Sociodemographics of Patient Populations Undergoing Gender-Affirming Surgery: A Systematic Review of All Cohort Studies

Abstract

Importance:

Sociodemographic and health characteristics of patients undergoing gender-affirming surgery (GAS) are currently unknown. Understanding these patient characteristics is vital to optimizing patient-centered care for transgender patients.

Objective:

To determine sociodemographic characteristics for the transgender population undergoing GAS. Collected sociodemographic information included the following: age, race/ethnicity, body metrics, hormone replacement therapy administration and duration, substance use, psychiatric comorbidities, and medical comorbidities.

Evidence Review:

A search of seven electronic databases (PubMed, PsycINFO, Embase, CINAHL, Web of Science, Cochrane, and Gender Studies) was used to find all articles on GAS from inception through May 2019. The 15,190 articles were then subjected to two levels of screening, and articles unrelated to gender-affirming care, unavailable in English, n<5, and with no outcomes reporting were excluded. Textbook chapters and letters were also excluded.

Findings:

A total of 406 studies were fully extracted, with 307 studies reporting age (n=22,727 patients), 19 reporting race/ethnicity (n=1184), 74 reporting body metrics (body mass index [BMI] n=6852, height n=416, and weight n=475), 58 reporting hormone therapies (n=5104), 56 reporting substance use (n=1146), 44 reporting psychiatric comorbidities (n=574), and 47 reporting medical comorbidities (n=573). From the 406 studies, 80 were done in the United States. Regarding U.S. studies, 59 studies reported age (n=5365), 10 reported race/ethnicity (n=709), 22 reported body metrics (BMI n=2519), 18 reported hormone therapies (n=3285), 15 reported substance use (n=478), 44 reported psychiatric comorbidities (n=394), and 47 reported medical comorbidities (n=293). Age was the most reported characteristic, reported in 75.62% of studies (73.75% of U.S. studies). Race/ethnicity was the least commonly reported data, reported in 4.68% of studies (12.50% of U.S. studies).

Conclusions and Relevance:

The type of sociodemographic information reported by GAS studies is inconsistently reported. To improve patient-centered care for transgender patients, further work is needed to create a standardization of collected sociodemographic information.

Introduction

People who are transgender have a gender identity that differs from their sex assigned at birth. Many transgender or nonbinary (TGNB) persons may undergo hormone replacement therapy (HRT) or gender-affirming surgeries (GAS) to mediate the dysphoria caused by this incongruence. The volume of GAS annually has increased dramatically in the last few years, with the American Society of Plastic Surgeons (ASPS) reporting a 155% increase in GAS from 2016 to 2017.¹ The number of patients seeking GAS increased over 200% from 2012 to 2014 due to expanding coverage of Medicaid and Medicare.²

In surgical literature, the type of demographic information reported varies greatly and often depends on the type of surgery (e.g., weight loss and breast procedures are more likely to report body mass index [BMI]). Similarly, there is a lack of standardization in reporting sociodemographic information in GAS articles. Indeed, very little is known about the sociodemographics of TGNB populations overall, with only a few studies having made a significant, concerted effort to collect such information.^3–5

There are several persistent obstacles to collection or standardization of these data, especially on a global scale. Different countries and cultures recognize and process gender changes in differing manners and, on an even more basic level, many TGNB individuals are reluctant to self-report in governmental surveys due to societal discrimination and risk of violence.⁶

The transgender population undergoing GAS is undoubtedly very different sociodemographically from the overall transgender population, due to financial barriers and other considerations. According to the U.S. Transgender Survey, only 25% of transgender people have undergone any form of GAS, with 42% of trans men, 28% of trans women, and 9% of nonbinary people having undergone procedures. Regardless, there is high interest in GAS; for example, 97% of transgender men surveyed had either had (36%) or wanted to have (61%) chest masculinizing mastectomy.⁴ As more multidisciplinary transgender health centers form, it will be critical to appropriately identify and gather sociodemographics of patient populations to identify needs and move toward more equitable access to care.

To date, no study has amassed and analyzed sociodemographic information across transgender surgical patients in the literature. The purpose of this study is to systematically assess the literature for all studies concerning GAS and evaluate sociodemographic information related to the following: age, ethnicity, height, weight, BMI, psychosocial and medical comorbidities, and HRT status.

