Residency Match Rates in Orthopaedic Surgery Based on Sex, Under-Represented in Medicine Status, and Degree Type

Introduction:

Orthopaedic surgery is well recognized as one of the most competitive and least diverse medical specialties. Despite efforts toward improving diversity, studies have shown that gender and racial/ethnic disparities continue to persist in orthopaedic graduate medical education. Therefore, we sought to identify the match rates of traditionally under-represented groups within orthopaedic surgery—female candidates, racial and ethnic minorities under-represented in medicine (URiM), and osteopathic physicians—compared with their application rates.

Methods:

A retrospective review of the Electronic Residency Application Service (ERAS) application data from the 2017 to 2021 application cycles was performed, and the total number of applicants, sex, race/ethnicity, and degree type of all students applying for orthopaedic surgery were recorded. A separate database, the Orthopaedic Residency Information Network (ORIN), which is a database self-reported by residency programs to provide information to applicants, was also queried to identify the total number of residents, sex, race/ethnicity, and degree type of all current residents in June 2022, corresponding to those residents who matched in 2017 to 2021.

Results:

From the ERAS application data, a total of 7,903 applicants applied to orthopaedic surgery during the study period. A total of 1,448 applicants (18%) were female, 1,307 (18%) were URiM, and 1,022 (15%) were from an osteopathic medical school. Based on the ORIN database, 688 of 3,574 residents (19%) were female, 1,131 of 7,374 (19%) were URiM, and 1,022 of US medical school graduates (12%) had a DO degree. The application and match rates were not significantly different for female (p = 0.249) and URiM (0.187) applicants; however, there was a significant difference in the application and match rates (15% vs 12%; p = 0.035) for US medical graduates with a DO degree.

Conclusion:

In recent years, there has been a significant and necessary push to increase diversity in the field of orthopaedic surgery. From 2017 to 2021, match rates of female and URiM candidates are reflective of their application rates. Osteopathic applicants in orthopaedic surgery have a lower match rate than their allopathic counterparts.

Level of Evidence:

III

Introduction

Orthopaedic surgery is well recognized as one of the most competitive¹ and least diverse medical specialties^2-4. Numerous studies have highlighted the under-representation of women and racial/ethnic minorities in orthopaedics^5-8. Over the past few decades, there has been some increase in the percentage of women and minorities in orthopaedic surgery^9,10, albeit both groups are still well below their representation in medicine and the general public.

In June 2020, the Accreditation Council for Graduate Medical Education (ACGME) and the American Osteopathic Association (AOA) completed their transition to a single accreditation system for all residency programs¹¹. Although the goal was to create nationwide standards for graduate medical education, the transition to a single accreditation system and unified residency match was not without criticism, especially regarding residency selection^12,13. Critics of the single accreditation entity were largely concerned that osteopathic physicians would be more likely to match into primary care specialties and less would enter competitive surgical subspecialties compared with their allopathic counterparts^13-16.

Therefore, the purpose of this study was to compare recent application rates and match rates for traditionally under-represented groups within orthopaedic surgery such as female candidates and racial and ethnic minorities under-represented in medicine (URiM) as well as among osteopathic physicians.

Methods

A retrospective study was performed to determine application and match rates for those applying into orthopaedic surgery during the 2017 through 2021 application cycles. Data before 2017 were unavailable. Application data were collected from the Electronic Residency Application Service (ERAS) annual report released by the American Association of Medical Colleges (AAMC)¹⁷. Data collected included sex, URiM status, and degree type for US medical school graduates (MD vs. DO). Data were not available by double variables, such as URiM male and female. All data collected by this system are self-reported by the applicants during their residency applications. For the analysis of sex, applicants with unknown sex (n = 1) were excluded, leaving 7,902 included for evaluation.

Regarding race/ethnicity, applicants in the latter 2 years (2020 and 2021) of the study period were allowed to select 1 or more races/ethnicities in the ERAS application, vs. only 1 in the initial 3 years of the study period. Applicants listed with unknown (n = 449) or other (n = 285) race were excluded, leaving 7,350 applicants for evaluation. Categorization of URiM status was based on the AAMC's definition of “racial and ethnic populations that are underrepresented in the medical profession relative to their numbers in the general population.” This includes the following ERAS categories: Black or African American, Hispanic, Latino, or of Spanish Origin; American Indian or Alaskan Native; and Native Hawaiian or other Pacific Islander¹⁸.

