Experiences of LGBTQ+ Residents in US General Surgery Training Programs

Key Points

Question  How do rates of self-reported mistreatment and sources of mistreatment vary between lesbian, gay, bisexual, transgender, queer, and other sexual and gender minority (LGBTQ+) general surgery residents and their non-LGBTQ+ peers?

Findings  In this survey study of 6381 surgical residents, LGBTQ+ general surgery residents reported higher rates of discrimination, harassment, and bullying than their non-LGBTQ+ peers, with attending surgeons as the most common source. Despite reporting similar career satisfaction, LGBTQ+ residents were twice as likely to consider leaving their program and/or have thoughts of suicide.

Meaning  Mistreatment is a common experience for LGBTQ+ general surgery residents, and multifaceted interventions are necessary to develop safer and more inclusive learning environments.

Importance  Previous studies have shown high rates of mistreatment among US general surgery residents, leading to poor well-being. Lesbian, gay, bisexual, transgender, queer, and other sexual and gender minority (LGBTQ+) residents represent a high-risk group for mistreatment; however, their experience in general surgery programs is largely unexplored.

Objective  To determine the national prevalence of mistreatment and poor well-being for LGBTQ+ surgery residents compared with their non-LGBTQ+ peers.

Design, Setting, and Participants  A voluntary, anonymous survey adapting validated survey instruments was administered to all clinically active general surgery residents training in Accreditation Council for Graduate Medical Education–accredited general surgery programs following the 2019 American Board of Surgery In-Training Examination.

Main Outcomes and Measures  Self-reported mistreatment, sources of mistreatment, perceptions of learning environment, career satisfaction, burnout, thoughts of attrition, and suicidality. The associations between LGBTQ+ status and (1) mistreatment, (2) burnout, (3) thoughts of attrition, and (4) suicidality were examined using multivariable regression models, accounting for interactions between gender and LGBTQ+ identity.

Results  A total of 6956 clinically active residents completed the survey (85.6% response rate). Of 6381 respondents included in this analysis, 305 respondents (4.8%) identified as LGBTQ+ and 6076 (95.2%) as non-LGBTQ+. Discrimination was reported among 161 LGBTQ+ respondents (59.2%) vs 2187 non-LGBTQ+ respondents (42.3%; P < .001); sexual harassment, 131 (47.5%) vs 1551 (29.3%; P < .001); and bullying, 220 (74.8%) vs 3730 (66.9%; P = .005); attending surgeons were the most common overall source. Compared with non-LGBTQ+ men, LGBTQ+ residents were more likely to report discrimination (men: odds ratio [OR], 2.57; 95% CI, 1.78-3.72; women: OR, 25.30; 95% CI, 16.51-38.79), sexual harassment (men: OR, 2.04; 95% CI, 1.39-2.99; women: OR, 5.72; 95% CI, 4.09-8.01), and bullying (men: OR, 1.51; 95% CI, 1.07-2.12; women: OR, 2.00; 95% CI, 1.37-2.91). LGBTQ+ residents reported similar perceptions of the learning environment, career satisfaction, and burnout (OR, 1.22; 95% CI, 0.97-1.52) but had more frequent considerations of leaving their program (OR, 2.04; 95% CI, 1.52-2.74) and suicide (OR, 1.95; 95% CI, 1.26-3.04). This increased risk of suicidality was eliminated after adjusting for mistreatment (OR, 1.47; 95% CI, 0.90-2.39).

Conclusions and Relevance  Mistreatment is a common experience for LGBTQ+ surgery residents, with attending surgeons being the most common overall source. Increased suicidality among LGBTQ+ surgery residents is associated with this mistreatment. Multifaceted interventions are necessary to develop safer and more inclusive learning environments.

