What Is the Representation of Sexual and Gender Minority Identities Among Orthopaedic Professionals in the United States?

Abstract

Background

There is substantial corroborating evidence that orthopaedic surgery has historically been the least diverse of all medical and surgical specialties in terms of race, ethnicity, and sex. Growing recognition of this deficit and the benefits of a diverse healthcare workforce has motivated policy changes to improve diversity. To measure progress with these efforts, it is important to understand the existing representation of sexual and gender minorities among orthopaedic professionals.

Questions/purposes

(1) What proportion of American Academy of Orthopaedic Surgeons (AAOS) members reported their identity as a sexual or gender minority? (2) What demographic factors are associated with the self-reporting of one’s sexual orientation and gender identity?

Methods

The AAOS published the updated membership questionnaire in January 2022 to collect information from new and existing society members regarding age and race or ethnicity and newly added categories of gender identity, sexual orientation, and pronouns. The questionnaire was updated with input from a committee of orthopaedic surgeons and researchers to ensure face validity. The AAOS provided a deidentified dataset that included the variables of interest: membership type, gender identity, sexual orientation, pronouns, age, race, and ethnicity. Of 35,427 active AAOS members, 47% (16,652) updated their membership questionnaire. To answer our first study question, we calculated the prevalence of participants who self-reported as lesbian, gay, bisexual, transgender, queer, or another sexual or gender minority identity (LGBTQ+) and other demographic characteristics of the 16,652 respondents. Categorical demographic data are described using frequencies and proportions. Median and IQR were used to describe the central tendency and variability. To answer our second study question, we conducted a stratified analysis to compare demographic characteristics between those who self-reported LGBTQ+ identity and those who did not. Visual methods (quantile-quantile plots) and statistical tests (Kolmogorov-Smirnov and Shapiro Wilk) confirmed that the age of AAOS member was not normally distributed. Therefore, a Kruskal Wallis test was used to determine the statistical associations between age and self-reported LGBTQ+ status. Chi-square tests were used to determine bivariate statistical associations between categorical demographic characteristics and self-reported LGBTQ+ status. A multivariable logistic regression model was developed to identify the independent demographic characteristics associated with respondents who self-reported LGBTQ+ identity. Further stratified analyses were not conducted to protect the anonymity of AAOS members. An alpha level of 5% was established a priori to define statistical significance.

Results

Overall, 3% (109 of 3679) and fewer than 1% (3 of 16,182) of the AAOS members (surgeons, clinicians, allied healthcare providers, and researchers) who updated their membership profiles reported identifying as a sexual (lesbian, gay, bisexual, queer) or gender minority (nonbinary or transgender), respectively. No individual self-identified as transgender. Five percent (33 of 603) of women and 3% (80 of 3042) of men self-identified as a sexual minority (such as lesbian, gay, bisexual, or queer). AAOS members who self-identified as LGBTQ+ were younger (OR 0.99 [95% confidence interval (CI) 0.98 to 0.99]; p < 0.001), less likely to self-identify as women (OR 0.86 [95% CI 0.767 to 0.954]; p < 0.001), less likely to be underrepresented in medicine (OR 0.49 [95% CI 0.405 to 0.599]; p < 0.001), and less likely to be an emeritus or honorary member (OR 0.75 [95% CI 0.641 to 0.883]; p < 0.003).

Conclusion

The proportion of self-reported LGBTQ+ AAOS members is lower than the 7% of the general US population. The greater proportion of younger AAOS members reporting this information suggests progress in the pursuit of a more-diverse field.

Clinical Relevance

The study findings support standardized collection of sexual orientation and gender identity data to better identify and address diversity gaps. As orthopaedic surgery continues to transform to reflect the diversity of musculoskeletal patients, all orthopaedic professionals (surgeons, clinicians, allied healthcare providers, and researchers), regardless of their identities, are essential in the mission to provide equitable and informed orthopaedic care. Sexual and gender minority individuals may serve as important mentors to the next generations of orthopaedic professionals; individuals from nonminority groups should serve as important allies in achieving this goal.

