Update on workforce diversity in vascular surgery

Abstract

Objective:

Creating a diverse workforce is paramount to the success of the surgical field. A diverse workforce allows us to meet the health needs of an increasingly diverse population and to bring new ideas to spur technical innovation. The purpose of this study was to assess trends in workforce diversity within vascular surgery (VS) and general surgery (GS) as compared with orthopedic surgery (OS)—a specialty that instituted a formal diversity initiative over a decade ago.

Methods:

Data on the trainee pool for VS (fellowships and integrated residencies), GS, and OS were obtained from the U.S. Graduate Medical Education reports for 1999 through 2017. Medical student demographic data were obtained from the Association of American Medical Colleges U.S. medical school enrollment reports. The representation of surgical trainee populations (female, Hispanic, and black) was normalized by their representation in medical school. We also performed the χ² test to compare proportions of residents over dichotomized time periods (1999–2005 and 2013–2017) as well as a more sensitive trend of proportions test.

Results:

The proportion of female trainees increased significantly between the time periods for the three surgical disciplines examined (P < .001). Hispanic trainees also represented an increasing proportion of all three disciplines (P ≤ .001). The proportion of black trainees did not significantly change in any discipline between the two periods. Relative to their proportion in medical school, Hispanic trainees were well represented in all surgical specialties studied (normalized ratio [NR], 0.95–1.52: 0.95 OS, 1.00 GS, 1.53 VS fellowship, and 1.23 VS residency). Compared with their representation in medical school, women were under-represented as surgical trainees (NR: 0.32 OS, 0.82 GS, 0.56 VS fellowship, and 0.78 VS residency) as were black trainees (NR: 0.63 OS, 0.90 GS, 0.99 VS fellowship, and 0.81 VS residency).

Conclusions:

Although there were significant increases in the number of women and Hispanic trainees in these three surgical disciplines, only Hispanic trainees enter the surgical field at a rate higher than their proportion in medical school. The lack of an increase in black trainees across all specialties was particularly discouraging. Women and black trainees were under-represented in all specialties as compared with their representation in medical school. The data presented suggest potential problems with recruitment at multiple levels of the pipeline. Particular attention should be paid to increasing the pool of minority medical school graduates who are both interested in and competitive for surgical specialties.

Research shows that minority and female physicians contribute to the medical workforce in critical ways. For example, minorities and women serve low-income and racial minority populations at higher rates than their majority colleagues.^1–3 These contributions are important given that racial and ethnic disparities in vascular disease outcomes are well documented. Minorities are more likely to present with advanced peripheral arterial disease, undergo amputation, and experience complications and death after endovascular aneurysm repair.^4–6 The causes of these disparities are complex with social and economic inequities, decreased access to health care, physician bias, and systemic discrimination among the many contributing factors. One proposed solution to aid in decreasing racial disparities is increasing the proportions of racial and ethnic minorities entering the medical field.^1,7 However, despite their rapidly growing proportion of the U.S. population, minorities continue to be under-represented among the vascular surgery (VS) workforce. The Society of Vascular Surgery reports that black and Hispanic vascular surgeons comprise just 2% and 3.8% of its members, respectively.⁸ Although these percentages may slightly under-represent these groups given that members may choose not to divulge their race and/or ethnicity, it at least provides a starting point for further conversation. Similarly, although there have been great strides in achieving gender equity among medical school matriculants,⁹ there continues to be disproportionately low representation of women in surgical specialties. In particular, only 13% of active vascular surgeons are women.¹⁰ It is worth noting that the Accreditation Council for Graduate Medical Education (ACGME) recently established guidelines that reprioritize this issue of diversity among trainees.^11,12

The existing literature regarding diversity among VS trainees is discouraging. Kane et al¹³ demonstrated that no significant increases in black, Hispanic, or female trainees occurred between 1999 and 2005 in VS fellowships, whereas programs with formal diversity recruitment initiatives like orthopedic surgery (OS) demonstrated positive trends in at least one under-represented group. More specifically, women and Hispanic trainees saw a small, but significant, increase over the study period. Abelson et al¹⁴ demonstrated that VS had the second-lowest rate of growth in women trainees between 1994 and 2014 among general surgery (GS) and surgical subspecialties—only thoracic surgery was lower. The Abelson study conducted no analysis on racial/ethnic minorities, and neither study differentiated integrated vascular residents from vascular fellows. Therefore, the current study aims to perform an updated analysis of racial and gender diversity among integrated VS residents and VS fellows in comparison to GS and OS trainees. We hypothesize the following: there will be a significant increase in women across the time period in all specialties, there will be a minimal increase in black and Hispanic trainees, and all groups will be under-represented as you move from medical school to surgical training programs.