Lesbian, gay, bisexual, transgender, queer, and other sexual and gender minority (LGBTQ+) individuals face high rates of harassment and discrimination in the workplace.^1,2 In June 2020, the US Supreme Court ruled that LGBTQ+ employees are protected by the 1964 Civil Rights Act,³ rendering such mistreatment not only unethical but also illegal. Clinicians who identify as LGBTQ+ are increasing in visibility, and stories of their experiences in discriminatory training and working environments are emerging.⁴ However, the LGBTQ+ experience in health care has yet to be characterized at large scale.

Mistreatment contributes substantially to burnout,^5-7 which negatively impacts physicians’ personal and professional lives,⁸ patients’ health outcomes,⁹ and the health care system.¹⁰ Underrepresented populations have long been known to have higher rates of burnout, and recent evidence attributes this disparity to higher rates of mistreatment.^6,7,11,12 In surgical training, mistreatment is a common experience, particularly for underrepresented groups.^6,11-14 Thus, LGBTQ+ surgical trainees may be at particular risk of mistreatment and poor well-being. However, the existing data consist of a single survey of 44 convenience-sampled residents with limited generalizability.¹⁵

To more comprehensively characterize the experiences of LGBTQ+ surgery residents, a national survey on mistreatment, perceptions of learning environment, and other measures of wellness was administered to all residents training in accredited US general surgery residency programs. The objectives of this study were to (1) assess the national prevalence of mistreatment for LGBTQ+ surgery residents, (2) identify sources of mistreatment toward LGBTQ+ residents, and (3) compare LGBTQ+ resident perceptions of workplace environment and experiences with burnout, career satisfaction, attrition, and thoughts of suicide with their non-LGBTQ+ peers.

A voluntary survey (eAppendix in the Supplement) was offered to 8907 general surgery residents training in all Accreditation Council for Graduate Medical Education (ACGME)–accredited programs taking the 2019 American Board of Surgery In-Training Examination (ABSITE). Survey responses were deidentified by the American Board of Surgery prior to transfer to Northwestern University for analysis.^16,17 All clinically active residents were considered for analysis. Respondents were excluded if they did not have an American Board of Surgery (ABS)–provided gender or if they omitted or selected “prefer not to answer” on the LGBTQ+ identity question. The Northwestern University Institutional Review Board deemed this study exempt from review. ABSITE examinees were informed that the purpose of the survey was for research and that their responses would remain confidential. Participation implied informed consent. The survey software was constructed such that participants could exit the survey at any time. This study followed the American Association for Public Opinion Research Best Practices for Survey Research guidelines.

The post-ABSITE survey was constructed from previously validated survey instruments.^6,18-30 When no instrument was available, new items were developed. The survey was evaluated for clarity and coherence via cognitive interviews with a sample of general surgery residents from across the US, then iteratively revised through multiple rounds of pilot testing and feedback.

Gender was queried by the ABS on its registration form with the response options of male or female. A single survey item, adapted from the single-pass LGBT identity question,²² asked respondents to identify their sexual orientation (straight, lesbian or gay, bisexual, other) and gender identity (transgender, other); multiple response options could be simultaneously selected. Residents were categorized as LGBTQ+ if they self-identified as gay or lesbian, bisexual, other sexual orientation, transgender, and/or other gender identity. Residents who indicated that their sexual orientation was straight and did not identify as transgender or other gender identity were classified as non-LGBTQ+.

Respondents were queried regarding their race (American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or other Pacific Islander, White, other, prefer not to say; more than 1 response option could be selected), ethnicity (Hispanic or non-Hispanic), relationship status (married, not married and in a relationship, not married and not in a relationship, divorced, or widowed), and family status (no children and not expecting a child this academic year, 1 or more children younger than 18 years, or pregnant, adopting, or expecting a child). The ABS provided resident postgraduate year (categorized as 1, 2 or 3, or 4 or 5) as well as program characteristics, including program type (academic, community, or military), program size (total number of general surgery residents, quartiled: less than 26, 26 to 36, 37 to 50, and 51 or more), and geographic location (Northeast, Southeast, Midwest, Southwest, and West).