Introduction

In 2022, at least 7% of the US population identified as lesbian, gay, bisexual, transgender, queer, or another sexual or gender minority identity (LGBTQ+) [21]. Among adults aged 18 to 26, 21% identified as LGBTQ+ [21]. Despite overall advancements in the rights of LGBTQ+ individuals in the United States, sexual and gender minorities continue to experience substantial health and healthcare disparities [18, 22, 23, 31-33, 44]. Some of these disparities are directly relevant to musculoskeletal health. For instance, national population-based studies have demonstrated poorer bone health among gay men than among heterosexual men, independent of known disparities in lifestyle and psychosocial risk factors, such as smoking [13]. Similarly, transgender women receiving feminizing hormone therapy are at higher risk of low bone mineral density [39, 43]. Such disparities among sexual and gender minorities require unique consideration and knowledge in the setting of orthopaedic care. There is a positive association between a diverse healthcare workforce and the quality of patient care. In contrast, a lack of workforce diversity has been associated with poorer health outcomes, lower healthcare access, decreased patient trust, diminished workplace satisfaction, and difficulties with employee retention [1, 16]. The positive impact of a diverse workplace is most felt among patients and employees from historically marginalized groups [18].

Orthopaedic surgery remains the least diverse of all medical and surgical specialties in terms of gender and race [2, 7], and one that has demonstrated several barriers to increasing diversity. As observed among people with other marginalized identities, LGBTQ+ medical students are less likely to pursue the most competitive specialties for many reasons, including lack of mentorship and perceived lack of inclusivity and belonging [12, 37]. In fact, graduating LGBTQ+ medical students were less likely to pursue orthopaedic surgery than any other specialty [28]. Furthermore, LGBTQ+ individuals experience professional barriers to pursuing certain medical specialties, including higher rates of mistreatment, fear of discrimination during the residency or fellowship application process, and discrimination from patients and colleagues [26, 28, 37, 42]. To address these concerns and diversify orthopaedic surgery, local and national institutions, including the American Academy of Orthopaedic Surgeons (AAOS), have prioritized the advancement of diversity, equity, and inclusion efforts [2, 11, 25]. The current proportion of LGBTQ+ orthopaedic professionals is unknown, as is the extent to which they self-report their sexual orientation and gender identity. This information is essential to measure progress and address persistent diversity gaps.

We therefore asked: (1) What proportion of AAOS members reported their identity as a sexual or gender minority? (2) What demographic factors are associated with the self-reporting of one’s sexual orientation and gender identity?

Patients and Methods

Study Design and Setting

This was a cross-sectional study and secondary analysis of 47% (16,652) of AAOS members who responded to the updated AAOS membership form from the 35,427 active AAOS members between January 2022 and May 2023 (described in “Survey Design”) [3]. The AAOS provided us with a deidentified dataset that included selected variables of interest from the AAOS membership directory (membership type, gender identity, sexual orientation, pronouns, age, and race or ethnicity).

To answer our first study question regarding the proportion of AAOS members who reported their identity as a sexual or gender minority, we calculated the proportion of individuals who self-reported as LGBTQ+ among the number of active AAOS members. To answer our second study question regarding demographic factors associated with the self-reporting of one’s sexual or gender identity, we conducted a stratified analysis to compare demographic characteristics between those who self-reported LGBTQ+ identity and those who did not.

Throughout this study, sexual orientation is defined as how a person characterizes their attraction to other people. A sexual minority includes all responders who self-identified as lesbian, gay, bisexual, or queer. Gender identity is defined as a person’s inner sense of being a woman, man, or another gender, and this may not align with their sex assigned at birth (male, female, or intersex) [6]. A gender minority in this study includes all responders who self-identified as transgender (identifying as a gender that does not align with their sex assigned at birth) or nonbinary (not ascribing to the gender binary of man or woman) [6].

Survey Design

The AAOS published an updated membership form in January 2022 to collect information from new and existing society members. In addition to existing questions regarding self-reported date of birth, race, and ethnicity, the updated membership form included questions regarding self-identified gender identity, sexual orientation, and pronouns. This questionnaire was updated with input from a committee of orthopaedic surgeons and researchers to ensure face validity (two study authors were members of this committee). All new membership applicants and existing members received emails prompting them to review and update their membership profiles using the new membership questionnaire. The AAOS data were provided “as-is” with the disclaimer that they make no representation as to the accuracy or completeness of the data. Of the 35,427 active AAOS members, 47% (16,652) updated their membership questionnaires.