Materials and Methods

Literature search

A comprehensive search strategy was developed in consultation with a medical informationist. The primary aim of the search was to conduct a systematic review on GAS procedures and outcomes; primary results are reported elsewhere (Oles et al., 2022a,b in Annals of Surgery).^7,8 This article reports the results of the secondary aim, to assess the GAS literature for sociodemographic characteristics. Methods are fully reported in the Supplementary Data S1.

Briefly, seven electronic databases (PubMed, PsycINFO, Embase, CINAHL, Web of Science, Cochrane, and Gender Studies) were used to identify studies on GAS from inception through May 2019. Procedures included genital reconstruction surgeries, top surgeries, hysterectomy, vaginectomy, orchiectomy, facial feminization or masculinization surgeries, reduction thyrochondroplasty, vocal cord surgeries, and any combination of the aforementioned surgeries in line with gender affirmation. There was no exclusion based on age group or country of origin, other than excluding articles that were unavailable in English.

Selection criteria

Two independent reviewers used DistillerSR to select articles at the title, abstract, and full-text stages. Types of studies excluded were case studies (k<5), non-English studies without an available English translation, and textbook chapters. Articles that did not specify sample size or surgical procedures were excluded.

Data extraction and analysis

Two reviewers extracted sociodemographic information (e.g., age, race, and height) and medical and psychological conditions and comorbidities (e.g., diabetes and depression) from each study onto a standardized template. Identification numbers were assigned to each patient cohort, and duplicate cohorts were accounted for during calculations. Extracted sociodemographic information was divided into physical patient characteristics, substance use, medical comorbidities (psychosocial and medical), and information on HRT.

Information from U.S. studies was also extracted separately so that results could be compared to those of studies that have looked at the demographics of the overall U.S. transgender population.^2–5 Please see Supplementary Data S2 for the full Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines checklist and Supplementary Figure S1.

Ethical review

No human subject data were collected for this review, so it was exempt from institutional review board review.

Results

The multi-database search strategy initially yielded 15,186 studies, which had been de-duplicated by a medical librarian. After title and abstract screening, there were 1826 for full-text review. From these articles, 1420 articles were removed because they did not meet eligibility criteria (i.e., were nonsurgical gender-affirming care). Four studies were manually added, leading to a total of 406 articles relevant to GAS, including information on patient sociodemographics.

Vaginoplasty was the most common surgical type in the literature, with 171 articles. The other surgeries were less represented in the literature, with 82 phalloplasty, 16 metoidioplasty, 23 hysterectomy/vaginectomy, 35 mastectomy, 6 mammoplasty, 21 facial feminization, 31 voice and/or cartilage, and 21 multiple surgeries together.

Sociodemographic information reported across studies was extremely variable. Please see Supplementary Table S1 for sociodemographic variables stratified by surgical type. The most commonly reported demographic variable was patient age, which was reported by 75.62% of studies. Data collected on other demographic factors were heterogenous: race/ethnicity was reported in only 4.68% of studies, body metrics in 18.23%, HRT administration and duration in 14.29%, substance use in 13.79%, psychiatric comorbidities in 10.84%, and medical comorbidities in 11.58% (Table 1). Please refer to Supplementary Table S2 to view all included studies.

Table 1.

Summary Breakdown of Reported Sociodemographic Data

Total 406 publications, 80 from the United States
	k of articles	%
Age data, ALL	307	75.62
Age data, United States	50	73.75
Race/ethnicity data, ALL	19	4.68
Race/ethnicity data, United States	10	12.50
Body metrics, ALL	74	18.23
Body metrics, United States	22	27.50
Hormone therapies, ALL	58	14.29
Hormone therapies, United States	18	22.50
Substance use, ALL	56	13.79
Substance use, United States	15	18.75
Psychiatric comorbidities, ALL	44	10.84
Psychiatric comorbidities, United States	15	18.75
Sexual function, ALL	13	3.20
Sexual function, United States	3	3.75
Medical comorbidities, ALL	47	11.58
Medical comorbidities, United States	15	18.75

Age

The 307 studies (75.62%) that collected information on patient age reported an overall mean age of 33.17 (standard deviation [SD] 8.21), and median age of 33.05. The youngest patient was 14 years old (mastectomy) and the oldest was 78 years old (vaginoplasty). The 59 U.S. articles that reported age had a mean patient age of 35.60 (SD 10.40) and median age of 30.22.