Items querying the learning environment were adapted from the Areas of Worklife Survey³¹ and the Voice Climate Survey.²⁴ Our conceptual model of the learning environment³² is based on previously published models.²⁸ Items were developed to query themes undescribed by prior work (eg, protected educational time), as described above. Responses for all learning environment measures were dichotomized into strongly disagree, disagree, or neutral vs strongly agree or agree.

Respondents were queried about experiences during that academic year (July 2018 to January 2019) with any of (1) 7 discriminatory behaviors they perceived to be based on gender, gender identity, or sexual orientation^{19,20,25,33-37}; (2) 6 sexual harassment behaviors^{19-21,23,25,26}; and (3) 9 bullying behaviors from the Short-Negative Acts Questionnaire,¹⁸ a previously validated instrument, with the addition of being cursed or sworn at.²⁵ All behaviors were queried in terms of frequencies; occurrences were dichotomized into never vs any.

Those responding affirmatively to any particular behavior were subsequently asked to identify a source, using the following options: (1) patients and their families, (2) attending surgeons, (3) administrators, (4) colleagues (coresidents and fellows), and (5) nurses or support staff. Those who reported experiencing a particular behavior but who did not answer the source question were classified as source not identified. Residents were considered to have experienced discrimination, sexual harassment, or bullying if they indicated ever experiencing any behavior within that subscale.

Residents were asked about their satisfaction with their decision to become a surgeon.³⁰ Burnout was defined as any symptom of emotional exhaustion or depersonalization, occurring at least weekly, from an abbreviated, modified Maslach Burnout Inventory.^6,13 Thoughts of leaving the program during the current academic year and thoughts of suicide in the past 12 months were assessed with previously published instruments.^6,27,29

Resident and program characteristics, specific types of mistreatment, and perceptions of the learning environment were compared between LGBTQ+ and non-LGBTQ+ residents using χ² tests. For each question, respondents with missing data were excluded from analyses. The most common source of each type of mistreatment was determined for non-LGBTQ+ and LGBTQ+ residents.

A multivariable logistic regression model with robust standard errors and program-level clustering was created for each type of mistreatment, adjusting for resident and program characteristics. Given prior evidence about gender differences in mistreatment experiences⁶ and the centrality of gender identity to LGBTQ+ status, an interaction term between gender and LGBTQ+ status was included in each model. Linear combinations of the coefficients were estimated to examine gender differences in the association of LGBTQ+ status and mistreatment; odds ratios (ORs) as a function of gender and LGBTQ+ status are reported. Multivariable logistic regression models with program-level clustering, adjusting for resident and program characteristics, tested whether LGBTQ+ status was associated with burnout, thoughts of attrition, and suicidality, first without and then with mistreatment as a covariate.

Analyses were performed using Stata version 14.2 (StataCorp) and R version 3.6.0 (The R Foundation). All tests were 2-sided with a predetermined significance level set at P < .05. Within mistreatment domains, P value thresholds were Bonferroni-corrected for multiple comparisons.

A total of 6956 clinically active residents responded (85.6% response rate). A total of 575 respondents were excluded who either did not have an ABS-provided gender (n = 181), declined to answer (n = 137), or selected prefer not to answer to the LGBTQ+ identity question (n = 257). Quiz Ref IDOf 6381 respondents included in the final analyses, 305 respondents (4.8%) identified as LGBTQ+, of which 191 (62.6%) identified as gay or lesbian, 98 (32.1%) as bisexual, 15 (4.9%) as other sexual orientation, and 16 (5.3%) as transgender or other gender identity; 6076 (95.2%) identified as non-LGBTQ+. There was a higher proportion of female residents among LGBTQ+ respondents (157 [51.5%]) than non-LGBTQ+ respondents (2494 [41.1%]). LGBTQ+ residents were less frequently married (64 [21.0%] vs 2666 [44.1%]) and more frequently in a nonmarriage relationship (133 [43.6%] vs 1887 [31.2%]), not in a relationship (96 [31.5%] vs 1392 [23.0%]), or divorced or widowed (12 [3.9%] vs 103 [1.7%]) than their non-LGBTQ+ counterparts (P < .001). LGBTQ+ residents less frequently had children (19 [6.2%] vs 1165 [19.2%]) or were expecting (11 [3.6%] vs 703 [11.6%]) (P < .001). More LGBTQ+ residents were training in the Northeast (109 [35.7%] vs 2007 [33.0%]) and West (60 [19.7%] vs 792 [13.0%]) than other regions (Southeast, 56 [18.4%] vs 1202 [19.8%]; Midwest, 53 [17.4%] vs 1343 [22.1%]; Southwest, 27 [8.9%] vs 732 [12.1%]) (P = .003) (Table 1).

