent Resident IMGs | P value | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Gender | ||||||||||
| Male | 3,918 | 47.91% | 1,554 | 56.45% | 544 | 51.47% | 1,442 | 45.84% | ||
| Female | 4,259 | 52.09% | 1,199 | 43.55% | 513 | 48.53% | 1,055 | 33.53% | ||
| Subtotal | 8,177 | 2,753 | 1,057 | 3,146 | p < 0.001 | |||||
| Age | Mean (SD) | 32.20 (3.21) | 34.28 (4.60) | 35.76 (4.74) | 33.13 (3.65) | p < 0.001 | ||||
| Educational Debt | ||||||||||
| No debt | 1,562 | 23.36% | 1,048 | 47.99% | 727 | 77.18% | 1,599 | 74.41% | ||
| < $50,000 | 366 | 5.47% | 182 | 8.33% | 95 | 10.08% | 245 | 11.40% | ||
| $50,000-$350,000 | 3,296 | 49.29% | 572 | 26.19% | 102 | 10.83% | 271 | 12.61% | ||
| > $350,000 | 1,463 | 21.88% | 382 | 17.49% | 18 | 1.91% | 34 | 1.58% | ||
| Subtotal | 6,687 | 2,184 | 942 | 2,149 | p < 0.001 | |||||
| Race | ||||||||||
| Non-Hispanic White | 4,472 | 55.71% | 926 | 34.09% | 194 | 20.38% | 353 | 16.43% | ||
| Non-Hispanic Asian | 2,021 | 25.17% | 1,019 | 37.52% | 440 | 46.22% | 959 | 44.65% | ||
| Non-Hispanic Black | 532 | 6.63% | 152 | 5.60% | 86 | 9.03% | 134 | 6.24% | ||
| Non-Hispanic Other | 428 | 5.33% | 311 | 11.45% | 131 | 13.76% | 477 | 22.21% | ||
| Hispanic | 575 | 7.16% | 308 | 11.34% | 101 | 10.61% | 225 | 10.47% | ||
| Subtotal | 8,028 | 2,716 | 952 | 2,148 | p < 0.001 | |||||
| Practice Location | ||||||||||
| Inner City | 2558 | 32.69% | 759 | 28.72% | 356 | 34.97% | 718 | 30.12% | ||
| Major City | 1889 | 24.14% | 522 | 19.75% | 204 | 20.04% | 378 | 15.86% | ||
| Suburban | 2434 | 31.10% | 916 | 34.66% | 305 | 29.96% | 522 | 21.90% | ||
| Small City | 677 | 8.65% | 329 | 12.45% | 124 | 12.18% | 533 | 22.36% | ||
| Rural | 268 | 3.42% | 117 | 4.43% | 29 | 2.85% | 233 | 9.77% | ||
| Subtotal | 7826 | 2643 | 1018 | 2384 | p < 0.001 | |||||
| New Primary Care Physicians (PCPs) | 1,690 | 20.59% | 1,206 | 43.51% | 456 | 46.91% | 1,112 | 50.52% | p < 0.001 | |
| New PCPs in Rural Ares | 85 | 5.03% | 60 | 4.98% | 19 | 4.17% | 109 | 9.80% | p < 0.001 | |
| New PCPs in HPSAs | 281 | 16.63% | 155 | 12.85% | 33 | 7.24% | 473 | 42.54% | p < 0.001 | |
In 2010, 54.83% of new physicians were USMGs compared to 17.50% US citizen IMGs, 8.68% permanent resident IMGs, and 18.98% NCNP IMGs. In 2019, the proportion of USMGs reached 63.36%, an increase of 8.53% from 2010, while the proportion of NCNP IMGs fell by 6.07% to 12.92%. Throughout the study period, US citizen IMGs remained relatively stable comprising of 18.86% of new physicians in 2019, while permanent resident IMGs decreased to 4.86% (Fig. 1A). Approximately 31.71% of all new physicians entered primary care in 2019. Among USMGs, 23.11% chose primary care in 2019 with an increase of 7.15% from 2010. 39.11% of NCNP IMGs entered primary care in 2019, a 20% decline from a decade earlier. Both US citizen IMGs and permanent resident IMGs also saw declines with 37.29% and 38.16% respectively entering primary care in 2019 (Fig. 1B).