Ethical Approval

After reviewing the study protocol, the Wake Forest University School of Medicine Institutional Review Board determined that the study does not meet the federal definition of research involving human subject research as outlined in the federal regulations 45 CFR 46. 45 CFR 46.102(f) because the data are not individually identifiable and were not obtained for research purposes.

Statistical Analyses

We performed all data management and statistical analyses using SAS/STAT® Version 15.1 (SAS Institute). Categorical demographic data (membership type, self-reported gender identity, sexual orientation, pronouns, race, and ethnicity) have been described using frequencies and proportions, and chi-square tests were used for statistical comparisons between subgroups (LGBTQ+ and non-LGBTQ+). The age of AAOS members was the only numeric variable included, and visual methods (quantile-quantile plots) and statistical tests (Kolmogorov-Smirnov and Shapiro Wilk) confirmed that age was not normally distributed. Therefore, we used median and IQR to describe the central tendency and variability, and Wilcoxon and Kruskal-Wallis tests were used to determine statistical associations between age and other characteristics (gender identity, sexual orientation, pronouns, race, and ethnicity). A multivariable logistic regression model was developed to identify the independent demographic characteristics associated with respondents who self-reported LGBTQ+ identity. An a priori significance level of less than 0.05 was applied for all analyses. Because triangulating multiple variables for subset analyses yields small samples, we determined that it was possible to identify the respondent. Therefore, we did not conduct extensive stratified analyses to protect the anonymity of the AAOS members. To protect responder anonymity when stratifying responses by self-reported LGBTQ+ identity, we consolidated race into three categories: underrepresented minority (URM), Asian, and White. As defined by the Association of American Medical Colleges, URM included Black, Mexican-American, Native American (including American Indian, Alaska Native, and Native Hawaiian), and Puerto Rican people [4]. Because there were missing responses, we created binary variables to define whether the following demographic information was reported: gender identity, race, ethnicity, sexual orientation, and pronouns.

Results

What Proportion of AAOS Members Reported They Identify as a Sexual or Gender Minority?

Overall, 3% (109 of 3679) and fewer than 1% (3 of 16,182) of the AAOS members who updated their membership profiles reported identifying as a sexual or gender minority, respectively. Of the 109 orthopaedic professionals who self-identified as a sexual minority, no members reported self-identifying as transgender. Three percent (3 of 109) reported having a nonbinary gender identity. In addition to the 72 LGBTQ+ individuals, 47 reported they were questioning or identified with another sexual minority identity.

Most responders were men (84% [13,655 of 16,182]), White (69% [9095 of 13,115]), non-Hispanic (91% [11,053 of 12,165]), heterosexual (89% [3260 of 3679]), and use “he/him/his” pronouns (72% [2312 of 3196]) (Table 1). The median (IQR) age of the 16,652 orthopaedic professionals was 49 years (37 to 67).

Table 1.

Respondent characteristics

Characteristics	% (frequency)
Race (n = 13,115)
White	69 (9095)
Asian	13 (1665)
Chose not to report	7 (950)
Black	6 (737)
Multiracial	4 (532)
American Indian or Alaskan Native	1 (107)
Native Hawaiian or Pacific Islander	< 1 (29)
Ethnicity (n = 12,165)
Non-Hispanic	91 (11,053)
Hispanic	9 (1112)
Gender identity (n = 16,182)
Men	84 (13,655)
Women	16 (2524)
Nonbinary	< 1 (3)
Transgender	0 (0)
Sexual orientation (n = 3679)
Heterosexual	89 (3260)
Choose not to report	8 (300)
Other	1 (37)
Gay	1 (28)
Lesbian	1 (22)
Bisexual	< 1 (16)
Not sure	< 1 (10)
Queer	< 1 (6)
Pronouns (n = 3196)
He/him/his	72 (2312)
She/her/hers	20 (631)
Choose not to report	7 (215)
Another set of pronounsa	1 (31)
They/them/theirs	< 1 (7)
AAOS membership category (n = 16,652)
Active fellow member	46 (7690)
Emeritus or honorary	20 (3360)
Resident or fellow member	18 (3054)
Associate member	15 (2548)
Candidate AAOS member (n = 16,652)
No	87 (14,523)
Yes	13 (2129)

Data presented as % (n). The following data were self-reported by responders: race, ethnicity, gender identity, sexual orientation, and pronouns.