LGBTQ+ residents reported similar perceptions of camaraderie and support within their programs’ learning environment (Table 2); Quiz Ref IDthey agreed as frequently as non-LGBTQ+ residents that their work is appreciated by their coresidents (242 [82.0%] vs 4761 [83.8%]; P = .41) and attending surgeons (205 [69.5%] vs 4201 [74.0%]; P = .09), that residents in their program cooperate with one another (256 [87.7%] vs 5007 [88.3%]; P = .74) and are among their closest friends (189 [64.1%] vs 3639 [64.2%]; P = .96), that they have a mentor within their department who cares about them and their career (194 [65.8%] vs 3818 [67.3%]; P = .59), and that favoritism determines decisions during residency (80 [27.2%] vs 1309 [23.1%]; P = .10).

However, LGBTQ+ residents more frequently reported experiencing all 3 forms of mistreatment compared with their non-LGBTQ+ peers (Table 3). Quiz Ref IDDiscrimination based on gender, gender identity, or sexual orientation was reported by 161 LGBTQ+ residents (59.2%) compared with 2187 non-LGBTQ+ respondents (42.3%; P < .001). The most common forms were being mistaken for a nonphysician (LGBTQ+, 130 [48.0%]; non-LGBTQ+, 1751 [34.1%]; P < .001), slurs or negative comments (LGBTQ+, 77 [28.6%]; non-LGBTQ+, 625 [12.1%]; P < .001), and different standards of evaluation (LGBTQ+, 67 [24.7%]; non-LGBTQ+, 1199 [23.2%]; P = .56). The sources of discrimination were similar for LGBTQ+ and non-LGBTQ+ residents; attending surgeons and coresidents were the most commonly attributed sources for all forms of discrimination, with the exception of being mistaken for a nonphysician, which was most commonly attributed to patients and their families for both LGBTQ+ and non-LGBTQ+ residents (Table 3).

Sexual harassment was reported by 131 LGBTQ+ residents (47.5%) compared with 1551 non-LGBTQ+ residents (29.3%; P < .001), with the most commonly reported behaviors being crude, demeaning, or explicit comments (117 [42.4%] vs 1378 [26.1%]; P < .001), offensive body language (46 [16.8%] vs 536 [10.2%]; P = .001), and unwanted verbal sexual attention (46 [16.7%] vs 669 [12.7%] P = .05). Additionally, LGBTQ+ residents more frequently reported being shown unwanted sexual imagery (24 [8.7%] vs 221 [4.2%]; P < .001). Quiz Ref IDAttending surgeons were the most common source of crude, demeaning, or explicit comments and offensive body language for LGBTQ+ residents, while coresidents were the most common source for non-LGBTQ+ residents. For both groups, patients and families were the most common source of unwanted verbal sexual attention (Table 3).

