Abstract
Purpose
This study describes graduate medical education (GME) placement outcomes for recent U.S. medical school graduates and examines racial and ethnic differences in GME placement among these graduates.
Method
This retrospective, observational study used data collected from and about U.S. medical school graduates for academic years 2015–2016 through 2021–2022. An individual-level, deidentified database was constructed to examine GME placement at graduation in association with race and ethnicity, as well as other demographic and academic and professional development variables. Multilevel (nested by school) logistic regression models identified variables independently associated with GME placement at graduation, reporting unadjusted odds ratios (UORs) and adjusted odds ratios (AORs) with 95% CIs.
Results
The study sample included 140,072 of 140,073 eligible graduates (> 99.9%; 1 graduate missing gender information was excluded), of whom 136,022 (97.1%) were placed in GME at graduation. Proportions of graduates placed in GME varied by race and ethnicity and by each covariable examined. In addition, proportions of graduates placed in GME varied by school (N = 152; mean [SD], 96.9% [3.4%]; P < .001). In multilevel (nested by school) models, GME placement UORs were lower for (among other groups examined) Asian (UOR, 0.76; 95% CI, 0.70–0.83), Black or African American (UOR, 0.44; 95% CI, 0.39–0.49), and Hispanic (UOR, 0.70; 95% CI, 0.60–0.80) graduates (vs White). The GME placement AORs, adjusted for all covariables, were similar for Asian (AOR, 0.96; 95% CI, 0.87–1.07), Black or African American (AOR, 0.89; 95% CI, 0.77–1.02), and Hispanic (AOR, 1.06; 95% CI, 0.89–1.25) graduates (vs White).
Conclusions
The proportion of graduates placed in GME at graduation during the 7 years of the study was high. However, there were racial and ethnic differences in this outcome during the study period.
The United States needs a larger physician workforce to address projected physician shortages for an increasing and aging population1 as well as a more racially and ethnically diverse physician workforce to better meet health care needs of an increasingly diverse population and reduce long-standing health inequities.2–4 U.S. medical schools have taken steps to address these needs by expanding class sizes, creating new schools,5 and matriculating increasingly diverse classes,6 particularly of racial and ethnic groups historically underrepresented in medicine (URiM), including Black or African American, Hispanic, American Indian or Alaska Native, and Native Hawaiian or Other Pacific Islander.
For medical school graduates, graduate medical education (GME) placement is a critical next step toward entry into the practicing physician workforce. Historically, GME placement rates at graduation have been high. According to results of a study of MD-granting schools’ 2004–2005 through 2014–2015 graduates, 97.0% of graduates were placed in GME at graduation.7 However, there were racial and ethnic differences in GME placement in this study. Graduates placed in GME at graduation comprised a less racially and ethnically diverse group than all graduates generally: those placed in GME at graduation included 97.4% of Asian, 94.3% of Black, 94.4% of Hispanic, and 98.1% of White graduates.7
Since this study was conducted, medical school graduate numbers have increased annually amid concerns about GME position availability for graduates.5 In 2020, implementation of a single accreditation system for MD and doctor of osteopathy (DO) GME programs was completed, increasing both numbers of applicants to and programs with positions offered in the National Resident Matching Program (NRMP) Main Residency Match,8 the process by which most students at U.S. MD-granting schools who are seeking to enter GME secure positions to do so. Coincident with these changes, an increasing body of work has continued to articulate the need for greater health care workforce diversity,4,9,10 and, in 2019, the Accreditation Council for Graduate Medical Education (ACGME) introduced Common Program Requirements addressing recruitment and retention of a diverse and inclusive workforce.11 Finally, recent COVID-19 pandemic and social movements to address systemic racism have placed a focus on the role of racial and ethnic diversity within medical education and the broader health care workforce.12 In the context of this evolving medical education landscape, we examined recent U.S. MD-granting school graduate cohorts to describe proportions of graduates placed in GME at graduation and to determine whether racial and ethnic differences in these proportions, which affect the diversity of the GME workforce, persisted.
Method
In this retrospective, observational study, we constructed a national database of individual-level, deidentified data collected from and about 2015–2016 through 2021–2022 U.S. MD-granting school graduates. Along with the main predictor variable, race and ethnicity, we included other demographic as well as academic and professional development covariables, with selection informed by others’ findings regarding variables associated with U.S. MD graduates’ entry into GME,7,13,14 program directors’ reports of variables used in resident selection,15,16 and reports from and about matching programs.8,17–19
On the basis of Association of American Medical Colleges (AAMC) information, we created a 6-category race and ethnicity variable (see Supplemental Digital Appendix 1 at http://links.lww.com/ACADMED/B630), including Asian, Black or African American, Hispanic, other URiM, other (non-URiM) or unknown, and White, placing each graduate in our study in 1 of these categories. Information from the AAMC Student Records System (SRS)20 was used for graduation year, gender, U.S. citizen or permanent resident status, degree program at graduation, and medical school leave of absence (LOA) covariables.
We created a 12-category specialty covariable using AAMC Graduation Questionnaire (GQ)21 and Electronic Residency Application Service (ERAS)22 data (see Supplemental Digital Appendix 2 at http://links.lww.com/ACADMED/B630) and used ERAS data for Gold Humanism Honor Society (GHHS) and Alpha Omega Alpha (AOA) Honor Society membership covariables. The National Board of Medical Examiners released graduates’ United States Medical Licensing Examination (USMLE) Step l and Step 2 Clinical Knowledge (CK) results. We created a 6-category covariable for the Step 1 and Step 2 CK results individually, separating graduates without a first-attempt pass into a category (fail or missing) and dividing graduates with a first-attempt pass into quintiles based on within-graduation-year score distribution (first through fifth quintile pass scores). Using the AAMC Organizational Characteristics Database23 information, we created covariables for research-intensive (National Institutes of Health top 40 ranked medical school, 2018 ranking based on federal research expenditures) and for community-based (medical school that does not have an integrated teaching hospital, received full accreditation in 1972 or later, and is nonfederal) medical school graduation. Using SRS data (reported by medical school registrars at graduation), we created a dichotomous GME placement at graduation (in an ACGME-accredited GME position) outcome and tabulated postgraduation plans for graduates not placed in GME.
We merged deidentified records into a single file, using χ2 tests of association, correlation coefficients, and tests for equality between 2 proportions for bivariate analyses and used random intercept multilevel (nested by school) logistic regression models. Models among all graduates and among those graduates with ERAS records in their final medical school year (ERAS-applicant graduates) identified variables independently associated with GME placement. We examined models for race and ethnicity alone and adjusted for all covariables, reporting unadjusted odds ratios (UORs) and adjusted odds ratios (AORs) and 95% CIs. A 2-sided P < .05 indicates statistical significance. The AAMC Human Subjects Protection Program staff reviewed this study and determined it exempt from further institutional board review. Data analyses were performed using Stata software, version 18 (StataCorp, College Station, Texas).
Results
The study sample included 140,072 of all 140,073 medical school graduates from 2015–2016 through 2021–2022 (> 99.9%; 1 graduate missing gender information was excluded), including 136,022 (97.1%) placed in GME at graduation. As indicated in Table 1, proportions of graduates placed in GME varied by race and ethnicity and by each covariable examined. In addition (not shown), proportions of graduates placed in GME varied by school (N = 152; mean [SD], 96.9% [3.4%]; P < .001).
Table 1.