For degree type, only applicants from a US medical school were included in the study, resulting in 1,240 international medical graduates being excluded, leaving 6,663 applicants available for study. To determine the demographic characteristics of current residents, a separate database, the Orthopaedic Residency Information Network (ORIN), was queried in June 2022. This was performed to identify current orthopaedic surgery residents throughout the United States, corresponding to those residents who largely applied during the 2017-2021 application cycles. This database is managed by the Council of Orthopaedic Residency Directors committee of the American Orthopaedic Association, relies on the self-reporting by residency programs, and was created to provide information to prospective orthopaedic surgery applicants¹⁹. At the time of query, 171 residency programs were listed. Sex was identified for 155 programs (90.6%) for a total of 3,574 residents. URiM and degree type were listed for 143 (83.6%) and 125 (73%) programs, respectively.

Statistics were performed with STATA 14.0. Descriptive statistics for all categorical variables included counts and proportions. Statistical testing performed included chi-square tests for all categorical variables, and Pearson's correlation was performed for sex, URiM status, and degree type over time. Statistical significance was defined as p < 0.05.

Results

Sex

Across the study period, there were a total of 7,902 applicants, of which 6,454 (82%) were identified as male and 1,448 (18%) were female. From 2017 to 2021, there were increasing trends in the total number of applicants (r = 0.91, p = 0.037) as well as the number of both male and female candidates applying into orthopaedic surgery (r = 0.90, p = 0.039; and r = 0.90, p = 0.039, respectively). Of the 3,574 residents identified in the ORIN database, 688 (19%) were female and 2,886 (81%) were male. There was no significant difference in the application rate when the 2 databases were compared with the match rate based on sex (p = 0.249, Table I).

TABLE I.

Percentage of Applicants and Residents in Orthopaedic Surgery Based on Sex

	Sex
	Male	Female
Applicants	81.7%	18.3%
Residents	80.7%	19.3%
		p = 0.24

Race/Ethnicity

Race/ethnicity was reported by 7,374 applicants, including 6,043 non-URiM and 1,331 URiM candidates (18% URiM, Table II). There was an increase in the number of URiM and non-URiM applicants during the study period (r = 0.92; p = 0.026; and r = 0.93; p = 0.023, respectively). There were 3,345 residents with reported race/ethnicity, with 2,714 residents not URiM and 631 URiM residents (19% URiM, p = 0.156, Table II).

TABLE II.

Percentage of Applicants and Residents in Orthopaedic Surgery Based on Race/Ethnicity^*

	Race/Ethnicity
	Non-URiM	URiM
Applicants	81.9%	18.1%
Residents	81.1%	18.9%
		p = 0.16

^*URiM = under-represented in medicine.

Degree

A total of 6,663 allopathic and osteopathic applicants from US medical schools were identified, including 5,641 allopathic and 1,022 osteopathic candidates (15% osteopathic, Table III). During the study period, there was a significant increase in the number of DO applicants (r = 0.88; p = 0.047). There were 2,856 residents identified who graduated from US medical schools, 2,521 with an MD degree and 335 with a DO degree (12% osteopathic, p < 0.002, Table III).

TABLE III.

Percentage of Applicants and Residents in Orthopaedic Surgery Based on Degree Type

	Degree
	MD	DO
Applicants	84.7%	15.3%
Residents	88.3%	11.7%
		p = 0.002

Discussion

This study sought to examine the application and match rates among various under-represented demographics within the field of orthopaedic surgery. From 2017 to 2021, we found that match rates for female and racial and ethnic minorities URiM candidates are reflective of their application rates. However, we found that osteopathic applicants in orthopaedic surgery have a lower match rate than their allopathic counterparts.

Our analysis found that match rates of female candidates are reflective of their application rates. However, just 19% of orthopaedic surgery residents were female. This is higher compared with a previous study that demonstrated only 14% female residents during the 2016-2017 academic year⁹. Although there has been an increase in female residents in the orthopaedic community, female candidates still make up a considerable minority of orthopaedic residents. A previous study showed a compound annual growth rate of just 2% in the proportion of practicing female orthopaedic surgeons from 2010 to 2019²⁰. Assuming that growth continued at this rate, they estimated that it would take 217 years to achieve gender parity (36% women in orthopaedic surgery). Thus, although match rates of female candidates are currently reflective of their application rates, the number of female residents pursuing orthopaedic surgery is far less than male residents. With increased early exposure in medical education with programs such as the Perry Initiative and improved female mentorship through the Ruth Jackson Orthopaedic Society and other organizations, the orthopaedic community can improve on the proportion of female candidates pursuing careers within orthopaedic surgery^21-23.