Bullying was reported by 220 LGBTQ+ residents (74.8%) compared with 3730 non-LGBTQ+ residents (66.9%; P = .005). The most commonly reported types included being repeatedly reminded of errors or mistakes (136 [46.4%] vs 2326 [41.9%]; P = .13), being shouted at or the target of spontaneous anger (132 [44.9%] vs 2159 [38.9%]; P = .04), and persistent criticism of work or effort (107 [36.4%] vs 1655 [29.8%] P = .02). Sources of bullying were similar for LGBTQ+ and non-LGBTQ+ residents, with most behaviors attributed to attending surgeons and coresidents (Table 3). Sources for each behavior are shown in the eTable in the Supplement. Compared with non-LGBTQ+ men, the likelihood of all types of mistreatment was higher for LGBTQ+ men (discrimination: OR, 2.57; 95% CI, 1.78-3.72; sexual harassment: OR, 2.04; 95% CI, 1.39-2.99; bullying: OR, 1.51; 95% CI, 1.07-2.12), non-LGBTQ+ women (discrimination: OR, 21.51; 95% CI, 18.50-25.02; sexual harassment: OR, 2.79; 95% CI, 2.46-3.16; bullying: OR, 1.55; 95% CI, 1.38-1.75), and LGBTQ+ women (discrimination: OR, 25.30; 95% CI, 16.51-38.79; sexual harassment: OR, 5.72; 95% CI, 4.09-8.01; bullying: OR, 2.00; 95% CI, 1.37-2.91), after adjusting for other individual and program characteristics (Table 4).

Quiz Ref IDLGBTQ+ residents reported satisfaction with their decision to become a surgeon with similar frequency as their non-LGBTQ+ peers (239 [80.7%] vs 4701 [81.5%]; P = .74) and were equally as likely to report symptoms of burnout (OR, 1.22; 95% CI, 0.97-1.52). However, LGBTQ+ residents were more likely to report considering leaving their program (61 [21.1%] vs 613 [10.9%]; OR, 2.04; 95% CI, 1.52-2.74) and having thoughts of suicide (24 [8.2%] vs 242 [4.2%]; OR, 1.95; 95% CI, 1.26-3.04). After adjusting for discrimination, sexual harassment, and bullying, LGBTQ+ residents continued to be more likely to consider leaving their program (OR, 1.79; 95% CI, 1.31-2.45) but did not demonstrate increased burnout (OR, 1.02; 95% CI, 0.81-1.29) or suicidality (OR, 1.47; 95% CI, 0.90-2.39) compared with non-LGBTQ+ residents (Table 5).

By surveying all residents training in ACGME-accredited general surgery programs and achieving a high response rate, this study constitutes, to our knowledge, the most comprehensive analysis of the workplace environment for LGBTQ+ trainees in any field of medicine. Discrimination was reported by 59.2%, harassment by 47.5%, and bullying by 74.8% of LGBTQ+ general surgery residents. We note that mistreatment of non-LGBTQ+ residents is also quite prevalent, particularly among residents identifying as female; these data are explored in more detail in other publications.^6,11,12,38 Despite experiencing higher rates of mistreatment, LGBTQ+ residents were equally likely to report satisfaction with the decision to become a surgeon and symptoms of burnout as their non-LGBTQ+ peers. They were nearly twice as likely to report considering leaving their program or having thoughts of suicide compared with their non-LGBTQ+ peers. This association with suicidality was no longer evident after adjusting for mistreatment.

To our knowledge, a single prior study of LGBTQ+ general surgery residents exists: of the 44 LGBTQ+ respondents to an online survey, 21% had been subject to targeted homophobic remarks by residents and 12% by attending surgeons.¹⁵ In another survey of 427 LGBTQ+ physicians across multiple specialties, 15% reported harassment, 22% social ostracization, 27% witnessing discrimination against an LGBTQ+ employee, and 65% having heard derogatory comments about LGBTQ+ individuals.³⁹ Both studies were limited by small sample sizes and convenience sampling methods, raising concerns about nonresponse bias and generalizability. Yet despite a sample size that more closely approximates the underlying population, we report higher rates of mistreatment than either study. Our findings more closely approximate those of the General Social Survey and other national probability samples, in which 42% of LGBTQ+ employees report discrimination based on sexual orientation and 58% have heard derogatory comments about sexual orientation and gender identity at work.²