METHODS

For surgical trainees, data from VS (fellowship and residency), GS, and OS were derived from the U.S. Graduate Medical Education reports from 1999 to 2005 and 2013 to 2017. Data on the gender, race, and ethnicity of all residents and fellows of accredited training programs are publicly available in this report. In addition to evaluating changes in female representation, we chose to focus our analysis on black and Hispanic trainees given that they are the largest minority groups in the United States. Similar to methods employed by Kane et al,¹³ the proportion of women, black, and Hispanic trainees for each specialty was calculated by dividing the number of trainees in each subgroup by the total number of trainees in that specific specialty for that year. The proportions, expressed as percentages, were calculated over the two time periods. We also evaluated changes in the proportion of women, black, and Hispanic trainees across time periods as a continuous analysis by testing for trends comparing the aggregated 1999–2005 time period and each year between 2013 and 2017.

We also set out to evaluate our hypothesis that a “leaky pipeline” for minorities exists; that is, in transitioning from medical school to residency, minority representation further decreases. To evaluate this hypothesis, we compared medical school representation of women and minorities to their representation in surgical specialties. Medical student data were derived from the American Medical Association United States medical enrollment reports. Minority representation was then reported as a normalized ratio whereby the ratio of a minority group within a surgical specialty was divided by their respective ratio in medical school. This comparison allows us to evaluate change across groups in a meaningful way, given large variation in the baseline numbers of women, black, and Hispanic trainees in medical school and residency.

χ² tests were used to compare proportions of women, black, and Hispanic trainees across the two time periods. A more sensitive test—the test for trends in proportions—was used to evaluate for significant trends across time. A P value of less than .05 was considered statistically significant.

RESULTS

The percentages of women, black, and Hispanic trainees in the VS, GS, and OS for the two time periods (1999–2005 and 2013–2017) are shown in the Table and the Supplementary Table (online only). For women, there was a statistically significant increase (P < .001) in representation for the VS fellowship (13.8%−27%), GS (25%−38.5%), and OS (10%−14.6%). For Hispanic trainees, a significant increase in representation was also seen across all disciplines (VS, 5.2%−8.7%; GS, 6.1%−8.4%; OS, 3.3%−5.3%; P < .001). Black trainees saw no significant increase in representation in the specialties analyzed across these two time periods. The VS residencies could not be compared in the same way given their introduction after the first time period. Overall, when comparing the previous era with a more contemporary era, there have been significant increases in the representation of women and Hispanic surgical trainees; however, there has not been the same increase in black trainees.

Table.

Proportion of women, black, and Hispanic trainees in each surgical discipline, 1999–2005 and 2013–2017

	Specialty	1999–2005 (%)	2013–2017 (%)	P value
Women	Vascular residency	N/A	36.7	–
	Vascular fellowship	13.8	27	<.001
	General surgery	25	38.5	<.001
	Orthopedic surgery	10	14.6	<.001
Black	Vascular residency	N/A	5.1	–
	Vascular fellowship	4.3	5.4	NS
	General surgery	5.9	5.7	NS
	Orthopedic surgery	4	4	NS
Hispanic	Vascular residency	N/A	7.3	–
	Vascular fellowship	5.2	8.7	<.001
	General surgery	6.1	8.4	<.001
	Orthopedic surgery	3.3	5.3	<.001

N/A, Not applicable; NS, not significant.

The trends in proportions over time were also analyzed for each subgroup across the four training programs (Fig 1). Women have seen a significant increase in all programs except the VS residency (P < .0001). For Hispanic trainees, there has been a significant increase across time in GS and OS (P < .0001); an increase was not seen in either the VS fellowship or residency. For black trainees, there was no significant increase over time for the VS residency or OS. In contrast, there was an appreciable increase in black representation in the VS fellowship (P < .05) and GS (P < .0001). Overall, the VS integrated residency program has started and continues to train a relatively large proportion of women. Although there has been a small, but significantly increased trend in the number of black trainees in VS fellowship, there has been a more significant decline in the number of black trainees in GS and no significant changes in representation in the VS residency or OS.

Fig 1.

Trends over time in the representation of women, black, and Hispanic trainees by surgical specialty. *P <.05, **P < .0001. Arrows indicative of a general trend (up or down). NS, Not significant.