Our data found that match rates of URiM candidates are reflective of their application rates as well; however, just 18.9% of residents are URiM and just 18.1% of applicants are URiM. Previous studies have examined this lack of diversity in race/ethnicity among orthopaedic residency programs. A study by Poon et al. found applicants from Asian, Black, Hispanic, or other race groups had lower odds of admission into residency compared with White applicants¹⁰. Another study found that 61% of minority applicants were accepted into an orthopaedic residency vs. 73% of White applicants²⁴. Although this does conflict with our data, we did confirm the continued need for increased URiM representation within orthopaedic surgery applicants and residents. Previous studies have identified a lack of minority faculty and many other potential barriers to recruiting Black and minority applicants^25,26. To further improve diversity in orthopaedic surgery, efforts should focus on early exposure in medical education and improved mentorship with programs such as the Nth Dimension and the J. Robert Gladden Orthopaedic Society^23,25. In addition, programs must take a proactive approach to create a safe and fair working environment for URiM residents²⁵.

We found a significant difference in the match rates of osteopathic applicants compared with their allopathic counterparts, with osteopathic students significantly less likely to match into orthopaedic surgery residency. To the best of our knowledge, this is the first study to examine and compare match rates between osteopathic and allopathic applicants in orthopaedic surgery residency training. One previous study looked at the effect of the single accreditation system on DO medical student match rates in surgical subspecialties²⁷. From 2018 to 2020, they found a decrease of 3% in the total number of matched spots for DO applicants. In addition, they found that only 51.7% of DO applicants matched compared with 67.7% of MD applicants in competitive surgical specialties. This study coupled with our present study may suggest that there is an element of selection bias against DO applicants. This is also consistent with the findings of the National Ranking Matching Program's program director survey from the first 2 years after the merger of the AOA and ACGME in a single graduate medical education accreditation system. In 2021, 53% of programs would seldomly or never interview a US DO fourth year applicant compared with 9% for US MD fourth year seniors²⁸. In 2022, this trend worsened with 63% for programs now indicating that they would seldomly or never interview a US DO fourth year applicant vs. 16% for US MD fourth year applicants²⁹. As medical school accreditation status was ranked the fifth most important education and academic characteristic for a program director when determining who to interview in 2021, this may suggest the increasing value going forward and possible increasing selection bias. Another plausible explanation is that recent studies have highlighted the importance of away rotations at matching into orthopaedic surgery and there being a discord in opportunities and fees among DO and MD applicants applying for residency, leading to a difference in application and match rates^30-36. Further studies are needed to elucidate the etiology and nature of this selection bias.

This study is not without limitations. ORIN is self-reported, and thus gender, race/ethnicity, and degree data may not exactly match ERAS data. Second, at the time the ORIN database was queried, approximately 80% of all residents nationally were accounted for in the database. The residents who are unaccounted for may bias the results, especially in the race/ethnicity and degree categories, which were slightly less accounted for in ORIN than sex. In addition, it is assumed that all current residents in June 2022 (when ORIN was queried) applied to orthopaedic surgery residency in the 2017-2021 application cycles. This may not exactly be the case for a small number of residents on a 6-year track, those who took a leave of absence, or those who left their residency program and thus are not included in the ORIN database. Given how ERAS and ORIN data are presented, we are unable to differentiate application or residency characteristics in those applicants/residents who fall under more than 1 minority category. Fourth, during the study period, the single accreditation system finalized its merger among residency programs with all traditionally ACGME and AOA programs being unified under a single system. Although this may confound the findings, ORIN has been reported to be the most reliable publicly accessible database for reporting the degree types of current residents³⁷. Longer term studies are needed to validate the results of this study. Finally, we did not control for characteristics that have been previously associated with matching in orthopaedic surgery, such as Alpha Omega Alpha status, US Medical Licensing Exam scores, publications, and volunteer experience¹⁰. Future studies should attempt to control for these variables in looking at differences in orthopaedic surgery match rates. However, given the lack of availability of National Residency Match Program data, this type of research is inherently difficult.

Conclusion

In recent years, there has been a significant and necessary push to increase diversity in the field of orthopaedic surgery. From 2017 to 2021, match rates of female and URiM candidates are reflective of their application rates. Osteopathic applicants in orthopaedic surgery have lower match rates than their allopathic counterparts.