^aSome individuals use another set of pronouns to describe themselves. Such examples include “she/they” and “ze/hir/hirs.”

What Demographic Factors Are Associated With the Self-reporting of One’s Sexual Orientation and Gender Identity?

After adjusting for potentially confounding variables, we found that AAOS members whose self-reported identity was LGBTQ+ were younger (OR 0.99 [95% confidence interval (CI) 0.98 to 0.99]; p < 0.001), less likely to self-identify as women [OR 0.86 [95% CI 0.77 to 0.95]; p < 0.001), less likely to be a URM (OR 0.49 [95% CI 0.41 to 0.60]; p < 0.001), and less likely to be an emeritus or honorary member (OR 0.75 [95% CI 0.64 to 0.88]; p < 0.003) (Table 2). The unadjusted results comparing demographic and membership characteristics between those whose self-reported identity was LGBTQ+ were, in most ways, similar to those of the adjusted analysis (Table 3).

Table 2.

Multivariable logistic regression modeling odds of self-reporting as a sexual or gender minority

Effect	OR (95% CI)	p value
Gender
Women (comparator group is men)	0.86 (0.77 to 0.95)	< 0.001
Ethnicity
Hispanic (comparator group is non-Hispanic)	1.01 (0.88 to 1.16)	< 0.001
Racea (comparator group is White)
URM + Asian	0.49 (0.41 to 0.60)	< 0.001
AAOS membership group (comparator group is active fellow member)
Associate member	0.99 (0.83 to 1.19)	0.45
Emeritus or honorary member	0.75 (0.64 to 0.88)	0.003
Learner (resident, fellow)	1.07 (0.93 to 1.23)	0.03
Candidate member
Yes (comparator group is noncandidate members)	0.72 (0.60 to 0.85)	0.08
Age in years	0.99 (0.98 to 0.99)	< 0.001

^aTo protect anonymity when stratifying by self-reported LGBTQ+ identity, a URM in medicine category was implemented as described by the AAMC. The AAMC defines URM as “those racial and ethnic populations that are underrepresented in the medical profession relative to their numbers in the general population [4]; these groups include Black, Mexican-American, Native American (including American Indian, Alaska Native, and native Hawaiian), and Puerto Rican; URM = underrepresented minority.

Table 3.

Association between gender identity, race or ethnicity, and membership classifications, stratified by self-reported LGBTQ+ identity

	All	Chose not to report	Not LGBTQ+	LGBTQ+	p value
Age in years	49 (37-67)	58 (42-68)	45 (35-62)	45 (36-65)	< 0.001
Gender identity					< 0.001
Woman	603	4 (26)	90 (544)	5 (33)
Man	3042	8 (252)	89 (2710)	3 (80)
Nonbinary	3	0 (0)	0 (0)	3 (100)
Ethnicity					0.02
Hispanic	332	4 (12)	95 (314)	2 (6)
Non-Hispanic	3165	7 (219)	90 (2838)	3 (108)
Racea					0.01
Asian	361	5 (19)	92 (333)	2 (9)
URM	391	4 (14)	93 (365)	3 (12)
White	2745	7 (198)	89 (2454)	3 (93)
AAOS membership category					< 0.001
Active fellow member	1868	9 (175)	88 (1648)	2 (45)
Associate member	545	6 (34)	90 (489)	4 (22)
Emeritus or honorary	467	13 (62)	81 (378)	6 (27)
Resident or fellow (trainee) member	799	4 (29)	93 (745)	3 (25)
Candidate AAOS member					0.059
No	3214	9 (275)	88 (2837)	3 (102)
Yes	465	5 (25)	91 (423)	4 (17)

Data presented as median (IQR), % (n) or number; the following data were self-reported by responders: race, ethnicity, gender identity, sexual orientation, and pronouns.

^aTo protect anonymity when stratifying by self-reported LGBTQ+ identity, a URM in medicine category was implemented as described by the AAMC. The AAMC defines URM as “those racial and ethnic populations that are underrepresented in the medical profession relative to their numbers in the general population [4]; these groups include Black, Mexican-American, Native American (including American Indian, Alaska Native, and native Hawaiian), and Puerto Rican; URM = underrepresented minority.