Study Sample Characteristics of All Graduates Grouped by GME Placement at Graduation, 2015–2016 Through 2021–2022
| Characteristic | GME placement at graduation | Total (N = 140,072), no. (%)a | P value | |
|---|---|---|---|---|
| Not placed (n = 4,050), no. (%)a |
Placed (n = 136,022), no. (%)a |
|||
| Demographic | ||||
| Race and ethnicityb | < .001 | |||
| Asian | 875 (21.6) | 30,679 (22.6) | 31,554 (22.5) | |
| Black or African American | 527 (13.0) | 8,489 (6.2) | 9,016 (6.4) | |
| Hispanic | 471 (11.6) | 7,356 (5.4) | 7,827 (5.6) | |
| Other (non-URiM) or unknown | 281 (6.9) | 7,423 (5.5) | 7,704 (5.5) | |
| Other URiM | 303 (7.5) | 7,313 (5.4) | 7,616 (5.4) | |
| White | 1,593 (39.3) | 74,762 (55.0) | 76,355 (54.5) | |
| Gender | < .001 | |||
| Men | 2,772 (68.4) | 69,104 (50.8) | 71,876 (51.3) | |
| Women | 1,278 (31.6) | 66,918 (49.2) | 68,196 (48.7) | |
| U.S. citizen or permanent resident | < .001 | |||
| No | 193 (4.8) | 1,860 (1.4) | 2,053 (1.5) | |
| Yes | 3,857 (95.2) | 134,162 (98.6) | 138,019 (98.5) | |
| Academic and professional development | ||||
| Degree program at graduation | < .001 | |||
| MDc | 3,317 (81.9) | 126,954 (93.3) | 130,271 (93.0) | |
| MD and PhD | 195 (4.8) | 4,130 (3.0) | 4,325 (3.1) | |
| MD and other advanced degreed | 538 (13.3) | 4,938 (3.6) | 5,476 (3.9) | |
| Leave of absencee | < .001 | |||
| No | 3,098 (76.5) | 129,047 (94.9) | 132,145 (94.3) | |
| Yes | 952 (23.5) | 6,975 (5.1) | 7,927 (5.7) | |
| Gold Humanism Honor Society memberf | < .001 | |||
| Nonmember, not available at the school, or unknown | 3,874 (95.7) | 118,313 (87.0) | 122,187 (87.2) | |
| Yes | 176 (4.3) | 17,709 (13.0) | 17,885 (12.8) | |
| Alpha Omega Alpha memberf | < .001 | |||
| Nonmember, not available at the school, or unknown | 3,914 (96.7) | 114,175 (83.9) | 118,089 (84.3) | |
| Yes | 136 (3.4) | 21,847 (16.1) | 21,983 (15.7) | |
| Intended specialty categoryg | < .001 | |||
| Anesthesiology | 97 (2.4) | 8,234 (6.1) | 8,331 (5.9) | |
| Emergency medicine | 97 (2.4) | 12,002 (8.8) | 12,099 (8.6) | |
| Family medicine | 244 (6.0) | 11,063 (8.1) | 11,307 (8.1) | |
| Internal medicine | 303 (7.5) | 27,495 (20.2) | 27,798 (19.8) | |
| Obstetrics and gynecology | 94 (2.3) | 8,073 (5.9) | 8,167 (5.8) | |
| Pediatrics | 105 (2.6) | 12,660 (9.3) | 12,765 (9.1) | |
| Psychiatry | 181 (4.5) | 7,692 (5.7) | 7,873 (5.6) | |
| Radiology or radiation oncology | 76 (1.9) | 6,588 (4.8) | 6,664 (4.8) | |
| Surgery–general | 164 (4.1) | 8,030 (5.9) | 8,194 (5.9) | |
| Other surgical specialties | 450 (11.1) | 13,722 (10.1) | 14,172 (10.1) | |
| All other specialties | 1,048 (25.9) | 19,424 (14.3) | 20,472 (14.6) | |
| No specialty indicator | 1,191 (29.4) | 1,039 (0.8) | 2,230 (1.6) | |
| USMLE Step 1 first attempt | < .001 | |||
| First quintile pass | 1,458 (36.0) | 26,588 (19.6) | 28,046 (20.0) | |
| Second quintile pass | 605 (14.9) | 26,751 (19.7) | 27,356 (19.5) | |
| Third quintile pass | 472 (11.7) | 26,502 (19.5) | 26,974 (19.3) | |
| Fourth quintile pass | 431 (10.6) | 27,262 (20.0) | 27,693 (19.8) | |
| Fifth quintile pass | 345 (8.5) | 25,643 (18.9) | 25,988 (18.6) | |
| Fail or missingh | 739 (18.3) | 3,276 (2.4) | 4,015 (2.9) | |
| USMLE Step 2 CK first attempt | < .001 | |||
| First quintile pass | 1,452 (35.9) | 26,878 (19.8) | 28,330 (20.2) | |
| Second quintile pass | 615 (15.2) | 27,031 (19.9) | 27,646 (19.7) | |
| Third quintile pass | 438 (10.8) | 28,168 (20.7) | 28,606 (20.4) | |
| Fourth quintile pass | 336 (8.3) | 26,246 (19.3) | 26,582 (19.0) | |
| Fifth quintile pass | 196 (4.8) | 25,333 (18.6) | 25,529 (18.2) | |
| Fail or missingi | 1,013 (25.0) | 2,366 (1.7) | 3,379 (2.4) | |
| Community-based medical school graduationj | < .001 | |||
| No | 3,380 (83.5) | 118,141 (86.9) | 121,521 (86.8) | |
| Yes | 670 (16.5) | 17,881 (13.1) | 18,551 (13.2) | |
| Research-intensive medical school graduationk | < .001 | |||
| No | 2,885 (71.2) | 92,920 (68.3) | 95,805 (68.4) | |
| Yes | 1,165 (28.8) | 43,102 (31.7) | 44,267 (31.6) | |
Abbreviations: CK, clinical knowledge; GME, graduate medical education; URiM, underrepresented in medicine; USMLE, United States Medical Licensing Examination.
aPercentages shown are column percentages within each variable category.
bThe 6-category variable for race and ethnicity includes Asian (non-Hispanic) alone; Black or African American (non-Hispanic) alone; Hispanic, Latino, or of Spanish origin (no race identified) alone (Hispanic); other URiM, which included American Indian or Alaska Native and Native Hawaiian or Other Pacific Islander, alone or in combination with any other race or ethnicity, Black or African American in combination with any other race or ethnicity, and Hispanic ethnicity in combination with any race; other (non-URiM) or unknown; and White (non-Hispanic) alone.
cCategory includes those who graduated with a BA and MD or BS and MD degree.
dIncludes, for example, MD and MBA, MD and DDS, and MD and MPH degrees.
eA leave of absence is defined as a student taking a leave of absence from medical school for any of the following reasons: academic remediation, financial reasons, health reasons, or other reasons. This does not include students taking leave from medical school for research or participation in a joint degree.
fNot all schools had Gold Humanism Honor Society or Alpha Omega Alpha chapters.
gGraduates placed in GME were not necessarily placed in positions in their intended specialties.
hIncludes 3,812 graduates with first-attempt fail scores and 203 graduates without scores because graduate did not take the examination or opted to have their score withheld from being shared for research purposes.
iIncludes 3,192 graduates with first-attempt fail scores and 187 graduates without scores because graduate did not take the examination or opted to have their score withheld from being shared for research purposes.
jA nonfederal medical school that does not have an integrated teaching hospital and received full accreditation in 1972 or later.
kNational Institutes of Health top 40 ranked medical school according to 2018 ranking based on federal research expenditures.
