Patient and Surgery Characteristics of Inpatient Hysterectomies Among Transgender Individuals

Abstract

Purpose:

The purpose of this study is to estimate population-based rates of inpatient hysterectomy and accompanying bilateral salpingo-oophorectomy by indication and evaluate surgical patient characteristics by indication, year, patient age, and hospital location.

Methods:

We used 2016 and 2017 cross-sectional data from the Nationwide Inpatient Sample to estimate the hysterectomy rate for individuals aged 18–54 years with a primary indication for gender-affirming care (GAC) compared to other indications. Outcome measures were population-based rates for inpatient hysterectomy and bilateral salpingo-oophorectomy by indication.

Results:

The population-based rate of inpatient hysterectomy for GAC per 100,000 was 0.05 (95% confidence interval [CI] = 0.02–0.09) in 2016 and 0.09 (95% CI = 0.03–0.15) in 2017. For comparison, the rates per 100,000 for fibroids were 85.76 in 2016 and 73.25 in 2017. Rates of bilateral salpingo-oophorectomy in the setting of hysterectomy were higher in the GAC group (86.4%) than in comparison groups (22.7%–44.1% for all other benign indications, 77.4% for cancer) across all age ranges. A higher rate of hysterectomies performed for GAC was done laparoscopically or robotically (63.6%) than other indications, and none was done vaginally, as opposed to comparison groups (0.7%–9.8%).

Conclusion:

The population-based rate for GAC was higher in 2017 compared to 2016 and low compared to other hysterectomy indications. Rates of concomitant bilateral salpingo-oophorectomy were more prevalent for GAC than for other indications at similar ages. The patients in the GAC group tended to be younger, insured, and most procedures occurred in the Northeast (45.5%) and West (36.4%).

Introduction

Gender-affirming surgery has helped transgender and gender nonconforming individuals alleviate gender dysphoria, which is defined as distress caused by an incongruity between one's gender identity and sex assigned at birth.¹ The utilization of gender-affirming surgery has markedly increased in the past decade, as it became covered by the Centers for Medicare and Medicaid Services and other health insurance companies in 2014.^1–4 For transgender and gender nonconforming individuals assigned female at birth, the most common gender-affirming surgeries are chest surgery (bilateral mastectomy or breast reduction) and hysterectomy.^2,3

However, data on indications for hysterectomy alone or hysterectomy with bilateral salpingo-oophorectomy for transgender and gender nonconforming patients are lacking. Within the National Inpatient Sample, the recommended two-step method of self-reported gender identity is not consistently implemented.⁵ Transgender national surveys have assessed the prevalence of hysterectomy in transgender and gender nonconforming individuals,^6,7 but not indications or surgical characteristics.⁷ Past research that utilized the National Inpatient Sample to identify trends of gender-affirming surgery in the United States focused on transgender and gender nonconforming mental health disorders and only documented up until 2014.¹

Hysterectomy with bilateral salpingo-oophorectomy is an area of interest for those considering fertility preservation and the potential health benefits of ovarian retention.⁸ The rate and clinical contexts of bilateral salpingo-oophorectomy in the transgender and gender nonconforming population have not been adequately described, which limits research on the long-term outcomes of transgender and gender nonconforming individuals who do or do not undergo ovarian removal.

The objectives of this study are to utilize the National Inpatient Sample to estimate population-based rates of inpatient hysterectomy and accompanying bilateral salpingo-oophorectomy by indication and evaluate patient and surgical characteristics of these surgeries, including year, patient age, and geographic hospital location, during a time of expanding insurance coverage but also U.S. political backlash.

Methods

Data source

We used cross-sectional data from the 2016 and 2017 National Inpatient Sample. The National Inpatient Sample database is maintained by the Agency for Healthcare Research and Quality's Heathcare Cost and Utilization Project. As of 2012, the National Inpatient Sample approximates a 20% stratified sample of all discharges from United States community hospitals, comprising more than 7 million inpatient records per year. National Inpatient Sample data include basic demographics, International Classification of Diseases (ICD)-10 diagnoses, procedural coding, payment source and financial data, and hospital characteristics.

The National Inpatient Sample is sampled from the State Inpatient Databases, which includes inpatient data currently contributed to Healthcare Research and Quality's Heathcare Cost and Utilization Project, covering more than 97% of the U.S. population. National Inpatient Sample sampling methods have previously been described by Yi et al.,⁹ and detailed information on the design is available online.¹⁰ The unit of analysis was the discharge or encounter rather than the individual. As hysterectomy should only occur once per patient, we assumed that each discharge or encounter appeared once per patient. A unique hospital identifier allows linkage of discharge data to National Inpatient Sample data with hospital geographic characteristics.