Figure 1.
A New physician composition by international medical graduate status, 2010–2019, (%). Figure note: We used linear probability models to test the trends over time (Coef: -0.003, 95% CI, -0.006, 0.003, p = 0.07). B Percentage of new physicians entering primary care by international medical graduate status. Figure note: We used linear probability models to test the trends over time (Coef: -0.005, 95% CI, -0.007, -0.002, p < 0.001). C Percentage of new primary care physicians working in rural areas by international medical graduate status. Figure note: We used linear probability models to test the trends over time (Coef: 0.004, 95% CI, 0.002, 0.006, p < 0.001). D Percentage of new primary care physicians working in health professional shortage areas by international medical graduate status. Figure note: We used linear probability models to test the trends over time (Coef: 0.005, 95% CI, -0.002, 0.011, p = 0.15).
Among new primary care physicians, the proportion of them working in rural areas grew. In 2019, 8.22% of USMG primary care physicians worked in a rural area, an increase of 2.96% from 2010, while the percentage of US citizen IMGs grew from 5.32% in 2010 to 14.29% in 2019. Among permanent resident IMGs in primary care, 14.81% worked in rural areas in 2019 compared to 1.85% in 2010. Among NCNP IMGs in primary care, 13.92% worked in rural areas in 2019 compared to 7.43% in 2010 (Fig. 1C). The proportion of new primary care physicians who were USMGs and chose to work in HPSAs increased from 15.79% in 2010 to 18.26% in 2019, compared with a decrease for US citizen IMGs from 12.77% in 2010 to 10.48% in 2019 and relatively stable proportions for permanent resident IMGs (12.96% in 2010 vs 11.11% in 2019) and NCNP IMGs (45.95% in 2010 vs. 43.04% in 2019) (Fig. 1D).
Our logistic regression models showed that US citizen IMGs were five times (Adjusted Odds Ratio [AOR] = 5.10, 95% CI, 3.47–7.49), permanent resident IMGs seven times (AOR = 7.52, 95% CI, 4.73–11.97), and NCNP IMGs nine times (AOR = 8.87, 95% CI, 6.26–12.58) more likely to enter primary care compared to USMGs, controlling for gender, age, race, and educational debt. Additionally, physicians of non-Hispanic Asian (AOR = 1.34, 95% CI, 1.21–1.49), non-Hispanic Black (AOR = 2.15, 95% CI, 1.81–2.56), and Hispanic origin (AOR = 1.56, 95% CI, 1.33–1.82) were more likely to enter primary care compared to non-Hispanic White physicians (Table 2). When analyzing the IMG-year interaction term, we found that each year from 2014–2019, there has been a significant decline in NCNP IMGs entering primary care compared to the trend in USMGs entering primary care over the same time period (each AOR < 1.0). To help interpret the interaction term, we used the Stata Margins command to create an adjusted figure (Fig. 2) showing a downward trend in NCNP IMGs entering primary care, which is consistent with Fig. 1B.
Table 2.