As indicated in Table 2, overall GME placement percentages (7 years combined) were lower for all other race and ethnicity groups compared with the White graduates group. Within-year percentages for Hispanic and other URiM graduates were lower in every year; for Black or African American graduates, lower in every year except 2021–2022; for Asian graduates, lower in every year except 2017–2018 and 2021–2022; and for graduates of other (non-URiM) or unknown race and ethnicity, lower in every year except 2021–2022. Within-group correlations of GME placement percentage with more recent year were significant among Black or African American graduates (r = 0.96; P = .001) and graduates of other (non-URiM) or unknown race and ethnicity (r = 0.78; P = .04).
Table 2.
Graduate Medical Education Placement by Graduation Year (N = 140,072), 2015–2016 Through 2021–2022
| Race and ethnicitya | No. placed/no. of graduates (% of graduates) by academic yearb | Total no. placed/no. of graduates (% of graduates) | Correlation between academic year and the percentage placed (r) | P value | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 2015– 2016 |
2016–2017 | 2017–2018 | 2018–2019 | 2019–2020 | 2020–2021 | 2021– 2022 |
||||
| Asian | 4,060/4,190 (96.9)c | 4,057/4,184 (97.0)c | 4,186/4,296 (97.4) | 4,371/4,484 (97.5)c | 4,558/4,674 (97.5)d | 4,698/4,841 (97.0)d | 4,749/4,885 (97.2) | 30,679/31,554 (97.2)c | 0.30 | .51 |
| Black or African American | 986/1,081 (91.2)c | 1,038/1,117 (92.9)c | 1,077/1,162 (92.7)c | 1,208/1,290 (93.6)c | 1,321/1,404 (94.1)c | 1,420/1,486 (95.6)c | 1,439/1,476 (97.5) | 8,489/9,016 (94.2)c | 0.96 | .001 |
| Hispanic | 868/953 (91.1)c | 916/975 (93.9)c | 999/1,064 (93.9)c | 1,017/1,082 (94.0)c | 1,153/1,214 (95.0)c | 1,238/1,301 (95.2)c | 1,165/1,238 (94.1)c | 7,356/7,827 (94.0)c | 0.73 | .06 |
| Other (non-URiM) or unknown | 844/879 (96.0)c | 946/986 (95.9)c | 993/1,030 (96.4)c | 1,003/1,047 (95.8)c | 1,108/1,150 (96.3)c | 1,228/1,270 (96.7)d | 1,301/1,342 (96.9) | 7,423/7,704 (96.4)c | 0.78 | .04 |
| Other URiM | 839/893 (94.0)c | 962/1,000 (96.2)c | 939/972 (96.6)d | 1,027/1,066 (96.3)c | 1,064/1,101 (96.6)c | 1,183/1,233 (95.9)c | 1,299/1,351 (96.2)c | 7,313/7,616 (96.0)c | 0.51 | .24 |
| White | 10,710/10,947 (97.8) | 10,788/11,000 (98.1) | 10,798/11,038 (97.8) | 10,780/10,966 (98.3) | 10,641/10,847 (98.1) | 10,545/10,795 (97.7) | 10,500/10,762 (97.6) | 74,762/76,355 (97.9) | −0.33 | .46 |
| Total | 18,307/18,943 (96.6) | 18,707/19,262 (97.1) | 18,992/19,562 (97.1) | 19,406/19,935 (97.3) | 19,845/20,390 (97.3) | 20,312/20,926 (97.1) | 20,453/21,054 (97.1) | 136,022/140,072 (97.1) | 0.56 | .19 |
Abbreviation: URiM, underrepresented in medicine.
aThe 6-category variable for race and ethnicity includes Asian (non-Hispanic) alone; Black or African American (non-Hispanic) alone; Hispanic, Latino, or of Spanish origin (no race identified) alone (Hispanic); other URiM, which included American Indian or Alaska Native and Native Hawaiian or Other Pacific Islander, alone or in combination with any other race or ethnicity, Black or African American in combination with any other race or ethnicity, and Hispanic ethnicity in combination with any race; other (non-URiM) or unknown; and White (non-Hispanic) alone.
bAcademic year was July 1 to June 30.
cP < .01 for 2-sample t test comparing the percentage for the indicated race and ethnicity group with the percentage for the White group.
dP < .05 for 2-sample t test comparing the percentage for the indicated race and ethnicity group with the percentage for the White group.
Table 3 gives the GME placement odds among all graduates. The GME placement UORs were lower for Asian (UOR, 0.76; 95% CI, 0.70–0.83), Black or African American (UOR, 0.44; 95% CI, 0.39–0.49), Hispanic (UOR, 0.70; 95% CI, 0.60–0.80), other (non-URiM) or unknown race and ethnicity (UOR, 0.57; 95% CI, 0.50–0.65), and other URiM (UOR, 0.62; 95% CI, 0.55–0.71) graduates (vs White). The GME placement AORs, adjusting for all covariables, were similar for Asian (AOR, 0.96; 95% CI, 0.87–1.07), Black or African American (AOR, 0.89; 95% CI, 0.77–1.02), Hispanic (AOR, 1.06; 95% CI, 0.89–1.25), and other URiM (AOR, 0.96; 95% CI, 0.82–1.12) graduates and lower for graduates of other (non-URiM) or unknown race and ethnicity (AOR, 0.81; 95% CI, 0.69–0.95) (vs White graduates).
Table 3.
