现在的位置: 首页 >时讯速递, 进展交流>正文
[JAMA发表论文]:美国内科人力资源中的非美国出生医生情况
2026年11月23日 ⁄ 时讯速递, 进展交流 ⁄ [JAMA发表论文]:美国内科人力资源中的非美国出生医生情况已关闭评论

Research Letter 

Non–US-Born Physicians in the US Internal Medicine Workforce

Giacomo Meille, Bradley M. Gray, Jonathan L. Vandergrift, et al

JAMA Published Online: August 12, 2026

doi: 10.1001/jama.2026.13268

Recent US immigration policies, including travel bans for 39 countries,1 social media vetting of applicants,2 and $100 000 H-1B visa fees for applicants residing outside the US,3 have disrupted resident applications and current physician workforce.4 Although the 39-country travel ban later exempted physicians, hospitals placed some noncitizen physicians from these countries on administrative leave,1 and the 2026 postgraduate year 1 match rate for non–US citizen international medical graduates (IMGs) reached a 5-year low.5

These immigration policies are concerning because the US health care system relies extensively on non–US-born physicians. Each year, residency positions outnumber US medical graduates (USMGs),6 and IMGs, especially noncitizens, address workforce shortages by specializing in primary care and practicing in underserved areas at higher rates than USMGs.7,8

Internal medicine is the largest physician specialty and one of the most reliant on IMGs.9 However, research on non–US-born physicians in internal medicine subspecialties is limited, and to our knowledge, no studies have examined the characteristics of patients treated by non–US-born physicians. We estimated the share of US-born and non–US-born USMGs and IMGs in the internal medicine workforce, including 11 subspecialties. We also examined the share of underserved patients cared for overall and during the first 3 years after completing training, the period covered by the Conrad 30 waiver program, which allows physicians with J-1 visas to remain in the US if they practice in underserved areas.

Methods

The study population included all physicians initially certified in internal medicine by the American Board of Internal Medicine (ABIM) during 1990-2025. A census of these physicians was examined using ABIM administrative data. Physicians without birth country or medical school data were excluded (2.0%).

To estimate the 2026 workforce, physicians with inactive licenses and those reported as retired or deceased were excluded (eAppendix in Supplement 1). The share of US-born and non–US-born USMGs and IMGs was calculated by specialty, year of certification, and birth country. USMGs included US-trained MDs and DOs.

The patient analysis linked ABIM data to 100% traditional Medicare claims from 2024 using National Provider Identifiers. For this analysis, we excluded physicians who were missing National Provider Identifier data (2.3%), who were in fellowship in 2024, or who were initially certified in 2024-2025 and Medicare beneficiaries who were younger than 65 years or lived outside the 50 US states and District of Columbia. We calculated the proportion of patients from underserved populations for the 4 groups of physicians (eAppendix in Supplement 1).

Analyses used Stata/MP version 19 (StataCorp). The study was deemed exempt from consent by the Advarra Institutional Review Board. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline was followed.

Results

In 2026, the total US workforce included 249 679 board-certified internal medicine physicians. The 2024 Medicare analysis included 178 782 internal medicine physicians.

Non–US-born IMGs made up 36.1% and non–US-born USMGs made up 11.1% (9.5% MDs and 1.6% DOs) of the 2026 internal medicine workforce (Table 1). Among internal medicine–certified physicians, non–US-born IMGs made up at least 27% of each subspecialty and more than 50% of internal medicine–certified geriatricians, sleep medicine physicians, and nephrologists. India and Pakistan contributed the most non–US-born IMGs—10.8% and 3.8% of the internal medicine workforce, respectively. Physicians born in the 39 travel ban countries comprised 6.0% of the workforce. The proportion of non–US-born physicians by initial certification year grew from 34.9% in 1990-1994 to 54.2% in 1995-1999 and decreased after 2015, reaching 44.5% in 2020-2025.