Characteristics of New Physicians Entering Primary Care–Results from a Multivariate Logistic Regression
| ADJUSTED ODDS RATIO | P VALUE | 95% CONFIDENCE INTERVAL | |||
|---|---|---|---|---|---|
| GENDER | |||||
| Male | [Reference] | ||||
| Female | 1.17 | < 0.001 | 1.08 | 1.28 | |
| AGE | 0.87 | < 0.001 | 0.86 | 0.88 | |
| MEDICAL GRADUATE STATUS | |||||
| US Medical Graduate | [Reference] | ||||
| US Citizen International Medical Graduate | 5.10 | < 0.001 | 3.47 | 7.49 | |
| Permanent Resident International Medical Graduate | 7.52 | < 0.001 | 4.73 | 11.97 | |
| Noncitizen Non-Permanent Resident International Medical Graduate | 8.87 | < 0.001 | 6.26 | 12.58 | |
| EDUCATIONAL DEBT | |||||
| No debt | [Reference] | ||||
| < $50,000 | 1.01 | 0.891 | 0.86 | 1.19 | |
| $50,000-$350,000 | 0.92 | 0.138 | 0.82 | 1.03 | |
| > $350,000 | 1.03 | 0.678 | 0.90 | 1.18 | |
| RACE AND ETHNICITY | |||||
| Non-Hispanic White | [Reference] | ||||
| Non-Hispanic Asian | 1.34 | < 0.001 | 1.21 | 1.48 | |
| Non-Hispanic Black | 2.15 | < 0.001 | 1.81 | 2.56 | |
| Non-Hispanic Other | 1.15 | 0.078 | 0.98 | 1.34 | |
| Hispanic | 1.56 | < 0.001 | 1.33 | 1.82 | |
| Medical Graduate Status—Year Interaction | |||||
| US CITIZEN IMG | 2011 | 1.06 | 0.826 | 0.62 | 1.81 |
| US CITIZEN IMG | 2012 | 1.01 | 0.959 | 0.60 | 1.72 |
| US CITIZEN IMG | 2013 | 0.81 | 0.442 | 0.47 | 1.39 |
| US CITIZEN IMG | 2014 | 1.10 | 0.732 | 0.64 | 1.88 |
| US CITIZEN IMG | 2015 | 0.91 | 0.734 | 0.52 | 1.59 |
| US CITIZEN IMG | 2016 | 0.48 | 0.011 | 0.27 | 0.85 |
| US CITIZEN IMG | 2017 | 0.67 | 0.101 | 0.42 | 1.08 |
| US CITIZEN IMG | 2018 | 0.65 | 0.075 | 0.40 | 1.04 |
| US CITIZEN IMG | 2019 | 0.49 | < 0.001 | 0.30 | 0.79 |
| PERMENENT RESIDENT IMG | 2011 | 1.05 | 0.889 | 0.56 | 1.97 |
| PERMENENT RESIDENT IMG | 2012 | 1.32 | 0.399 | 0.69 | 2.51 |
| PERMENENT RESIDENT IMG | 2013 | 1.17 | 0.623 | 0.62 | 2.21 |
| PERMENENT RESIDENT IMG | 2014 | 0.66 | 0.238 | 0.34 | 1.31 |
| PERMENENT RESIDENT IMG | 2015 | 0.58 | 0.128 | 0.29 | 1.17 |
| PERMENENT RESIDENT IMG | 2016 | 0.70 | 0.353 | 0.33 | 1.49 |
| PERMENENT RESIDENT IMG | 2017 | 0.48 | 0.029 | 0.24 | 0.93 |
| PERMENENT RESIDENT IMG | 2018 | 0.38 | 0.005 | 0.20 | 0.75 |
| PERMENENT RESIDENT IMG | 2019 | 0.36 | 0.004 | 0.18 | 0.72 |
| NCNP IMG | 2011 | 0.62 | 0.051 | 0.39 | 1.00 |
| NCNP IMG | 2012 | 0.96 | 0.851 | 0.59 | 1.54 |
| NCNP IMG | 2013 | 0.68 | 0.126 | 0.41 | 1.12 |
| NCNP IMG | 2014 | 0.53 | 0.012 | 0.32 | 0.87 |
| NCNP IMG | 2015 | 0.44 | < 0.001 | 0.26 | 0.74 |
| NCNP IMG | 2016 | 0.36 | < 0.001 | 0.21 | 0.61 |
| NCNP IMG | 2017 | 0.26 | < 0.001 | 0.16 | 0.41 |
| NCNP IMG | 2018 | 0.33 | < 0.001 | 0.20 | 0.54 |
| NCNP IMG | 2019 | 0.26 | < 0.001 | 0.16 | 0.42 |
Figure 2.
Adjusted percentage of new physicians entering primary care by international medical graduate status. Figure note: The percentage of new physicians entering primary care by international medical graduate status were adjusted by gender, age, educational debt level and race and ethnicity.
Among new primary care physicians, NCNP IMGs were more likely (AOR = 2.85, 95% CI, 1.89–4.29) to work in rural areas compared to USMGs, while US citizen IMGs and permanent resident IMGs acted similarly to USMGs (Table 3). Our supplementary Tables 1–4 show that NCNP IMGs were more likely to work in a small city (AOR: 2.75, 95% CI, 2.02, 3.74) with a population less than 50,000 people but were less likely to work in an inner city (AOR: 0.55, 95% CI, 0.44, 0.69) or major city (AOR: 0.69, 95% CI, 0.52, 0.91) compared to USMGs.