Multilevel Logistic Regression Results for GME Placement at Graduation (N = 140,072), 2015–2016 Through 2021–2022
| Multilevel model with only race and ethnicitya,b | Placed (vs not placed) in GME at graduation | |
|---|---|---|
| Unadjusted OR (95% CI) |
P value | |
| Asian | 0.76 (0.70–0.83) | < .001 |
| Black or African American | 0.44 (0.39–0.49) | < .001 |
| Hispanic | 0.70 (0.60–0.80) | < .001 |
| Other (non-URiM) or unknown | 0.57 (0.50–0.65) | < .001 |
| Other URiM | 0.62 (0.55–0.71) | < .001 |
| White | 1.00 [reference] | NA |
| Placed (vs not placed) in GME at graduation | ||
|---|---|---|
| Full multilevel model with all covariatesc | Adjusted OR (95% CI) | P value |
| Demographic | ||
| Graduation year, per more recent year | 1.00 (0.98–1.02) | .98 |
| Race and ethnicityb | ||
| Asian | 0.96 (0.87–1.07) | .46 |
| Black or African American | 0.89 (0.77–1.02) | .10 |
| Hispanic | 1.06 (0.89–1.25) | .52 |
| Other (non-URiM) or unknown | 0.81 (0.69–0.95) | .01 |
| Other URiM | 0.96 (0.82–1.12) | .60 |
| White | 1.00 [reference] | NA |
| Gender | ||
| Men | 1.00 [reference] | NA |
| Women | 1.85 (1.71–2.01) | < .001 |
| US citizen or permanent resident | ||
| Yes | 1.00 [reference] | NA |
| No | 0.27 (0.22–0.33) | < .001 |
| Academic and professional development | ||
| Degree program at graduation | ||
| MDd | 1.00 [reference] | NA |
| MD and PhD | 0.79 (0.65–0.95) | .01 |
| MD and other advanced degreee | 0.36 (0.31–0.41) | < .001 |
| Leave of absencef | ||
| No | 1.00 [reference] | NA |
| Yes | 0.42 (0.38–0.47) | < .001 |
| Gold Humanism Honor Society memberg | ||
| Nonmember, not available at the school, or unknown | 1.00 [reference] | NA |
| Yes | 1.73 (1.47–2.04) | < .001 |
| Alpha Omega Alpha memberg | ||
| Nonmember, not available at the school, or unknown | 1.00 [reference] | NA |
| Yes | 2.22 (1.83–2.69) | < .001 |
| Intended specialty categoryh | ||
| Anesthesiology | 0.80 (0.63–1.02) | .07 |
| Emergency medicine | 1.05 (0.83–1.33) | .68 |
| Family medicine | 0.85 (0.71–1.02) | .08 |
| Internal medicine | 1.00 [reference] | NA |
| Obstetrics and gynecology | 0.61 (0.47–0.78) | < .001 |
| Pediatrics | 1.25 (0.99–1.58) | .06 |
| Psychiatry | 0.64 (0.53–0.78) | < .001 |
| Radiology or radiation oncology | 0.70 (0.54–0.91) | .007 |
| Surgery–general | 0.41 (0.33–0.50) | < .001 |
| Other surgical specialties | 0.18 (0.15–0.21) | < .001 |
| All other specialties | 0.25 (0.22–0.29) | < .001 |
| No specialty indicator | 0.01 (0.01–0.01) | < .001 |
| USMLE Step 1 first attempt | ||
| First quintile pass | 0.69 (0.61–0.79) | < .001 |
| Second quintile pass | 1.03 (0.90–1.19) | .66 |
| Third quintile pass | 1.00 [reference] | NA |
| Fourth quintile pass | 0.99 (0.85–1.15) | .89 |
| Fifth quintile pass | 0.93 (0.78–1.11) | .41 |
| Fail or missing | 0.31 (0.26–0.37) | < .001 |
| USMLE Step 2 CK first attempt | ||
| First quintile pass | 0.48 (0.42–0.55) | < .001 |
| Second quintile pass | 0.79 (0.69–0.91) | .001 |
| Third quintile pass | 1.00 [reference] | NA |
| Fourth quintile pass | 1.06 (0.91–1.24) | .43 |
| Fifth quintile pass | 1.52 (1.25–1.85) | < .001 |
| Fail or missing | 0.10 (0.08–0.11) | < .001 |
| Community-based medical school graduationi | ||
| Yes | 0.75 (0.56–1.00) | .046 |
| No | 1.00 [reference] | NA |
| Research-intensive medical school graduationj | ||
| Yes | 1.00 (0.78–1.28) | .99 |
| No | 1.00 [reference] | NA |
Abbreviations: CK, clinical knowledge; GME, graduate medical education; NA, not applicable; OR, odds ratio; URiM, underrepresented in medicine; USMLE, United States Medical Licensing Examination.
aModel χ2 = 250.89 (P < .001).
bThe 6-category variable for race and ethnicity includes Asian (non-Hispanic) alone; Black or African American (non-Hispanic) alone; Hispanic, Latino, or of Spanish origin (no race identified) alone (Hispanic); other URiM, which included American Indian or Alaska Native and Native Hawaiian or Other Pacific Islander, alone or in combination with any other race or ethnicity, Black or African American in combination with any other race or ethnicity, and Hispanic ethnicity in combination with any race; other (non-URiM) or unknown; and White (non-Hispanic) alone.
cModel χ2 = 8,351.09 (P < .001). A test for multicollinearity among independent variables showed that all variance inflation factors were substantially below the threshold (values < 10) to indicate that collinearity was a concern.
dCategory includes those who graduated with a BA and MD or BS and MD degree.
eIncludes, for example, MD and MBA, MD and DDS, and MD and MPH degrees.
fA leave of absence is defined as a student taking a leave of absence from medical school for any of the following reasons: academic remediation, financial reasons, health reasons, or other reasons. This does not include students taking leave from medical school for research or participation in a joint degree.
gNot all schools had Gold Humanism Honor Society or Alpha Omega Alpha chapters.
hGraduates placed in GME were not necessarily placed in positions in their intended specialties.
iA nonfederal medical school that does not have an integrated teaching hospital and received full accreditation in 1972 or later.
jNational Institutes of Health top 40 ranked medical school according to 2018 ranking based on federal research expenditures.
Among covariables, GME placement AORs were lower for non–U.S. citizen or permanent residents (vs US citizen or permanent residents); MD and PhD and MD and other advanced degree (vs MD) graduates; graduates with (vs without) LOA records; graduates intending specialties (vs internal medicine) of obstetrics and gynecology, psychiatry, radiology or radiation oncology, surgery-general, other surgical specialties, all other specialties, and no specialty indicator graduates; graduates with Step 1 (vs third quintile pass) fail or missing and first quintile pass scores; graduates with Step 2 CK (vs third quintile pass) fail or missing, first quintile pass, and second quintile pass scores; and community-based (vs all other) medical school graduates. The GME placement AORs were higher for women (vs men); GHHS and AOA members (vs nonmembers, missing, or not available); and graduates with Step 2 CK fifth quintile pass (vs third quintile pass) scores. The GME placement AORs were similar for each of more recent years’ graduates; graduates intending specialties (vs internal medicine) of anesthesiology, emergency medicine, family medicine, and pediatrics; graduates with Step 1 second quintile pass, fourth quintile pass, and fifth quintile pass (vs third quintile pass) scores; graduates with Step 2 CK fourth quintile pass (vs third quintile pass) scores; and research-intensive (vs all other) medical school graduates.
Table 4 gives GME placement odds among ERAS-applicant graduates only. Race and ethnicity findings were largely similar to findings among all graduates (Table 3), except that whereas the AOR for Black or African American (vs White) graduates was similar among all graduates, the AOR was lower (AOR, 0.79; 95% CI, 0.68–0.92) for Black or African American (vs White) graduates among ERAS-applicant graduates. Covariate findings were similar to those among all graduates except that among ERAS-applicant graduates, the AOR (similar for those intending each of pediatrics and anesthesiology [vs internal medicine] among all graduates) was higher for those intending pediatrics (AOR, 1.38; 95% CI, 1.07–1.79) and lower for those intending anesthesiology (AOR, 0.75; 95% CI, 0.59–0.97); the AOR (similar among all graduates) was higher (AOR, 1.53; 95% CI, 1.20–1.95) for research-intensive (vs all other) medical school graduates; and the AOR (lower for MD and PhD [vs MD] graduates among all graduates) was similar (AOR, 1.24; 95% CI, 0.95–1.62) among ERAS-applicant graduates.
Table 4.