Table 1.  Certified Internal Medicine Physicians by Birth Country and Training, 2026 US Workforce

CategoryDenominator, No., in thousandsNo., in thousands (%)
US-born USMGaUS-born IMGNon–US-born USMGaNon–US-born IMG
Total249.7117.0 (46.9)14.9 (6.0)27.8 (11.1)90.0 (36.1)
Birth regionb249.7
AfricaNANA2.1 (0.8)8.3 (3.3)
Americas, excluding USNANA4.3 (1.7)11.2 (4.5)
AsiaNANA17.4 (7.0)62.0 (24.9)
EuropeNANA3.8 (1.5)8.4 (3.3)
OceaniaNANA0.1 (0.1)0.1 (0.1)
Selected country groupsb249.7
Restricted immigration countriescNANA3.0 (1.2)11.8 (4.7)
Spanish-speaking countriesNANA1.8 (0.7)7.1 (2.9)
Birth countryb,d249.7
IndiaNANA4.6 (1.8)27.0 (10.8)
PakistanNANA0.8 (0.3)9.4 (3.8)
PhilippinesNANA0.6 (0.2)3.1 (1.2)
NigeriaNANA0.5 (0.2)2.6 (1.0)
SyriaNANA0.1 (<0.1)2.5 (1.0)
ChinaNANA1.8 (0.7)2.5 (1.0)
IranNANA1.1 (0.4)2.3 (0.9)
CanadaNANA1.3 (0.5)2.2 (0.9)
EgyptNANA0.3 (0.1)1.9 (0.8)
LebanonNANA0.2 (0.1)1.8 (0.7)
All other countries, excluding USNANA16.4 (6.6)34.7 (13.9)
Specialtye
General internal medicine, not subspecialized
Outpatient focusf44.122.1 (50.2)2.6 (5.8)4.9 (11.1)14.5 (32.8)
Inpatient focusf39.215.6 (39.7)3.1 (8.0)4.2 (10.8)16.3 (41.6)
Subspecialties
Cardiology29.214.1 (48.4)1.2 (4.2)3.9 (13.4)9.9 (34.1)
Critical care/pulmonary medicine20.09.1 (45.5)1.5 (7.5)1.8 (8.8)7.6 (38.2)
Hematology/oncology17.58.2 (47.0)0.7 (4.0)2.0 (11.5)6.6 (37.6)
Gastroenterology16.58.8 (53.8)0.7 (4.4)2.4 (14.4)4.5 (27.4)
Nephrology11.83.8 (32.2)0.7 (5.6)1.2 (10.2)6.2 (52.0)
Infectious disease10.15.0 (49.1)0.6 (5.8)0.9 (9.2)3.6 (35.8)
Endocrinology8.33.2 (38.9)0.4 (5.4)0.9 (10.4)3.8 (45.4)
Rheumatology6.33.0 (46.9)0.3 (5.4)0.7 (11.2)2.3 (36.5)
Geriatrics5.72.0 (34.9)0.3 (5.4)0.4 (7.1)3.0 (52.7)
Hospice and palliative medicine5.02.8 (56.0)0.3 (6.2)0.5 (9.0)1.4 (28.8)
Sleep medicine3.81.3 (34.0)0.2 (5.2)0.3 (8.8)2.0 (52.0)
Initial internal medicine certification yearg
1990-199420.412.4 (60.7)0.9 (4.4)1.7 (8.6)5.4 (26.3)
1995-199932.513.9 (42.7)1.0 (3.1)3.6 (11.1)14.0 (43.1)
2000-200432.215.2 (47.4)1.3 (4.0)4.2 (13.0)11.5 (35.6)
2005-200933.414.5 (43.4)1.7 (5.0)3.6 (10.7)13.7 (40.9)
2010-201434.615.1 (43.5)2.1 (6.1)3.6 (10.3)13.9 (40.1)
2015-201939.118.1 (46.5)3.2 (8.1)4.6 (11.8)13.1 (33.6)
2020-202557.527.8 (48.3)4.7 (8.2)6.5 (11.4)18.5 (32.2)

In 2024, non–US-born USMGs and non–US-born IMGs treated approximately 4.8 million and 17.7 million Medicare beneficiaries, respectively. The share of patients from underserved populations was larger among non–US-born physicians (especially IMGs) than US-born USMGs (Table 2). Patients treated by non–US-born IMGs were more likely to be Black (9.5% vs 7.0%), Latino (6.6% vs 3.9%), originally Medicare eligible based on disability (13.8% vs 11.1%), and dual eligible for Medicare and Medicaid (20.9% vs 11.7%). Patients treated by non–US-born IMGs were also more likely to live in high-poverty zip codes (19.8% vs 15.9%), areas designated as medically underserved (21.5% vs 19.9%), and primary care shortage areas (30.3% vs 29.2%) compared with patients served by US-born USMGs. Non–US-born IMGs treated an especially high proportion of patients who were Medicare beneficiaries living in underserved locations during the first 3 years after completing training.