Discussion

Extensive evidence supports that a lack of diversity in healthcare is associated with poorer health and healthcare outcomes, barriers to healthcare access, decreased patient trust, and reduced workplace satisfaction and employee retention [1, 16]. It has been well established that orthopaedic surgery is the least-diverse medical specialty with regard to race, gender, and ethnicity [25], but what is less clear is where orthopaedic surgery stands with regard to LGBTQ+ diversity [2, 7]. This study determined that 3% (109 of 3679) and fewer than 1% (3 of 16,182) of the AAOS members who updated their membership profiles reported identifying as a sexual or gender minority, respectively; 89% identified as heterosexual, with 8% choosing not to report their sexual orientation. Furthermore, we found that age was the only demographic characteristic associated with self-report of one’s sexual orientation; responders who chose not to report their sexual identity were older than those who chose to report this information. These findings support standardized collection of sexual orientation and gender identity data to better measure progress in addressing diversity gaps. As the field of orthopaedic surgery continues to transform to reflect the diversity of musculoskeletal patients, all orthopaedic professionals (surgeons, clinicians, allied healthcare providers, and researchers), regardless of their identities, are essential in the mission to provide equitable and informed orthopaedic care.

Limitations

First, the AAOS data were provided “as-is” with the disclaimer that they make no representation as to the accuracy or completeness of the data. Members are not required to provide the AAOS with this information, and the information is self-reported. As a result, there were missing data that may have introduced selection bias. There are several reasons why respondents may have chosen not to complete certain demographic questions on the AAOS membership form. It is likely that some people simply do not want to provide what they may consider “unnecessary” information for an organization to possess about them. Other individuals may simply not want to take the additional time to complete voluntary questionnaires. Some members of AAOS may be unfamiliar with the concepts of gender identity, sexual orientation, and pronoun use, and they may have opted to leave these questions blank. Conversely, there are known personal and institutional barriers to disclosure of a sexual or gender minority identity, such as discrimination or mistreatment [5, 17]. Prior research has found that among lesbian, gay, and bisexual individuals, bisexual women and men are less likely to disclose their sexual identity in national studies [9]. For these reasons, we expect these data in fact underestimate the true proportion of sexual or gender minority members of AAOS. We further explore how to improve response rates from sexual and gender minorities in the “Other Considerations” subsection.

Finally, because of sample sizes, we limited the extent of stratified analyses to protect the respondents’ anonymity. In one instance, to stratify by LGBTQ+ identity, we aggregated several racial identities under the Association of American Medical Colleges definition of URM in medicine for this reason. As such, we were unable to analyze LGBTQ+ identity among racial minorities. However, we believe that reporting larger aggregates of data will build trust among respondents to feel safe self-disclosing sensitive personal information. We were able to report the representation of sexual and gender minorities among AAOS members without compromising this trust. Similar studies of professional organizations in other medical specialties have limited their analyses as well for this reason [27].

What Proportion of AAOS Members Reported They Identify as a Sexual or Gender Minority?

We found that 89% of AAOS members who updated their new membership questionnaire identify as heterosexual, with 8% choosing not to report their sexual orientation and approximately 3% of AAOS members identifying as queer, gay, lesbian, bisexual, another identity, or not sure. Considering that more than 7% of the US population identify as LGBTQ+ [21], our findings suggest there is likely a gap in the representation of sexual and gender minorities in orthopaedic surgery. Closing this gap can continue to be a key aim for diversity, equity, and inclusion efforts in orthopaedic surgery.

Overall, the healthcare environment continues to be unwelcoming to LGBTQ+ patients and providers. Sexual and gender minorities are less likely to seek healthcare because of fear of stigmatization, with more than half of LGBTQ+ individuals experiencing discrimination in the healthcare environment [24]. One-fifth of transgender and gender-nonconforming individuals have been refused care altogether [40]. In a 1994 survey of LGBTQ+ physicians [36], 34% had reported having experienced verbal harassment from colleagues, 37% felt socially ostracized, and 88% reported having heard their medical colleagues make disparaging remarks toward LGBTQ+ patients. In a follow-up study more than a decade later [10], 15% of LGBTQ+ physicians had experienced harassment by a colleague, 22% had felt socially ostracized, 34% had witnessed discriminatory care of an LGBTQ+ patient, and 65% reported having heard derogatory comments about LGBTQ+ individuals. It is not unexpected that LGBTQ+ physicians have been found to have less career satisfaction and lower rates of retention than peers who are not in the community [30].