Table 3.
Characteristics of New Primary Care Physicians Working in Rural Areas–Results from a Multivariate Logistic Regression
| ADJUSTED ODDS RATIO | P VALUE | 95% CONFIDENCE INTERVAL | |||
|---|---|---|---|---|---|
| GENDER | |||||
| Male | [Reference] | ||||
| Female | 0.75 | 0.043 | 0.57 | 0.99 | |
| AGE | 1.02 | 0.366 | 0.98 | 1.05 | |
| MEDICAL GRADUATE STATUS | |||||
| US Medical Graduate | [Reference] | ||||
| US Citizen International Medical Graduate | 1.13 | 0.562 | 0.75 | 1.68 | |
| Permanent Resident International Medical Graduate | 1.12 | 0.711 | 0.62 | 2.01 | |
| Noncitizen Non-Permanent Resident International Medical Graduate | 2.85 | < 0.001 | 1.89 | 4.29 | |
| EDUCATIONAL DEBT | |||||
| No debt | [Reference] | ||||
| < $50,000 | 0.79 | 0.376 | 0.47 | 1.33 | |
| $50,000-$350,000 | 0.99 | 0.966 | 0.68 | 1.44 | |
| > $350,000 | 1.83 | 0.009 | 1.16 | 2.86 | |
| RACE AND ETHNICITY | |||||
| Non-Hispanic White | [Reference] | ||||
| Non-Hispanic Asian | 1.05 | 0.769 | 0.74 | 1.49 | |
| Non-Hispanic Black | 1.15 | 0.606 | 0.68 | 1.92 | |
| Non-Hispanic Other | 0.54 | 0.026 | 0.31 | 0.93 | |
| Hispanic | 1.26 | 0.332 | 0.79 | 2.01 | |
US citizen IMGs (AOR = 0.66, 95% CI, 0.48–0.90) and permanent resident IMGs (AOR = 0.37, 95% CI, 0.23–0.61) were less likely while NCNP IMGs were five times (AOR = 5.52, 95% CI, 4.07–7.48) more likely to work in HPSAs compared to USMGs. Physicians with greater than $350,000 educational debt (AOR = 1.68, 95% CI, 1.18–2.40) were also more likely to work in HPSAs. Finally, physicians of non-Hispanic Black (AOR = 2.51, 95% CI, 1.68–3.76) and Hispanic origin (AOR = 2.70, 95% CI, 1.90–3.83) were more likely to work in HPSAs compared to non-Hispanic White primary care physicians (Table 4).
Table 4.
Characteristics of New Primary Care Physicians Working in Health Professional Shortage Areas–Results from a Multivariate Logistic Regression
| ADJUSTED ODDS RATIO | P VALUE | 95% CONFIDENCE INTERVAL | ||
|---|---|---|---|---|
| GENDER | ||||
| Male | [Reference] | |||
| Female | 1.22 | 0.059 | 0.99 | 1.50 |
| AGE | 1.03 | 0.012 | 1.01 | 1.06 |
| MEDICAL GRADUATE STATUS | ||||
| US Medical Graduate | [Reference] | |||
| US Citizen International Medical Graduate | 0.66 | 0.009 | 0.48 | 0.90 |
| Permanent Resident International Medical Graduate | 0.37 | < 0.001 | 0.23 | 0.61 |
| Noncitizen Non-Permanent Resident International Medical Graduate | 5.52 | < 0.001 | 4.07 | 7.48 |
| EDUCATIONAL DEBT | ||||
| No debt | [Reference] | |||
| < $50,000 | 1.22 | 0.299 | 0.84 | 1.77 |
| $50,000-$350,000 | 1.44 | 0.011 | 1.09 | 1.91 |
| > $350,000 | 1.68 | < 0.001 | 1.18 | 2.40 |
| RACE AND ETHNICITY | ||||
| Non-Hispanic White | [Reference] | |||
| Non-Hispanic Asian | 1.15 | 0.308 | 0.88 | 1.50 |
| Non-Hispanic Black | 2.51 | < 0.001 | 1.68 | 3.76 |
| Non-Hispanic Other | 1.31 | 0.141 | 0.91 | 1.88 |
| Hispanic | 2.70 | < 0.001 | 1.90 | 3.83 |
DISCUSSION
Overall, our results indicate that citizenship status has significant effects on IMGs’ choices of specialties and practice locations. Over the last decade, the percentage of NCNP IMG new physicians has declined, echoing the previously reported decrease in noncitizen IMGs due to the rapid growth of USMGs and US citizen IMGs.18 Furthermore, we find two important trends: the declining proportion of NCNP IMGs entering primary care and the increased likelihood of NCNP IMGs practicing in rural areas and HPSAs.