Table 2.  Medicare Beneficiaries From Underserved Populations and Residing in Underserved Areas by Physician Type, 2024 (N = 178 782)

No. of claims, in millions (% of physicians’ claims)a
All physicians1-3 y after completing trainingb
US-born USMG (n = 45.0)cUS-born IMG (n = 6.5)Non–US-born USMG (n = 11.6)cNon–US-born IMG (n = 46.1)US-born USMG (n = 4.3)cUS-born IMG (n = 0.9)Non–US-born USMG (n = 1.1)cNon–US-born IMG (n = 3.7)
Beneficiary characteristics
Non-Latino Black3.1 (7.0)0.5 (8.2)0.9 (8.0)4.4 (9.5)0.4 (8.4)0.1 (8.7)0.1 (8.8)0.4 (9.5)
Latino1.7 (3.9)0.4 (6.7)0.7 (6.3)3.0 (6.6)0.2 (4.8)0.1 (6.5)0.1 (6.1)0.2 (5.3)
Eligible through disabilityd5.0 (11.1)0.9 (13.3)1.3 (11.3)6.4 (13.8)0.5 (12.8)0.1 (14.0)0.1 (13.1)0.6 (15.2)
Dual eligiblee5.3 (11.7)1.3 (19.5)2.3 (19.7)9.7 (20.9)0.7 (16.2)0.2 (20.2)0.2 (18.7)0.7 (19.7)
Any underserved population11.4 (25.4)2.2 (34.2)3.9 (33.7)16.9 (36.7)1.3 (30.8)0.3 (35.2)0.4 (33.8)1.3 (35.7)
Beneficiary location
Nonmetropolitan countyf6.0 (13.4)0.6 (9.8)0.9 (7.7)5.9 (12.8)0.5 (12.8)0.1 (10.1)0.1 (9.2)0.7 (19.1)
High-poverty zip codeg7.2 (15.9)1.1 (16.8)1.8 (15.2)9.1 (19.8)0.7 (17.1)0.2 (16.8)0.2 (16.1)0.8 (21.9)
MUAPh9.0 (19.9)1.2 (18.6)1.7 (14.3)9.9 (21.5)0.8 (17.8)0.2 (20.4)0.2 (16.1)1.0 (26.2)
Primary care HPSAh13.2 (29.2)1.6 (24.3)2.3 (19.5)14.0 (30.3)1.2 (27.6)0.2 (26.0)0.2 (23.1)1.5 (39.0)
Any underserved location19.5 (43.2)2.6 (39.7)4.0 (34.7)21.6 (46.9)1.8 (42.4)0.4 (42.3)0.4 (38.6)2.1 (55.4)

Discussion

This study found that non–US-born physicians accounted for nearly half of the US internal medicine workforce and an even larger fraction in subspecialties such as geriatrics and nephrology. In 2024, non–US-born physicians treated 22.5 million traditional Medicare beneficiaries, and compared with US-born USMGs, a higher percentage of their patients were from low-income, disabled, Medicaid-eligible, and Black and Latino populations.

This study has limitations. It included physicians with initial ABIM certification; these represent 95% to 97% of US internists.10 Although citizenship data were not available, most IMGs are not US citizens when they begin residency.5 The patient analysis included only traditional Medicare beneficiaries; their demographics may differ from other Medicare populations.

These results suggest that recent changes in immigration policy may substantially affect internal medicine subspecialties that depend extensively on non–US-born IMGs. These policies may also further increase disparities in health care access among underserved populations.

抱歉!评论已关闭.

×
腾讯微博