Multilevel Logistic Regression Results for GME Placement at Graduation Among Graduates With Electronic Residency Application Service Records in Their Final Year of Medical School (N = 136,280a), 2015–2016 Through 2021–2022
| Multilevel model with only race and ethnicityb,c | Placed (vs not placed) in GME at graduation | |
|---|---|---|
| Unadjusted OR (95% CI) | P value | |
| Asian | 0.67 (0.60 to 0.75) | < .001 |
| Black or African American | 0.35 (0.30 to 0.40) | < .001 |
| Hispanic | 0.58 (0.49 to 0.68) | < .001 |
| Other (non-URiM) or unknown | 0.59 (0.50 to 0.70) | < .001 |
| Other URiM | 0.51 (0.44 to 0.60) | < .001 |
| White | 1.00 [reference] | |
| Placed (vs. not placed) in GME at graduation | ||
|---|---|---|
| Full multilevel model with all covariatesd | Adjusted OR (95% CI)e | P value |
| Demographic | ||
| Graduation year, per more recent year | 1.01 (0.99–1.03) | .27 |
| Race and ethnicityc | ||
| Asian | 0.90 (0.80–1.01) | .08 |
| Black or African American | 0.79 (0.68–0.92) | .003 |
| Hispanic | 0.93 (0.78–1.11) | .43 |
| Other (non-URiM) or unknown | 0.76 (0.64–0.92) | .004 |
| Other URiM | 0.85 (0.72–1.02) | .08 |
| White | 1.00 [reference] | NA |
| Gender | ||
| Men | 1.00 [reference] | NA |
| Women | 1.98 (1.81–2.18) | < .001 |
| U.S. citizen or permanent resident | ||
| Yes | 1.00 [reference] | NA |
| No | 0.25 (0.20–0.31) | < .001 |
| Academic and professional development | ||
| Degree program at graduation | ||
| MDe | 1.00 [reference] | NA |
| MD and PhD | 1.24 (0.95–1.62) | .11 |
| MD and other advanced degreef | 0.76 (0.60–0.95) | .02 |
| Leave of absenceg | ||
| No | 1.00 [reference] | NA |
| Yes | 0.44 (0.39–0.50) | < .001 |
| Gold Humanism Honor Society memberh | ||
| No, not available at the school, or unknown | 1.00 [reference] | NA |
| Yes | 1.43 (1.21–1.68) | < .001 |
| Alpha Omega Alpha memberh | ||
| No, not available at the school, or unknown | 1.00 [reference] | NA |
| Yes | 2.00 (1.64–2.44) | < .001 |
| Intended specialty categoryi | ||
| Anesthesiology | 0.75 (0.59–0.97) | .02 |
| Emergency medicine | 1.04 (0.81–1.34) | .75 |
| Family medicine | 0.93 (0.76–1.13) | .48 |
| Internal medicine | 1.00 [reference] | NA |
| Obstetrics and gynecology | 0.57 (0.44–0.74) | < .001 |
| Pediatrics | 1.38 (1.07–1.79) | .01 |
| Psychiatry | 0.66 (0.54–0.82) | < .001 |
| Radiology or radiation oncology | 0.68 (0.51–0.89) | .006 |
| Surgery–general | 0.59 (0.46–0.76) | < .001 |
| Other surgical specialties | 0.17 (0.15–0.21) | < .001 |
| All other specialties | 0.23 (0.19–0.26) | < .001 |
| No specialty indicator | 0.05 (0.03–0.06) | < .001 |
| USMLE Step 1 first attempt | ||
| First quintile pass | 0.72 (0.61–0.84) | < .001 |
| Second quintile pass | 1.13 (0.96–1.33) | .14 |
| Third quintile pass | 1.00 [reference] | NA |
| Fourth quintile pass | 0.95 (0.81–1.13) | .57 |
| Fifth quintile pass | 0.91 (0.75–1.11) | .35 |
| Fail or missing | 0.31 (0.26–0.38) | < .001 |
| USMLE Step 2 CK first attempt | ||
| First quintile pass | 0.43 (0.37–0.51) | < .001 |
| Second quintile pass | 0.78 (0.66–0.92) | .003 |
| Third quintile pass | 1.00 [reference] | NA |
| Fourth quintile pass | 1.01 (0.84–1.21) | .93 |
| Fifth quintile pass | 1.43 (1.15–1.80) | .002 |
| Fail or missing | 0.09 (0.07–0.10) | < .001 |
| Community-based medical school graduationj | ||
| Yes | 0.72 (0.55–0.95) | .02 |
| No | 1.00 [reference] | NA |
| Research-intensive medical school graduationk | ||
| Yes | 1.53 (1.20–1.95) | .001 |
| No | 1.00 [reference] | NA |
Abbreviations: CK, clinical knowledge; GME, graduate medical education; NA, not applicable; OR, odds ratio; URiM, underrepresented in medicine; USMLE, United States Medical Licensing Examination.
aOf these 136,280 ERAS-applicant graduates, 133,653 (98.1%) were placed in GME in graduation.
bModel χ2 = 266.02 (P < .001).
cThe 6-category variable for race and ethnicity includes Asian (non-Hispanic) alone; Black or African American (non-Hispanic) alone; Hispanic, Latino, or of Spanish origin (no race identified) alone (Hispanic); other URiM, which included American Indian or Alaska Native and Native Hawaiian or Other Pacific Islander, alone or in combination with any other race or ethnicity, Black or African American in combination with any other race or ethnicity, and Hispanic ethnicity in combination with any race; other (non-URiM) or unknown; and White (non-Hispanic) alone.
dModel χ2 = 4,617.22 (P < .001). A test for multicollinearity among independent variables showed that all variance inflation factors were substantially below the threshold (values < 10) to indicate that collinearity was a concern.
eCategory includes those who graduated with a BA and MD or BS and MD degree.
fIncludes, for example, MD and MBA, MD and DDS, and MD and MPH degrees.
gA leave of absence is defined as a student taking a leave of absence from medical school for any of the following reasons: academic remediation, financial reasons, health reasons, or other reasons. This does not include students taking leave from medical school for research or participation in a joint degree.
hNot all schools had Gold Humanism Honor Society or Alpha Omega Alpha chapters.
iGraduates placed in GME were not necessarily placed in positions in their intended specialties.
jA nonfederal medical school that does not have an integrated teaching hospital and received full accreditation in 1972 or later.
kNational Institutes of Health top 40 ranked medical school according to 2018 ranking based on federal research expenditures.
Table 5 summarizes the postgraduation plans of all 4,050 graduates not placed in GME. Plans varied by race and ethnicity.
Table 5.
Postgraduation Plans for Graduates Not Placed in Graduate Medical Education at Graduation (N = 4,050), 2015–2016 Through 2021–2022
| Race and ethnicitya,b | Medical activity, no. (%)c,d | Nonmedical activity or year off, no. (%)d,e | Oral (maxillofacial) surgery, no. (%)d | Research, no. (%)d | Undecided, unknown, or missing, no. (%)d,f | Total, no. (%)d |
|---|---|---|---|---|---|---|
| Asian | 121 (21.7) | 123 (27.4) | 95 (20.2) | 137 (22.4) | 399 (20.4) | 875 (21.6) |
| Black or African American | 57 (10.2) | 45 (10.0) | 9 (1.9) | 70 (11.4) | 346 (17.7) | 527 (13.0) |
| Hispanic | 68 (12.2) | 36 (8.0) | 7 (1.5) | 55 (9.0) | 305 (15.6) | 471 (11.6) |
| Other (non-URiM) or unknown | 41 (7.3) | 32 (7.1) | 66 (14.0) | 31 (5.1) | 111 (5.7) | 281 (6.9) |
| Other URiM | 35 (6.3) | 26 (5.8) | 16 (3.4) | 41 (6.7) | 185 (9.4) | 303 (7.5) |
| White | 236 (42.3) | 187 (41.6) | 278 (59.0) | 278 (45.4) | 614 (31.3) | 1,593 (39.3) |
| Total | 558 | 449 | 471 | 612 | 1,960 | 4,050 |
Abbreviations: URiM, underrepresented in medicine.
aThe 6-category variable for race and ethnicity includes Asian (non-Hispanic) alone; Black or African American (non-Hispanic) alone; Hispanic, Latino, or of Spanish origin (no race identified) alone (Hispanic); other URiM, which included American Indian or Alaska Native and Native Hawaiian or Other Pacific Islander, alone or in combination with any other race or ethnicity, Black or African American in combination with any other race or ethnicity, and Hispanic ethnicity in combination with any race; other (non-URiM) or unknown; and White (non-Hispanic) alone.
bThe association between race and ethnicity and postgraduation plans was statistically significant (χ2 = 321.90; P < .001).
cMedical activity group included 528 medical activity and 30 faculty or administrative positions. Medical activity group includes (among others) graduates placed in graduate medical education in locations outside the United States (e.g., in Canada).
dPercentages shown are column percentages.
eNonmedical activity or year off group included 307 who planned to pursue a nonmedical activity and 142 who planned to take a year off.
fUndecided, unknown, or missing group included 1,941 whose plans were reported to the Association of American Medical Colleges in the Student Records System as undecided or unknown and 19 with missing data (i.e., registrar’s office did not provide a response for the Student Records System item pertaining to plan for the graduate).