In terms of age differences by surgical type, patients who underwent vocal and facial feminization surgeries were on average significantly older, with a mean age of 39.56 and 38.67 years, respectively. Patients who underwent other surgeries were younger: vaginoplasty (34.87 years), phalloplasty (32.95 years), metoidioplasty (31.20 years), hysterectomy/vaginectomy (31.73 years), and breast augmentation (33.99 years). Patients who underwent mastectomy were the youngest, with a mean age of 28.10 years. Median age was much less likely to be reported. Mean and median ages are shown in Table 2.

Table 2.

Patient Age Stratified by Type of Surgery

Type of surgery	Mean age (years)	Median age (years)	Age range (years)
Vaginoplasty	34.87	34	17–78
Phalloplasty	32.95	36	17–72
Metoidioplasty	31.20	30.5	18–62
Ovariohysterectomy/vaginectomy	31.73	31.4	18–67
Facial feminization	38.67	25a	16–70
Voice	39.56	40	16–68
Mastectomy	28.10	28	14–71
Mammoplasty	33.99	b	17–76

^a Note that only 1 study reported age as a median.

^b Note that no study reported age as a median.

Race/ethnicity

The 19 (4.68%) studies reporting information on race/ethnicity sampled a total of 1184 patients. From that pool, the ethnic/racial breakdown was 56.17% White, 7.35% Black, 24.32% Asian, 1.86% Latinx, and 10.30% other or multiple races/ethnicities. Only 3.62% of the international articles reported patient race and/or ethnicity. The 10 U.S. studies that sampled a total of 709 patients had an ethnic/racial breakdown of 65.44% White, 11.85% Black, 2.82% Asian, 3.10% Latinx, and 16.78% other.

Body metrics

The 74 (18.23%) articles that collected information on body metrics most often collected information on BMI, with 70 articles reporting that information. The mean and median BMI were 25.08 and 24.55, respectively (SD 3.80). The mean BMI of patients who underwent top surgery was 25.99, and the mean BMI of patients who underwent bottom surgeries was comparable at 25.86. Only 9 articles reported height alone and 10 articles reported weight alone. The mean and median heights were 169.24 cm and 170.90 cm, respectively (SD 5.70 cm). The mean weight was 70.37 kg, and the median weight was 72.50 kg (SD 10.70 kg). BMI was most reported for mastectomy, vaginoplasty, and hysterectomy/vaginectomy (Supplementary Table S3).

HRT administration and duration

Fifty-eight (14.29%) studies reported information on hormone therapy. The patients receiving hormone therapy had done so for a mean of 47.47 months and a median of 37.04 months (SD 29.33 months). Information on HRT was more likely to be reported in articles related to breast procedures, such as augmentation (83.33% reported) or chest masculinization (51.43% reported). Mean HRT duration (in months) also varied depending on surgical procedure: metoidioplasty (43.64 months), hysterectomy/vaginectomy (36.02), mastectomy (15.84), breast augmentation (57.86), and multiple surgeries (48.65). None of the vaginoplasty nor phalloplasty article reported HRT duration.

Substance use

Only 56 (13.79%) articles reported information on any substance use. Information on current tobacco use was most likely to have been collected (6.77%). The percentages of studies that collected information on former tobacco use (1.69%), alcohol use (0.71%), marijuana use (0.29%), and other/unspecified substance use (0.99%) were much smaller.

Psychiatric comorbidities

Information on psychiatric comorbidities was reported by 44 (10.84%) studies. Of these studies, 34.09% did not specify psychiatric condition. In terms of reported psychiatric condition, 40.91% of these articles reported on depression, but information on anxiety (6.82%), phobia (4.54%), suicidal thoughts/ideation (9.09%), suicide attempts (13.64%), and completed suicide (during follow-up after surgery; 9.09%) was much less likely to have been reported.

Medical comorbidities

Information on medical comorbidities was reported by 47 articles (11.58%). These studies most commonly reported information on diabetes (42.55%), hypertension (27.79%), asthma (19.14%), hypothyroidism (11.90%), and unspecified cardiovascular problems (10.64%). Mastectomy articles were most likely to collect information on medical comorbidities, with 31.43% of those studies including such information.

Gender

Most (98.28%) of the studies reported information on patient gender, which was invariably reported as male-to-female (MTF) or female-to-male (FTM). Seven (1.72%) studies did not report any information on patient gender. MTF patients were included in 234 (56.64%) studies, and FTM patients were included in 187 (46.06%) studies. No study reported nonbinary patient inclusion.