Historically, IMGs have been entering primary care at a rate much higher than USMGs.19 While IMGs continue to enter primary care at a higher rate, that rate has decreased over the past decade as our analysis revealed. To our knowledge, no prior study has highlighted this declining proportion of IMGs entering primary care. Coupled with an increase in unmatched spots in primary care, this trend could affect long-term primary care access in the US.20 In certain states such as Massachusetts, visa incentives through the Conrad 30 waiver program prioritize NCNP IMG primary care physicians.21 However, those incentives may not be sufficient to meet the growing demand for primary care.22,23 A combination of factors including burnout, low reimbursement, complex chronic disease management, and dissatisfaction have driven the primary care physician shortage.24 Prioritizing primary care physicians across states through the Conrad program or creating new channels for NCNP IMG to practice primary care in the US could bolster the primary care workforce.
The percentage of primary care physicians working in rural areas grew over our study period, likely contributing to recent increases in the primary care workforce in rural areas.25 However, rural–urban disparities in primary care physicians remain. Citizenship status affects practice location among new primary care physicians as our study found that US citizen IMGs and permanent resident IMGs were similarly likely to practice in rural areas compared to USMGs and less likely to work in HPSAs while NCNP IMGs were more likely to practice in both rural areas and HPSAs than USMGs. These differences indicate that strategies to recruit primary care physicians to underserved areas must be tailored to the unique characteristics of new physicians. For example, expansion of the Conrad 30 waiver program and encouragement to states to fill all their waiver spots can help recruit more NCNP IMGs to rural areas and HPSAs. Furthermore, states with provisional licensing pathways such as Tennessee should require IMGs to work in rural areas and HPSAs to increase physician workforce in those areas.9 However, there is much uncertainty about the costs and benefits of state provisional licensure policies, and additional evaluation is needed prior to national adoption. Given our finding of the significantly positive relationship between educational debt and new primary care physicians’ likelihood to work in a HPSA, debt relief programs can target US citizens and permanent resident IMGs who graduate from Caribbean or other for-profit international medical schools with substantial debt.26 Expansion of the National Health Services Corp and targeted recruitment at large Caribbean medical schools could encourage US citizen IMGs and permanent resident IMGs to work in rural areas and HPSAs. Additionally, Caribbean medical schools in need of clinical rotation sites could partner with accredited rural hospitals in the US to provide their graduates with rural training experience and potentially increase their chances to practice in rural areas. Prior studies have proved effectiveness of rural partner training programs, such as their success in placing physicians in rural communities and improving sustainable recruitment and retention to such areas.27 Federal-state partnerships to provide low-interest loans to rural primary care providers can also provide financial benefit for physicians to practice in underserved areas.
Limitations of this study include its cross-sectional nature, the survey limited to new physicians produced in one state, and that initial practice location may not fully correspond to long-term practice location choice. However, a cohort study of family medicine physicians found that the majority of physicians initially practicing in rural areas remained at that clinical site six years later.28 Further research should study IMG retention in rural areas.