Normalized ratios were calculated to illustrate the differences between the proportions of surgical residents vs medical students. If minority populations are represented equally in residency and medical school, they will have a normalized ratio of 1. As illustrated in Fig 2, in OS, although women represent nearly 50% of medical students, only 15% of the OS trainee population are women, yielding a normalized ratio of 0.3 and indicating relatively poor representation at the residency level. In contrast, the ratio for GS, VS fellowship, and VS residency was 0.8, 0.6, and 0.8, respectively. For black students and surgical trainees, the ratio in OS, GS, VS fellowship, and VS residency was 0.6, 0.9, 0.98, and 0.9, respectively. For Hispanics, the ratios for OS, GS, VS fellowship, and VS residency were 0.9, 1.5, 1.5, and 1.2, respectively. What these numbers tell us in aggregate are that though there has been a large increase in women surgical trainees over the years, women are still not well represented in surgical training overall. Black surgical trainees are relatively well represented when compared with their proportion in medical school, indicating that their numbers are equally low in medical school. Interestingly, Hispanic trainees make up a larger proportion of surgical trainees in most fields studied relative to their proportion in medical school.

Fig 2.

Normalized ratios of women, blacks, and Hispanics in surgical training. Proportion of each group compared with their respective proportions in medical school.

DISCUSSION

Although there were significant increases in the number of women and Hispanic trainees in these three surgical disciplines over the study period, only Hispanic trainees enter the surgical field at a rate similar to their proportion in medical school. This finding regarding Hispanic trainees refutes our hypothesis, but was a pleasant surprise. Women trainees are under-represented in all specialties as compared with their representation in medical school. Black surgical trainees have not markedly increased over time, although this appears to be in part due to their relatively low numbers in medical school.

It has been widely documented that more women are pursuing surgical fields, although they continue to be under-represented among the surgical workforce.^15–17 Our results support these findings and are consistent with our hypothesis, as OS, GS, and VS fellowships all saw significant increases in the proportions of women trainees, though normalized ratios point to continued under-representation. Looking specifically at VS fellowship, the proportion of women trainees nearly doubled over the period examined. Although this growth is a promising indication that the VS workforce is trending toward gender equity, the current ratio of women in VS is still below those making up medical school classes where women now outnumber men.⁹ Interestingly, our study also found that integrated VS residencies had the second highest proportion of women trainees despite showing no significant change in representation between 2013 and 2017. This observation supports literature suggesting that integrated vascular programs attract women to the specialty.¹⁸

Extensive research has been conducted to elucidate why women choose career paths outside of surgery. Perceived incompatibility with family life, negative perceptions of surgical personalities, male-dominated culture, lifestyle concerns, and a lack of role models have all been cited as barriers.^19–21 One reason why integrated vascular residencies may have success in recruiting female candidates is that these programs accelerate training from the traditional 5+2 paradigm. Reaching career independence sooner equates to more flexibility and lifestyle control. Furthermore, women are more likely to experience sexual harassment and gender-based discrimination during training, which likely contributes to higher attrition rates from surgical residencies.^21,22 Therefore, integrated VS programs may aid in recruiting women for a more practical reason—the 0+5 pathway removes one point of attrition because trainees enter straight from medical school instead of after a GS residency.

Our study found that the representation of Hispanic trainees increased among VS fellowship, GS, and OS. In fact, Hispanic trainees are represented in these fields at greater proportions than those in medical school classes. One theory for this finding is that these specialties are recruiting international medical graduates (IMGs) from Hispanic ethnicities. This theory is supported by the fact that increased representation was found in vascular fellowships, which until recently struggled to fill training spots and were therefore more open to IMGs. OS, which we did not find to have an over-representation of Hispanics, is a notoriously competitive specialty among U.S. medical school graduates and therefore has been less accessible to IMGs. To illustrate, according to National Resident Matching Program data, international graduates made up 5% of GS trainees, 4% of VS fellows, and less than 1% of OS residents in 2018. Nevertheless, these findings suggest that Hispanics have the potential to make up a representative portion of the vascular surgical workforce in the future. Further research must be conducted to uncover how the recruitment of Hispanics has been successful. It is worth noting that though the proportions of Hispanic VS trainees were encouraging, they still have not reached the proportion seen in the general U.S. population (18.5%).²³

Unfortunately, the representation of black trainees did not improve in any surgical specialty examined. In addition, the range of black trainees in each specialty fell between 4% and 6%, which is well below the 13.4% of blacks estimated in the U.S. population.²³ However, we did find that VS fellowships recruit near equal proportions of black trainees as those represented in medical school classes. This is an encouraging finding, as it suggests that the pipeline leading to VS has been successful in retaining promising black candidates from medical school through to fellowship. In comparison, the normalized ratio of black trainees in GS and integrated VS is 0.9 and 0.8, respectively. On one hand, these statistics indicate relative success in black trainee retainment; however, they also highlight a “leaky” pipeline, as large percentages of eligible black candidates continue to be lost at different stages of training.