Discussion

This review summarizes reporting of sociodemographic variables in published article of GAS, and found there is no standardization in the presentation of sociodemographic data across studies. This is not something that is unique to GAS and the types of sociodemographic information collected in studies vary broadly—some less commonly reported variables that we did not extract include marital status, highest level of education, employment status, and income.

In the 2019 Plastic Surgery Statistics Report, the ASPS collected demographic information on patient sex (male or female only), age (13–19, 20–29, 30–39, 40–54, and 55+), and ethnicity (Caucasian, Hispanic, African American, and Asian American).⁹ While there is no standardization in type of information collected, these three variables could serve as a baseline for studies across all plastic surgery.

Age

The mean age of 35.60 (SD 10.40) and median age of 30.22 of patients in U.S. studies are comparable to the average reported ages collected by the U.S. Transgender Survey, which found that 75% of transgender persons have their first surgical transition-related procedure after age 25, with 43% having their first procedure after age 34.⁴ Given that some studies reported patients having had previous GAS, it is likely that the average age of first procedure for the U.S. patients across these studies was significantly younger than that reported by the U.S. Transgender Survey, possibly reflecting the difficulty in access to GAS among the general U.S. transgender population. As transgender awareness and acceptance grow, and access to GAS expands through insurance, the mean age is likely to decrease.

Race/ethnicity

White people are overrepresented in the GAS studies in the literature and Black and Latinx people are underrepresented when compared to the U.S. transgender population, which has an estimated breakdown of 55% White, 16% Black, 21% Latinx, and 8% other—although these numbers also likely underestimate people of color.^3,10–12 With the vast majority of studies not collecting information on patient race/ethnicity, it is difficult to say how truly underrepresented minority populations are in GAS studies. It is vital that researchers report this information so that accessibility to GAS by race/ethnicity may be more accurately assessed in the future.

Regarding international articles, there is a much higher representation of Asian patients, given the significant research output on GAS from several Asian countries. It is also important to note that this article defined racial/ethnic breakdowns in the same way as it is typically done in the United States (White, Black, Latinx, Asian, and other), which is not reflective of schema used in other countries, for example, the Brazilian government typically breaks down racial/ethnic groups into White, Black, Brown, Yellow, and Indigenous.^13,14 However, due to the fact that so few international articles reported on race/ethnicity, with most from European or Asian countries with relatively homogenous populations, we did not come across other categorizations of racial or ethnic classifications in the search.

Body metrics

The articles that reported information on BMI often did so in the context of using BMI as a risk factor for surgical procedures. The role of BMI in complications is controversial, with some single-center series reporting it is not a predictor.^15,16 while other data suggest poor healing issues, surgical site infection, and flap necrosis are associated with greater BMI.^17,18 Future multicenter research is needed in this area to study the impact of BMI cutoff requirements for GAS.

HRT duration/administration

The variability in reporting HRT duration and administration is important to note, considering that hormones are often discontinued before GAS to decrease the risk of thrombosis and other complications. in addition, there is inconsistent information about risks associated with estrogen use before GAS.¹⁹ Given the active research in this field, it is crucial to note in future studies whether patients continue or discontinue hormones before GAS, and what outcomes occur when (dis)continuing, including any possible negative impact on mental health of patients who are asked to discontinue.

Substance use

There are significantly higher rates of smoking in the transgender community than in the cisgender community, with transgender adults being 2.1 times more likely to smoke than cisgender adults.²⁰ These high rates of smoking are challenging due to the increased surgical risk of venous thrombosis, impaired wound healing, and flap necrosis.^21–23 More studies should report proportions of smoking and other substance use in their surgical patients, as well as smoking cessation and duration as requirements for surgery and associated outcomes to gain a better understanding of the impacts of smoking on GAS outcomes.

Psychiatric comorbidities

Transgender patients are at increased risk for psychological distress and psychiatric comorbidities.⁴ In countries with universal health coverage, it has been shown that utilization of mental health treatment decreases in transgender persons after GAS, likely reflecting an improvement in mental health after surgery.²⁴ This decrease may also be somewhat influenced by requirements for letters of support from mental health providers before surgery.

Medical comorbidities

With around 56% of transgender patients reporting mistreatment from a health care provider or being afraid to seek health care due to fear of mistreatment, transgender patients are often at a disadvantage when it comes to health care.³ Over 30% of patients did not seek needed medical care due to past incidents of discrimination within a health care setting, according to a 2007 study.²⁵ This can lead to a higher incidence of chronic comorbidities that are not medically managed.²⁶ Reporting information on patients' medical comorbidities could be an important step in improving knowledge of medical comorbidities and ways in which these could affect GAS outcomes in the transgender community.