Discussion
Despite concerns about GME position availability for increasing numbers of U.S. medical school graduates,5 our observed GME placement percentage (97.1%) was similar to that reported (97.0%) for earlier, smaller cohort years,7 and in our study, GME placement odds remained similar among more recent graduate cohorts. Our observations align with NRMP Main Residency Match data that indicate that in 2016–2022, despite steadily increasing numbers of U.S. MD graduates, there was a consistent excess of postgraduate year 1 positions compared with the number of active U.S. MD senior applicants.24 A recently published analysis of ACGME-accredited GME entry-year positions in pipeline specialty programs (i.e., programs that lead to initial specialty board certification) in 2010–2011 through 2022–2023 also documented a consistent excess of entry-year positions compared with annual numbers of U.S. MD and DO graduates.25 Because there remains an excess of all active applicants in the NRMP relative to postgraduate year 1 positions26 and increase of U.S. medical school enrollees continues,25,27 further expansion of ACGME-accredited GME positions will be needed to address projected physician shortages in the United States.1,25
In our study and others examining GME placement outcomes,7,13,14,28 graduates placed in GME comprise a heterogeneous group. The GME placement outcomes are related to, but not synonymous with, matching process outcomes. Although most U.S. graduates placed in GME at graduation obtain their positions by matching,19,24 others secure positions via the NRMP Supplemental Offer and Acceptance Program24 or other arrangements outside a matching program.14,29
Graduates not placed in GME at graduation also comprise a heterogeneous group, including some who attempt to secure a GME position to enter immediately after graduation but are unsuccessful30 and others who do not attempt to do so. On the basis of Liaison Committee for Medical Education Annual Questionnaire Part II data, 50% (1,508 of 3,038) of all potential graduates in 2016–2022 who were anticipated by their schools not to be entering residency training immediately after graduation were reported as having been unable to find a residency position (D.A., written communication, July 11, 2023). These previously unpublished Liaison Committee for Medical Education Annual Questionnaire Part II data (corresponding to our graduate cohorts) and information we report regarding plans of graduates not placed in GME at graduation provide national benchmarking data for U.S. medical schools.
Although 97.1% of all graduates in our study were placed in GME at graduation, this proportion differed by race and ethnicity. As have others for earlier, smaller cohorts,7 we observed lower proportions of graduates of all race and ethnicity groups other than White placed in GME. We did not examine matching process data so could not determine the extent to which, if any, racial and ethnic differences in matching process outcomes may have contributed to the differences we observed in GME placement. Current NRMP initiatives to examine applicant demographic characteristics and matching process outcomes31,32 will be informative in this regard. We also note that, unlike earlier trends toward increasing proportions of Black and Hispanic graduates not placed in GME at graduation,7 we observed a significant trend toward increasing proportions of Black or African American graduates and stable proportions of Hispanic graduates placed in GME at graduation. The ACGME Common Program Requirements addressing recruitment and retention of a diverse and inclusive workforce11 implemented in 2019 may be contributory, given current attention to recruiting a more diverse GME workforce as program directors develop strategies aligning application review approaches with institutional missions33,34 and holistic review frameworks.35
Our models’ covariables accounted for the lower GME placement UORs for most, although not all, graduate race and ethnicity groups examined (vs White). Of the 3 demographic covariables examined, gender and U.S. citizen or permanent resident status, but not graduation year, were independently associated with GME placement. Women physicians are reportedly at higher risk for career path disruptions.36 Our findings, along with a recent report of lower medical school attrition among women than among men,37 indicate that this is not so for women at earlier stages in their development as physicians. The lower GME placement odds we observed for non–U.S. citizen or permanent resident graduates may be, in part, due to additional steps many of these graduates must navigate regarding their Visa status to enter GME in the United States; non–U.S. citizen or permanent residents who do secure GME positions hold a range of visa types and other designations.38
Each academic and professional development covariable in our study was associated with GME placement. Intended specialty findings were consistent with specialty-specific differences in matching process outcomes for U.S. MD seniors.39 For example, among U.S. MD senior applicants in 2018 (an approximate midpoint in our study period) who ranked programs in the NRMP Main Residency Match in a single specialty, high proportions matched to positions in internal medicine (3,172 of 3,220 [98.5%]), pediatrics (1,679 of 1,693 [99.2%]), and family medicine (1,458 of 1,512 [96.4%]),39 and our observed GME placement odds for graduates intending family medicine or pediatrics specialties were similar, or higher than, GME placement odds for graduates intending internal medicine (reference group). In contrast, lower proportions matched to positions in obstetrics and gynecology (868 of 964 [90.0%]) and other surgical specialties (e.g., plastic surgery [114 of 127 (89.8%)], neurologic surgery [197 of 219 (90.0%)], orthopedic surgery [652 of 755 (86.4%)])39 and our observed GME placement odds for graduates intending obstetrics and gynecology or other surgical specialties were lower than odds for graduates intending internal medicine.
Among all MD and advanced degree program graduates in our study, our observations differed for MD and PhD and for MD and other advanced degree graduates. Because GME placement AORs for MD and PhD graduates were lower among all graduates but not among ERAS-applicant graduates, MD and PhD graduates who attempted to enter GME did not appear to be disadvantaged in doing so compared with MD graduates. In contrast, the lower GME placement AORs for MD and other advanced degree graduates among all graduates and among ERAS-applicant graduates are consistent with NRMP data during our study period that indicate that U.S. MD senior other (non-PhD) advanced degree holder applicants were more highly represented among those unmatched vs matched to their preferred specialty.17 Because most U.S. MD-granting schools offer dual MD and PhD degree and various other advanced degree programs,40 our observations may be useful to these many programs’ participants, their advisers, and program leaders.
Observed associations of Step 1 and Step 2 CK scores and AOA and GHHS membership with GME placement were expected, given extensive information about secondary use of these data by program directors in applicant selection during our study period.15,16,41 Among program directors surveyed in 2018 regarding factors considered in selecting applicants to interview, most considered Step 1 and Step 2 CK scores (94% and 80%, respectively), and many considered AOA and GHHS membership (60% and 47%, respectively),15 and a 2018 NRMP report documented higher Step 1 and Step 2 CK scores and higher proportions of AOA members among U.S. MD seniors matched than unmatched to their preferred specialty.17 Inclusion of these covariables (among others in our study) does not imply that they necessarily should continue to carry the same weight by program directors in applicant selection. Rather, we included them as likely contributors to GME placement and for future comparison as residency programs implement holistic selection processes,42 balancing USMLE performance and honors society membership with other applicant characteristics. Interpretations of our findings should consider these caveats and that Step 1 results are now reported as pass/fail only.43
The lower GME placement odds for graduates with LOA records align with program directors’ consideration of lack of gaps in medical education when selecting applicants to interview. In a 2018 survey, 53% of program directors reported doing so.15 There are many reasons for medical school LOAs; because the well-being of the health professions community at large has emerged as a national issue,44 care should be taken to ensure that applicants who took LOAs for personal well-being are not unduly penalized in the resident selection process.