CONCLUSION
Our findings confirm the important role that noncitizen IMGs play in providing primary care, especially in HPSAs and rural areas. While visa incentive programs may drive noncitizen IMGs to work in those areas, the declining proportion of them entering primary care indicates the need for policy interventions to make primary care more attractive to new physicians more generally. Further studies should explore the differences in subspeciality, salaries, job satisfaction, and incentives between US citizen, permanent resident, and NCNP IMGs. Future research also needs to assess the impacts of new state laws (e.g., Tennessee SB 1451, Florida SB 7016)9,10 that offer provisional licensing pathways to IMGs. As states increase IMG recruitment through these new laws to mitigate their physician shortages, the divergence in practice specialty and location between US citizen, permanent resident, and NCNP IMGs may become increasingly important.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements:
This study was supported by the National Institute on Minority Health and Health Disparities under Award Number R01MD013736 and the National Institute of Nursing Research under Award Number R01NR020859. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, Harvard Medical School, Harvard Pilgrim Health Care Institute, or the Center for Health Workforce Studies at University at Albany, State University of New York.
Funding
National Institute on Minority Health and Health Disparities and National Institute of Nursing Research.
Data Availability
Data was publically available.
Declarations:
Conflict of Interests:
The authors have no conflict of interests to report.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.American Medical Association. How IMGs have changed the face of American medicine. October 19, 2021. Available at: https://www.ama-assn.org/education/international-medical-education/how-imgs-have-changed-face-american-medicine. Accessed 3 Apr 2024.
- 2.Young A, Chaudhry H Pei X, Arnhart K, Dugan M, Steingard S. FSMB Census of Licensed Physicians in the United States, 2018. Journal of Medical Regulation VOL 105, NO 2. 2019 Available at: https://www.fsmb.org/siteassets/advocacy/publications/2018census.pdf. Accessed 3 Apr 2024.
- 3.Physician Licensure in 2022. Federation of State Medical Boards. Available at: https://www.fsmb.org/u.s.-medical-regulatory-trends-and-actions/u.s.-medical-licensing-and-disciplinary-data/physician-licensure/. Accessed October 2023.
- 4.Tsugawa Y, Jena AB, Orav EJ, Jha AK. Quality of care delivered by general internists in US hospitals who graduated from foreign versus US medical schools: observational study. BMJ. 2017;356:j273. Published 2017 Feb 2. 10.1136/bmj.j273. [DOI] [PMC free article] [PubMed]
- 5.Hagopian A, Thompson MJ, Kaltenbach E, Hart LG. The role of international medical graduates in America's small rural critical access hospitals. J Rural Health. 2004;20(1):52-58. 10.1111/j.1748-0361.2004.tb00007.x. [DOI] [PubMed] [Google Scholar]
- 6.Thompson MJ, Hagopian A, Fordyce M, Hart LG. Do international medical graduates (IMGs) “fill the gap” in rural primary care in the United States? A national study. J Rural Health. 2009;25(2):124–134. 10.1111/j.1748-0361.2009.00208.x. [DOI] [PubMed] [Google Scholar]
- 7.Khullar D, Blumenthal DM, Olenski AR, Jena AB. U.S. immigration policy and American medical research: The scientific contributions of Foreign Medical Graduates. Ann Intern Med. 2017;167(8):584-586. 10.7326/M17-1304. [DOI] [PubMed] [Google Scholar]
- 8.Al Ashry HS, Kaul V, Richards JB. The implications of the current visa system for foreign medical graduates during and after graduate medical education training. J Gen Intern Med. 2019;34(7):1337-1341. 10.1007/s11606-019-05027-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ramesh T, Brotherton SE, Wozniak GD, Yu H. Evaluation of the Conrad 30 waiver program’s success in attracting international medical graduates to underserved areas. JAMA Health Forum. 2023;4(7):e232021. 10.1001/jamahealthforum.2023.2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ramesh T, Brotherton SE, Wozniak GD, Yu H. An increasing number of states filled Conrad 30 waivers for recruiting international medical graduates. Health Aff Sch. 2024;2(9):qxae103. Published 2024 Aug 19. 10.1093/haschl/qxae103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Venkatesh K, Wyant WA, Zhang AL, Ramesh T, Nambudiri VE. Bringing international medical graduates into the dermatology workforce: New licensing pathways to expand dermatologist supply and access. J Am Acad Dermatol. Published online August 22, 2024. 10.1016/j.jaad.2024.07.1479. [DOI] [PubMed]
- 12.Ramesh T, Horvitz-Lennon M, Yu H. Opening the door wider to international medical graduates - The Significance of a New Tennessee Law. N Engl J Med. 2023;389(21):1925-1928. 10.1056/NEJMp2310001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Fiore K. More States Cut Training Requirements for Some International Medical Graduates . MedPage Today. March 14, 2024. Available at: https://www.medpagetoday.com/special-reports/exclusives/109168?trw=no. Accessed 3 Apr 2024.