Gender

All studies reported information on gender in a binary manner, with no study mentioning nonbinary patients. Instead, studies reported patient gender as MTF/FTM or trans female/trans male in accordance with procedure, for example, trans patients undergoing phalloplasty were typically reported as FTM. This is in contrast to patient-reported gender identity: only 62% of the U.S. trans community identifies as trans man (29%) or trans woman (33%), and up to 35% of the community identifies as nonbinary or another gender.⁴ Regardless, it is common to collect information on male versus female breakdown of surgical procedure, with the ASPS reporting distributions across all procedures.⁷

In addition, many communities and countries around the world have significant representation of members of a third gender, including, but not limited to two-spirit, hijra, we'wha, and berdache.^11,27,28 Internationally, quantifying this population of individuals is difficult; best estimates state that 25 million people around the world are transgender or gender diverse.²⁹

Another report estimated the worldwide transgender population as 9.2 per 100,000, but only 2.5 per 100,000 with medical diagnoses in line with gender dysphoria—this leaves an estimate of only 736,000 transgender individuals worldwide based on those who sought gender-affirming therapy, which is not inclusive information.³⁰ The United Kingdom Government Equities Office states they have no definitive data on this—their best estimates published in 2018 offer a number of 200,00–500,000, and their Office for National Statistics is researching how best to quantify this population moving forward.³¹

A subsequent report published in 2019 counted 14,053 transgender and gender nonconforming out of 108,100 respondents to a self-reporting survey from a National lesbian, gay, bisexual, and transgender survey from the same office.³² Other countries throughout Europe and Asia do not have estimates or numbers that have been validated or published. It is estimated that 490,000 hijra live in India, based on estimates from 2014.³³ Searches for population data in South America were not conclusive and could not be found; the same was true for data for the Asian transgender community. In Canada, data from TransPulse, a self-reporting survey in Canada, estimates 1 in 200 people are transgender or gender diverse.³⁴

To establish an equivalency in GAS, there is a need to rethink classifications of gender to be more inclusive. Current best practice for appropriately capturing gender identity is to use a two-step approach, asking (1) sex assigned at birth and (2) current gender identity.²⁹

This study has limitations. Most included studies (171) reported on vaginoplasty and are from the United States, so publication bias is likely and demographics cannot be assumed to be representative of all GAS patients. Furthermore, access to and desire for GAS are limited to a subset of the TGNB population. As such, these summary results do not reflect overall TGNB population demographics, as seen in the comparison to the national U.S. transgender survey.

Conclusions

There is a lack of standardization in the way in which sociodemographic information across surgical articles is currently being collected and GAS studies are no exception. With articles related to transgender care, it is important to collect information on gender in a mindful, inclusive way and to be cognizant of patient demographic characteristics and ways in which results may not be generalizable to the overall TGNB population. To improve evidence-based gender-affirming care, there is a need to standardize the collection of patient sociodemographic information collected moving forward.

Abbreviations Used

ASPS

American Society of Plastic Surgeons

BMI

body mass index

FTM

female-to-male

GAS

gender-affirming surgery

HRT

hormone replacement therapy

MTF

male-to-female

SD

standard deviation

TGNB

transgender or nonbinary

Authors' Contributions

N.O., C.T., and D.C. conceived of and planned the review. N.O. and C.T. wrote the protocol. M.F. wrote the article with direction from N.O. and S.W.B. M.N. and S.W.B. assisted with editing and contributed to the draft. N.O. and M.F. wrote the search terms and ran the searches. N.O. and M.F. screened titles and abstracts. N.O. and M.F. reviewed full texts. J.W. abstracted data with direction from S.W.B.

Author Disclosure Statement

None of the authors has any disclosure, financial or otherwise, relating to the contents of this article.

Funding Information

There is no funding to report for this submission.

Supplementary Material

Supplementary Data S1

Supplementary Data S2

Supplementary Figure S1

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Cite this article as: Fazal M, Oles N, Beckham SW, Wang J, Noyes M, Twose C, Coon D (2023) Sociodemographics of patient populations undergoing gender-affirming surgery: a systematic review of all cohort studies, Transgender Health 8:3, 213–219, DOI: 10.1089/trgh.2021.0111.