We found that although LGBTQ+ residents were as satisfied as their non-LGBTQ+ peers with their decision to become a surgeon, they were nearly twice as likely to consider leaving their program. Thus, thoughts of attrition for LGBTQ+ residents are likely driven by environmental factors rather than a lack of individual career motivation or fulfilment. The association with attrition persisted after adjusting for mistreatment, suggesting that safe learning environments require the presence of positive as well as the absence of abusive behaviors. In other fields, the inclusiveness of a work environment (eg, as reflected by its policies) plays an integral role in the decision of LGBTQ+ individuals to remain or leave.⁴⁰ As such, the continued presence of LGBTQ+ people in a workplace and in leadership roles may serve as an indicator of inclusivity; the prior study of 44 LGBTQ+ surgery residents found 80% of them ranked programs in which they knew of an LGBTQ+ faculty member or resident more favorably.¹⁵

We also found that LGBTQ+ residents were more likely to have thoughts of suicide, but not after adjusting for mistreatment. In our prior work, we have similarly found that adjusting for mistreatment eliminates increased burnout and suicidality in vulnerable populations.⁶ Notably, bullying demonstrated a stronger association with suicidality than either discrimination (which was not associated) or sexual harassment. Therefore, addressing mistreatment in the learning environment, particularly bullying, may have a considerable impact on rates of suicide, the second leading cause of death among residents.⁴¹

Surgical educators should be aware of the unique stressors faced by their LGBTQ+ trainees. Owing to a lack of safety or support in the work or learning environment, individuals may not disclose their LGBTQ+ identities.⁴⁰ In the survey of 44 LGBTQ+ surgery residents, more than half chose not to disclose their identity out of fear of rejection or poor evaluations.¹⁵ Although not disclosing a stigmatized identity may help individuals evade mistreatment, it is a burden to maintain and has been associated with negative psychological outcomes. Nondisclosure causes cognitive preoccupation (ie, must repeatedly evaluate stressful decisions about when and to whom to disclose their identities) and hypervigilance (ie, constant fear of discovery). Additionally, substantial effort may be expended on self-monitoring and impression management (eg, modifying speech, mannerisms, gait), producing not only stress and shame from living inauthentically but also barriers to connecting with others (eg, increased response latency, decreased eye contact).⁴² Additionally, LGBTQ+ resident distress may be further compounded by a lack of social support outside of the workplace (eg, spousal, parental). Consistent with population-level data,⁴³ we found that LGBTQ+ residents were less frequently married and more frequently not in a relationship, divorced, or widowed. Relational support is fundamental to coping with adversity and pursuing opportunities for growth and development.⁴⁴

Addressing mistreatment of LGBTQ+ residents in the workplace is a critical goal, now supported by US federal law. The identification of mistreatment behaviors and their sources allows for the development and implementation of targeted interventions. Most LGBTQ+-related mistreatment was attributed to other health care professionals, indicating a need for health care professional training and effective institutional systems for reporting and addressing instances of mistreatment. However, employee training would not mitigate mistreatment from patients or their families, who constitute a substantial source of discrimination and sexual harassment. Strategies to address mistreatment must therefore be multifaceted (eg, clear institutional policies around patient codes of conduct, bystander training for health care professionals).^45,46 Recommendations to improve workplace and learning environments for LGBTQ+ students and employees in health professional schools address admissions and recruitment, resources and support, ally training, and equitable institutional policies.⁴⁷