- 14.Cleaves M, Stiegler K, Martiniano R, Moore J. The health care workforce in New York State: Trends in the supply of and demand for health workers. Rensselaer, NY: Center for Health Workforce Studies, School of Public Health, SUNY Albany; March 2022. [Google Scholar]
- 15.Chen A, Lo Sasso AT, Richards MR. Supply-side effects from public insurance expansions: Evidence from physician labor markets. Health Econ. 2018;27(4):690-708. 10.1002/hec.3625. [DOI] [PubMed] [Google Scholar]
- 16.Ma Y, Armstrong D, Forte GJ, Yu H. Effects of the affordable care act medicaid expansion on the compensation of new primary care physicians. Med Care. 2022;60(8):636-644. 10.1097/MLR.0000000000001740. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lo Sasso AT, Armstrong D, Forte G, Gerber SE. Differences in starting pay for male and female physicians persist; explanations for the gender gap remain elusive. Health Aff. 2020;39(2):256-263. [DOI] [PubMed] [Google Scholar]
- 18.Traverso G, McMahon GT. Residency training and international medical graduates: coming to America no more. JAMA. 2012;308(21):2193-4. 10.1001/jama.2012.14681. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Salsberg E, Nolan J. The Posttraining Plans of International Medical Graduates and US Medical Graduates in New York State. JAMA. 2000;283(13):1749–1750. 10.1001/jama.283.13.1749-JMS0405-5-1. [PubMed] [Google Scholar]
- 20.Mitchell D; AAFP News. Match Day 2023. Ann Fam Med. 2023;21(3):288-289. 10.1370/afm.2989. [Google Scholar]
- 21.MASSACHUSETTS CONRAD 30/J-1 VISA WAIVER PROGRAM POLICY. Mass Department of Public Health. September 2023. Available at: https://www.mass.gov/doc/massachusetts-conrad-30-j-1-visa-waiver-program-policy-0/download. Accessed 3 Apr 2024.
- 22.Liu M, Wadhera RK. Primary care physician supply by county-level characteristics, 2010-2019. JAMA. 2022;328(19):1974–1977. 10.1001/jama.2022.15106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Gottlieb J, Polyakova M, Rinz K, Shiplett H, Udalova V. Who values Human Capitalists' Human Capital? The earnings and labor supply of U.S. Physicians. National Bureau of Economic Research. 2023. 10.3386/w31469.
- 24.McKinlay J, Marceau L. When there is no doctor: reasons for the disappearance of primary care physicians in the US during the early 21st century. Soc Sci Med. 2008;67(10):1481-1491. 10.1016/j.socscimed.2008.06.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Zhang D, Son H, Shen Y, et al. Assessment of changes in rural and urban primary care workforce in the United States From 2009 to 2017. JAMA Netw Open. 2020;3(10):e2022914. 10.1001/jamanetworkopen.2020.22914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Halperin EC, Goldberg RB. Offshore medical schools are buying clinical clerkships in U.S. hospitals: The problem and potential solutions. Acad Med. 2016;91(5):639-644. 10.1097/ACM.0000000000001128. [DOI] [PubMed] [Google Scholar]
- 27.MacQueen IT, Maggard-Gibbons M, Capra G, Raaen L, Ulloa JG, Shekelle PG, Miake-Lye I, Beroes JM, Hempel S. Recruiting rural healthcare providers today: a systematic review of training program success and determinants of geographic choices. J Gen Intern Med. 2018;33(2):191-199. 10.1007/s11606-017-4210-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Frisch L, Kellerman R, Ast T. A cohort study of family practice residency graduates in a predominantly rural state: initial practice site selection and trajectories of practice movement. J Rural Health. 2003;19(1):47-54. 10.1111/j.1748-0361.2003.tb00541.x. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data was publically available.