Our findings suggest that the current representation of sexual and gender minorities among the AAOS members who updated their membership profiles does not yet reflect the diversity of musculoskeletal patients who are treated. The standardized collection of sexual orientation and gender identity data will be important to measure progress and address ongoing diversity gaps. Furthermore, sexual minority and gender identity status should be considered in trainee recruitment efforts, and ongoing initiatives in institutions and across national orthopaedic institutions could cultivate an inclusive environment to recruit and retain LGBTQ+ professionals.

What Demographic Factors Are Associated With the Self-reporting of One’s Sexual Orientation or Gender Identity?

Of the demographic factors we investigated, perhaps the most interesting finding was that those who did not report their sexual orientation were older than those who did. This finding might be secondary to a lack of exposure and education about concepts of gender identity, sexual orientation, and pronouns among older generations of orthopaedic professionals. If a respondent is confused or unfamiliar with a question concept, they may choose not to answer these questions. Negative attitudes and prejudices are propagated when there is a lack of exposure to members of the LGBTQ+ community [29].

Simultaneously, the finding that younger responders were more likely to report their sexual orientation and gender identity suggests progress in shaping an environment where these identities can be shared and discussed. Although 72% of all trainees endorse the importance of having mentorship [34], sexual and gender minorities in healthcare often perceive it as difficult or impossible to find LGBT-identified mentors or LGBT allies [35]. The lack of LGBTQ+ mentor visibility in orthopaedics could be because of relatively few surgeons in the LGBTQ+ community but could also be because of fear of identity disclosure, bias in LGBTQ+ faculty promotion, hostility toward sexual and gender minorities, or an unsupportive institutional climate [34, 35]. In the professional setting, the openness suggested by our findings may facilitate the establishment of mentorships and community in the workplace. An institutional structure and culture supportive of sexual and gender minorities is essential to shift the culture to one where more sexual and gender minority surgeons feel comfortable about being open with their identity.

Other Considerations

There are ample reasons to pursue a more diverse and inclusive environment in orthopaedic surgery, particularly with regard to the LGBTQ+ community. A culture supportive of sexual and gender minorities demonstrates a commitment to equality, social justice, and human rights [8]. A welcoming environment will attract diverse trainees into the orthopaedic field, create a healthy learning and work environment, and improve patient care and outcomes [14, 18, 20, 41]. However, visibility of sexual and gender minority orthopaedic surgeons remains extremely limited. A 2016 study reported that orthopaedics is perceived as among the least inclusive and welcoming specialties to sexual and gender minorities [37].

As discussed, this study highlights the importance of collecting sexual orientation and gender identity data to track the progress of diversity, equity, and inclusion initiatives. However, efforts can be made to improve response rates among sexual and gender minorities in questionnaires. There is evidence that providing clear contextualization for the relevance and use of data can help build trust among sexual and gender minority responders [15, 38]. The AAOS membership questionnaire implemented a paragraph at the beginning of the questionnaire explaining the use of the data: “We strive to promote a culture of inclusivity that best serves the diversity of the AAOS community. To help inform our efforts, we are asking the following questions about gender, gender identity, sexual orientation, age, race, and ethnicity to ensure that we are advancing this goal. Your responses are anonymous and will be kept private and secure. Any data will only be reported in aggregate.” It is encouraging that a recent report found that sexual orientation nonresponse has declined as the identification as a sexual minority has increased; the authors posited that these findings suggest greater acceptability of sexual orientation assessment in questionnaires [19].

Conclusion

The proportion of self-reported LGBTQ+ AAOS members is lower than the 7% of the general US population. The greater proportion of younger AAOS members reporting this information suggests there has been progress in the pursuit of a more diverse field. The study findings support standardized collection of sexual orientation and gender identity data to better identify and address diversity gaps. As orthopaedic surgery continues to transform to reflect the diversity of musculoskeletal patients, all orthopaedic professionals (surgeons, clinicians, allied healthcare providers, and researchers), regardless of their identities, are essential in the mission to provide equitable and informed orthopaedic care. Sexual and gender minority individuals may serve as important mentors to the next generations of orthopaedic professionals; individuals from nonminority groups should be important allies in achieving this goal.