In July 2019, the ACGME mandated that programs “engage in practices that focus on mission-driven, ongoing, systematic recruitment and retention of a diverse workforce of residents, fellows (if present), faculty members, senior administrative staff members, and other relevant members of its academic community” in order to maintain accreditation.¹¹ These new guidelines were modeled after similar requirements for medical school accreditation by the Liaison Committee of Medical Education. There is promising evidence that diversity has increased among medical students since implementation in 2009.¹²

To meet new ACGME guidelines programs will need to rethink their recruitment strategies in order to attract more diverse candidates. A recent study by Ku et al²⁴ analyzed perceptions of gender and racial diversity by Stanford residency interviewees and the importance of diversity on rank list. They found that women were more sensitive to perceiving gender imbalances, and under-represented minority (URM) interviewees were more sensitive to perceiving racial imbalances among program residents and faculty. Most importantly, women and URM interviewees were more likely to state that perceived gender and racial diversity, respectively, had a positive influence on their ranking of the program.²⁴ These findings present a dilemma—if visible diversity is necessary in order to recruit women and URMs, how should programs lacking in diversity proceed?

First, programs must work to support their current URM and women trainees. URM residents report experiencing discrimination, lack of social support, differing expectations, and pressure to prove belonging in the workplace.^25,26 Similarly, evidence of gender-specific obstacles for women in medicine is regrettably abundant.^22,27,28 A recent national, multicenter study of surgical trainees reports that over 70% of women surveyed had experienced at least one episode of sexual harassment during surgical training.²² These realities are unacceptable. Academic surgical leadership must reflect on their own programs to ensure that they are providing training environments that allow minority and female trainees to succeed. The American Surgical Association published a paper outlining benchmarks and tools to aid in this mission of achieving equity, diversity, and inclusion within the surgical field—this is a strong resource for program leaders to begin honest self-evaluation of their training cultures.²⁹ Second, graduate medical education programs must recognize their role in recruiting diverse students at lower levels of training. Although residency programs are limited to the medical school graduate pool in order to select trainees, programs can directly influence the development of medical students, undergraduates, and even high school students through participation in pipeline programming and mentorship. Furthermore, it is critical that participation in these activities is distributed among all faculty, including nonminorities, so that URM trainees and faculty do not pay a “minority tax.”

In addition, it is imperative that surgical leaders regularly track the number of URMs and women who populate their training programs. Recruitment and retention strategies must continuously be assessed to ensure that they are ultimately leading to success in the given specialty. Finally, leadership buy-in is essential to the success of any initiative. Surgical leaders should state their commitment to increasing workforce diversity making clear that it is the responsibility of everyone involved in the delivery of surgical care to uphold that commitment.

This study has various limitations that should be acknowledged. First, the medical student data from the Association of American Medical Colleges do not take into consideration medical students graduating from osteopathic medical schools, IMGs, or nonsenior U.S. graduates. However, according to National Resident Matching Program data, the percentages of GS, VS, and OS trainees entering training from these programs are very small.³⁰ Second, because integrated vascular residencies are relatively new, there is no sufficient data to analyze between two different time periods. However, this is the first study that attempts to characterize the racial/ethnic and gender makeup of integrated vascular trainees. More studies are needed in the future to explore trends over time. It should also be noted that “Hispanic” is reported as an ethnicity and not a race. It is unclear how the U.S. Graduate Medical Education or American Medical Association categorizes black Hispanics. It is our assumption that those individuals are included in both the black and Hispanic categories for each report.

In summary, regarding gender and racial/ethnic equity in VS, there has been progress but there is more work to be done. Our study demonstrates that the proportion of historically under-represented groups like women and Hispanics is increasing in VS, but continued efforts are necessary to achieve gender and racial/ethnic equity. Integrated vascular programs may be a solution in increasing the number of women entering the field. Targeted efforts to increase and retain black trainees and medical school matriculants should be encouraged for all vascular faculty. Lastly, making strides toward greater diversity in VS will require honest, self-reflective efforts by surgical department leadership.