Finally, our findings of institutional characteristics among ERAS-applicant graduates align with program directors’ reported use of medical school reputation in resident selection.15 Fifty percent of program directors surveyed in 2018 considered whether an applicant was a graduate of a highly regarded medical school in selecting applicants to interview,15 and in 2018, U.S. MD seniors graduating from top 40 National Institutes of Health–ranked medical schools were more highly represented among those matched vs unmatched to their preferred specialty.17
Our study has several strengths. Unlike studies examining GME placement outcomes among all graduates only7 or among ERAS-applicant graduates only,13,14,28 we examined outcomes among both groups, highlighting both similarities and differences. To our knowledge, numerous variables in our study (e.g., intended specialty, Step 1 and Step 2 CK scores, degree program) have not previously been examined together in association with GME placement, and national data regarding characteristics and plans of graduates not placed in GME at graduation have not been previously reported.
Our study also has numerous limitations. Small numbers of American Indian or Alaska Native and Native Hawaiian or Other Pacific Islander graduates precluded separate categorization. The Asian-alone graduates category included graduates in some subcategories that are underrepresented in medicine,45 and there may be subcategory GME placement differences within this category. Additional variables considered by program directors in resident selection15,16,41 and other characteristics and choices of graduates likely contributed to GME placement outcomes. Our study included U.S. MD-granting school graduates only. The GME placement findings may not generalize to U.S. DO-granting school graduates, who comprise 17.8% of the ACGME-accredited GME workforce.46 The DO-granting school matriculant numbers steadily increased during our study period as did the racial and ethnic diversity of DO-granting school matriculants.47 We were unable to include DO-granting school graduates due to lack of data availability for several variables in our study (e.g., AAMC SRS20 GME placement at graduation, AAMC GQ21 intended specialty). On the basis of the American Association of Colleges of Osteopathic Medicine Graduating Seniors Survey data,48 2% to 4% of recent respondents annually were undecided about postgraduation plans or definitely did not plan to enter GME after graduation. Our findings also may not generalize to international medical school graduates, who comprise 22.9% of the ACGME-accredited GME workforce46; their contributions to GME workforce diversity have been well described.49
Conclusions
Within the context of the above limitations, we conclude that GME placement rates have remained high among recent U.S. MD graduate cohorts. There were racial and ethnic differences in this outcome during the study period, with lower proportions of each of Black or African American, Asian, Hispanic, and other URiM graduates and graduates of other (non-URiM) or unknown race and ethnicity, placed in GME at graduation compared with the proportion of White graduates placed in GME at graduation. Because not all U.S. senior students necessarily attempt to secure GME positions for entry immediately after graduation, those not placed in GME at graduation comprise a heterogeneous group, including some who had attempted unsuccessfully to secure a GME position for entry immediately after graduation and others who had not attempted to do so.
Acknowledgments
The authors would like to thank the following Association of American Medical Colleges (AAMC) data stewards, all full-time employees of the AAMC, for their assistance with data and coding: David Matthew, PhD, Brianna Gunter, Andrew Nees, Tomas Massari, and Tyler Litsch. The authors would also like to thank Michael Jodoin, PhD, senior vice president, Customer and Portfolio Management, National Board of Medical Examiners, for his review and critique of a preliminary draft of the manuscript.
Funding/Support
None reported.
Other disclosures
None reported.
Ethical approval
The AAMC Human Subjects Protection Program staff reviewed this study and determined it exempt from further institutional board review.
Data
The data collected for this study cannot be made available to others by the study authors. The data used in the study are sensitive, proprietary data. The authors were granted access to the data for the purposes of the described study only. Interested researchers can submit a request for Association of American Medical Colleges’ data at https://www.aamc.org/request-aamc-data and for National Board of Medical Examiners’ data at https://www.nbme.org/services/data-sharing.
Footnotes
Supplemental digital content for this article is available at http://links.lww.com/ACADMED/B630.
Contributor Information
Douglas Grbic, Email: dgrbic@aamc.org.
Daniel P. Jurich, Email: djurich@nbme.org.
Alex J. Mechaber, Email: amechaber@nbme.org.
Lindsay Roskovensky, Email: lroskovensky@aamc.org.
Geoffrey H. Young, Email: gyoung@aamc.org.
References
- 1.GlobalData Plc . The Complexities of Physician Supply and Demand: Projections From 2019 to 2036. Association of American Medical Colleges; March 2024. https://www.aamc.org/media/75236/download?attachment. Accessed June 26, 2024. [Google Scholar]
- 2.Kington R, Tisnado D, Carlisle L. Increasing racial and ethnic diversity among physicians: an intervention to address health disparities? In: Smedley BD, Stith AY, Colburn L, Evans CH, eds. The Right Thing to Do, The Smart Thing to Do: Enhancing Diversity in the Health Professions: Summary of the Symposium on Diversity in Health Professions in Honor of Herbert W. Nickens, M.D. National Academies Press; 2001:57–90. [PubMed] [Google Scholar]
- 3.Capers Q. Diversifying the physician workforce—from rhetoric to positive action. N Engl J Med. 2023;388(10):865–867. [DOI] [PubMed] [Google Scholar]
- 4.National Academies of Sciences, Engineering, and Medicine . Ending Unequal Treatment: Strategies to Achieve Equitable Health Care and Optimal Health for All. National Academies Press; 2024. [PubMed] [Google Scholar]
- 5.Association of American Medical Colleges . Medical School Enrollment Survey: 2020 Results. Association of American Medical Colleges; 2021. https://www.aamc.org/media/9936/download. Accessed June 26, 2024. [Google Scholar]
- 6.Association of American Medical Colleges . Diversity increases at medical schools in 2022. https://www.aamc.org/news/press-releases/diversity-increases-medical-schools-2022. Published December 13, 2022. Accessed June 26, 2024.
- 7.Sondheimer HM, Xierali IM, Young GH, Nivet MA. Placement of US medical school graduates into graduate medical education, 2005 through 2015. JAMA. 2015;314(22):2409. [DOI] [PubMed] [Google Scholar]
- 8.National Resident Matching Program, Results and Data: 2020 Main Residency Match. National Resident Matching Program; 2020. [Google Scholar]
- 9.Swartz TH, Palermo AGS, Masur SK, Aberg JA. The science and value of diversity: Closing the gaps in our understanding of inclusion and diversity. J Infect Dis. 2019;220(suppl 2):S33–S41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Gomez LE, Bernet P. Diversity improves performance and outcomes. J Natl Med Assoc. 2019;111(4):383–392. [DOI] [PubMed] [Google Scholar]
- 11.McDade WA. Increasing graduate medical education diversity and inclusion. J Grad Med Educ. 2019;11(6):736–738. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Humphrey HJ, Levinson D, Nivet MA, Schoenbaum SC. Addressing harmful bias and eliminating discrimination in health professions learning environments: An urgent challenge. Acad Med. 2020;95(12 suppl):S1–S4. [DOI] [PubMed] [Google Scholar]
- 13.Nguyen M Chaudry SI Desai MM, et al. Rates of medical student placement into graduate medical education by sex, race and ethnicity, and socioeconomic status, 2018-2021. JAMA Netw Open. 2022;5(8):e2229243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Poon SC, Nellans K, Gorroochurn P, Chahine NO. Race, but not gender, is associated with admissions into orthopaedic surgery residency programs. Clin Orthop Relat Res. 2022;480(8):1441–1449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.National Resident Matching Program Data Release and Research Committee . Results of the 2018 NRMP Program Director Survey. National Resident Matching Program; 2018. [Google Scholar]
- 16.National Resident Matching Program Data Release and Research Committee . Results of the 2020 NRMP Program Director Survey. National Resident Matching Program; 2020. [Google Scholar]
- 17.National Resident Matching Program . Charting Outcomes in the Match: U.S. Allopathic Seniors, 2018. National Resident Matching Program; 2018. [Google Scholar]
- 18.National Resident Matching Program . Charting Outcomes in the Match: Senior Students of U.S. Medical Schools, 2022. National Resident Matching Program; 2022. [Google Scholar]
- 19.Jiang J, Diebert CM. Urology match trends: Examining the urology match from 2006 to 2021. J Urol. 2021;206(3 suppl):e425. doi: 10.1097/JU.0000000000002017.01. [DOI] [Google Scholar]
- 20.Association of American Medical Colleges . Data and Reports: Student Records System (SRS). https://www.aamc.org/data-reports/students-residents/report/student-records-system-srs. Accessed June 26, 2024.