Some limitations should be acknowledged. First, as in all cross-sectional research, association is not necessarily causation. Second, self-reported outcomes are subject to recall biases. The trauma associated with mistreatment may lead to overreporting or underreporting. Moreover, survey coadministration with a lengthy examination may result in test-related anxiety or postexamination elation.⁴⁸ However, analyses of data from prior administrations of the ABSITE survey have demonstrated that the reporting of mistreatment, burnout, and suicidality is independent of examination performance, and the association between mistreatment and burnout is independent of emotion at the time of the survey. Third, we did not inquire whether residents had disclosed their LGBTQ+ identity within their programs. Despite reassurances of confidentiality, those who had not come out may have chosen not to disclose their LGBTQ+ status or respond to any portion of this survey, thus resulting in an underestimate of the number of LGBTQ+ residents and the prevalence of LGBTQ+-related mistreatment. A subanalysis of data from respondents for whom gender was not provided, those who chose prefer not to answer to the LGBTQ+ question, and those who did not respond at all to the LGBTQ+ question was performed. Responses from this group were intermediary between the LGBTQ+ and non-LGBTQ+ respondents, thus likely representing a mixed cohort of nondisclosing LGBTQ+ respondents and non-LGBTQ+ nonrespondents. Because those who do not disclose their identity are likely to have worse psychological outcomes,⁴² we may have underestimated the prevalence of thoughts of attrition and suicide among LGBTQ+ individuals. Fourth, we used the single-pass question for LGBTQ+ identity,²² which has precedent but may have prevented us from identifying fully transitioned individuals who do not identify as transgender or other gender identity. Fifth, we inquired about discrimination based on gender, sexual orientation, and gender identity. Although the nature of this question makes it difficult to disentangle which aspect of identity was targeted, these identities intersect in complex ways that may not be cleanly separable.

Mistreatment is pervasive and impacts well-being among residents identifying as LGBTQ+ in general surgery. This work adds to the mounting evidence and rationale for addressing mistreatment, a necessary step toward creating learning environments that are capable of supporting the educational growth and professional identity formation of the diverse trainee cohorts required to represent and meet the needs of the communities we serve.

Accepted for Publication: August 10, 2021.

Published Online: October 20, 2021. doi:10.1001/jamasurg.2021.5246

Corresponding Author: Yue-Yung Hu, MD, MPH, Surgical Outcomes and Quality Improvement Center (SOQIC), Feinberg School of Medicine, Northwestern University, 633 N St Clair St, 20th Floor, Chicago, IL 60611 (yue-yung.hu@northwestern.edu).

Author Contributions: Dr Hu had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Mr Heiderscheit and Dr Schlick contributed equally to this work.

Study concept and design: Heiderscheit, Ellis, Irizarry, Eng, Hoyt, Bilimoria, Hu.

Acquisition, analysis, or interpretation of data: Heiderscheit, Schlick, Ellis, Cheung, Amortegui, Eng, Sosa, Hoyt, Buyske, Nasca, Bilimoria, Hu.

Drafting of the manuscript: Heiderscheit, Hoyt, Hu.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Heiderscheit, Schlick, Eng, Bilimoria, Hu.

Obtained funding: Nasca, Bilimoria, Hu.

Administrative, technical, or material support: Ellis, Amortegui, Hoyt, Buyske, Nasca, Hu.

Study supervision: Ellis, Cheung, Sosa, Hoyt, Bilimoria, Hu.

Conflict of Interest Disclosures: Dr Sosa has received institutional research funding from Exelixis and Eli Lilly and is a member of the data monitoring committee of the Medullary Thyroid Cancer Consortium Registry, which is supported by GlaxoSmithKline, NovoNordisk, AstraZeneca, and Eli Lilly. Drs Bilimoria and Hu have received grants for serving as co–principal investigators of the SECOND trial from the American College of Surgeons and the Accreditation Council for Graduate Medical Education as well as nonfinancial support from the American Board of Surgery. No other disclosures were reported.

Funding/Support: Funding for this work was provided by the Accreditation Council for Graduate Medical Education and American College of Surgeons. The American Board of Surgery provided in-kind support. Dr Ellis was supported by a postdoctoral research fellowship from the Agency for Healthcare Research and Quality (grant 5T32HS000078).

Role of the Funder/Sponsor: The funding agencies had no role in the design and conduct of the study; analysis of data; preparation or review of the manuscript; and decision to submit the manuscript for publication. Authors affiliated with funding agencies were involved in the collection and interpretation of the data and review and approval of the manuscript.

Disclaimer: The views expressed in this work represent those of the authors only.

Meeting Information: This paper was presented at the Academic Surgical Congress; February 4 and 6, 2020; Orlando, Florida.