- 21.Association of American Medical Colleges . Medical School Graduation Questionnaire (GQ). https://www.aamc.org/data-reports/students-residents/report/graduation-questionnaire-gq. Accessed June 26, 2024.
- 22.Association of American Medical Colleges . ERAS for Medical Schools. https://www.aamc.org/services/eras-for-institutions/medical-schools. Accessed March 21, 2024.
- 23.Association of American Medical Colleges . Organizational Characteristics Database (OCD). https://www.aamc.org/data-reports/faculty-institutions/report/organizational-characteristics-database-ocd. Accessed June 26, 2024.
- 24.National Resident Matching Program . Results and Data: 2022 Main Residency Match. National Resident Matching Program; 2022. [Google Scholar]
- 25.Nasca TJ. Are there sufficient ACGME-accredited GME positions leading to eligibility for initial board certification to absorb UME growth in the United States? J Grad Med Educ. 2024;16(1):100–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.National Resident Matching Program . Results and Data: 2023 Main Residency Match. National Resident Matching Program; 2023. [Google Scholar]
- 27.Association of American Medical Colleges . Diversity Increases at Medical Schools in 2022 [press release]. https://www.aamc.org/news/press-releases/diversity-increases-medical-schools-2022. Published December 13, 2022. Accessed June 26, 2024.
- 28.Bowe SN, Bly RA, Wang X, Whipple ME. Racial and ethnic differences in resident selection in 11 specialties, 2013-2018. JAMA. 2022;327(24):2450–2452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Association of American Medical Colleges . The FindAResident Search Tool. https://students-residents.aamc.org/findaresident/findaresident-search-tool. Accessed June 25, 2024.
- 30.Bumsted T, Schneider BN, Deiorio NM. Considerations for medical students and advisors after an unsuccessful match. Acad Med. 2017;92(7):918–922. [DOI] [PubMed] [Google Scholar]
- 31.Curtin LS, Goldstein RB, Lamb DL. Applicant Demographics and the Transition to Residency: It’s Time to Leverage Data on Preferred Specialty and Match Outcomes to Inform the National Conversation about Diversity and Equity in Medical Education. Research Brief, National Resident Matching Program. https://www.nrmp.org/wp-content/uploads/2023/02/Demographic-data-perspectives-paper_FINAL.pdf. Published February 2023. Accessed June 25, 2024.
- 32.National Resident Matching Program . Charting Outcomes: Demographic Characteristics of Applicants in the Main Residency Match and SOAP. https://www.nrmp.org/match-data-analytics/interactive-tools/charting-outcomes-demographic-characteristics-of-applicants-in-the-main-residency-match. Accessed June 27, 2024.
- 33.Swails JL, Adams S, Hormann M, Omoruyi E, Aibana O. Mission-based filters in the electronic residency application service: Saving time and promoting diversity. J Grad Med Educ. 2021;13(6):785–794. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Aibana O, Swails JL, Flores RJ, Love L. Bridging the gap: Holistic review to increase diversity in graduate medical education. Acad Med. 2019;94(8):1137–1141. [DOI] [PubMed] [Google Scholar]
- 35.Marbin J Rosenbluth G Brim R, et al. Improving diversity in pediatric residency selection: Using an equity framework to implement holistic review. J Grad Med Educ. 2021;13(2):195–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Frank E, Zhao Z, Sen S, Guille C. Gender disparities in work and parental status among early career physicians. JAMA Netw Open. 2019;2(8):e198340. doi: 10.1001/jamanetworkopen.2019.8340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Nguyen M Chaudhry SI Desai MM, et al. Association of mistreatment and discrimination with medical school attrition. JAMA Pediatr. 2022;176(9):935–937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Brotherton SE, Etzel SI. Graduate medical education, 2022-2023. JAMA. 2023;330(10):988–1011. [DOI] [PubMed] [Google Scholar]
- 39.National Resident Matching Program . Results and Data: 2018 Main Residency Match. National Resident Matching Program; 2018. [Google Scholar]
- 40.Association of American Medical Colleges . Students and residents: Considering a combined degree: MD-PhD, MD-MBA, MD-MPH, MD-JD. https://students-residents.aamc.org/applying-medical-school/considering-combined-degree-md-phd-md-mba-md-mph-md-jd. Accessed June 27, 2024.
- 41.National Resident Matching Program, Data Release and Research Committee . Results of the 2021 NRMP Program Director Survey. National Resident Matching Program; 2021. [Google Scholar]
- 42.National Resident Matching Program, Data Release and Research Committee . Results of the 2022 NRMP Program Director Survey. National Resident Matching Program; 2022. [Google Scholar]
- 43.Chaudhry HJ, Katsufrakis PJ, Tallia AF. The USMLE Step 1 decision: An opportunity for medical education and training. JAMA. 2020;323(20):2017. [DOI] [PubMed] [Google Scholar]
- 44.National Academy of Medicine . National Plan for Health Workforce Well-Being. National Academies Press; 2022. [Google Scholar]
- 45.Ko M, Ton H. The not underrepresented minorities: Asian Americans, diversity, and admissions. Acad Med. 2020;95(2):184–189. [DOI] [PubMed] [Google Scholar]
- 46.Accreditation Council for Graduate Medical Education . ACGME Data Resource Book: Academic Year 2022–2023: Table C.15. https://www.acgme.org/globalassets/pfassets/publicationsbooks/2022-2023_acgme_databook_document.pdf. Accessed June 27, 2024.
- 47.Ali A. First-Year Enrollment by Race/Ethnicity 1976–2023. American Association of Colleges of Osteopathic Medicine; 2023. https://www.aacom.org/searches/reports/report/1976-23-FYEbyRE. Accessed June 27, 2024. [Google Scholar]
- 48.American Association of Colleges of Osteopathic Medicine . 2022–2023 Academic Year, Graduating Seniors Survey, Summary Report. https://www.aacom.org/docs/default-source/research-reports/2022-2023-academic-year-graduating-seniors-survey-summary-reportc0819886-a691-48f3-af21-19fa309fe529.pdf. January 2024. Accessed June 27, 2024.
- 49.Norcini JJ, van Zanten M, Boulet JR. The contribution of international medical graduates to diversity in the U.S. physician workforce: Graduate medical education. J Health Care Poor Underserved. 2008;19(2):493–499. [DOI] [PubMed] [Google Scholar]