Studies using the National Inpatient Sample database are considered minimal risk and thus exempt by the Duke University Institutional Review Board.

Outcome measures

We obtained estimates of the number of hysterectomies performed for individuals aged 18–54 years with a primary indication of gender identity disorder or gender dysphoria (referred to as gender-affirming care [GAC] hereafter as the aforementioned terminology used for the ICD codes during our study time period is now considered outdated) for each included study year compared to other common indications (Supplementary Table S1).¹¹ The comparison indications were categorized as fibroids, endometriosis, abnormal uterine bleeding (AUB), cancer, pelvic pain, and “other benign.” “Other benign” was defined as all other indications not covered by the above included hysterectomy indications.

Primary outcome measures consisted of population-based rates for inpatient hysterectomy by indication and year and population-based rates for bilateral salpingo-oophorectomy combining both years of data. Secondary outcome measures were patient and surgery characteristics of surgeries by indication and age. The primary comparisons of interest were population-based rates of inpatient hysterectomy and patient and surgery characteristics for GAC versus other indications.

Patient, surgical, and hospital geographic characteristics were described by indication for hysterectomy. Continuous variables were summarized with mean (standard error), median (interquartile range), and range. Categorical variables were reported as frequency and percent. Surgical approach and bilateral salpingo-oophorectomy rates by indication within the age groups 18–24, 25–34, 35–44, and 45–54 years were also described. Patient race and ethnicity are listed as designated in medical records. Supplementary Table S2 provides the amount of missingness for each hysterectomy indication.

Statistical analyses

National Inpatient Sample data were used to calculate the number of patients that had a hysterectomy due to each indication and estimate the standard errors. A count of the number of women aged 18–54 in the United States in 2016 and 2017 was taken from the U.S. Census Bureau.^12,13 To assess the population-based rates of inpatient hysterectomy, U.S. census data on the total number of women in the appropriate age group were used as a denominator for the rates, as this is the best approximation of the pool of individuals eligible for hysterectomy, despite the fact that many assigned female at birth transgender and gender nonconforming individuals do not identify as women and may not be listed as women in U.S. census data.

The population-based rates of inpatient hysterectomy for each indication per 100,000 population in 2016 or 2017 were then calculated using the following formula:

$$\frac{NumberofNationalInpatientSamplehysterectomiesbyindication,weighted}{PopulationofU.S.womenin2016or2017}\ast 100,000$$

The corresponding 95% confidence interval (CI) was calculated for each indication and year. All statistical analyses were completed in SAS 9.4 (SAS Institute, Cary, NC).

Results

Population based rates of inpatient hysterectomy by indication and year

The population-based rate of inpatient hysterectomy per 100,000 women aged 18–54 in the United States for the primary indication of GAC was 0.05 (95% CI = 0.02–0.09) in 2016 and 0.09 (95% CI = 0.03–0.15) in 2017. For comparison, the hysterectomy rate for a primary indication of fibroids was 85.76 (95% CI = 82.26–89.25) per 100,000 women in 2016 and 73.25 (95% CI = 70.10–76.40) per 100,000 women in 2017. There was a decrease in the rate of hysterectomy for all indications other than GAC. In both years, fibroids were the most common indication for hysterectomy, followed by “other benign,” AUB, endometriosis, cancer, pelvic pain, and finally GAC (Table 1).

Table 1.

Population-Based Rates of Inpatient Hysterectomy by Indication and Year Using the National Inpatient Sample, 2016–2017

Indication	2016		2017
	Weighted frequency	Population-based rate (95% CI) per 100,000 population	Weighted frequency	Population-based rate (95% CI) per 100,000 population
Gender-affirming care	40	0.05 (0.02–0.09)	70	0.09 (0.03–0.15)
Fibroids	67,445	85.76 (82.26–89.25)	57,470	73.25 (70.10–76.40)
“Other benign”	37,145	47.23 (45.28–49.18)	31,345	39.95 (38.23–41.68)
AUB	15,630	19.87 (18.68–21.07)	11,815	15.06 (14.15–15.97)
Endometriosis	10,000	12.72 (11.91–13.52)	8330	10.62 (9.90–11.33)
Cancer	8940	11.37 (10.43–12.30)	7795	9.94 (9.12–10.75)
Pelvic pain	4255	5.41 (4.95–5.87)	2910	3.71 (3.35–4.06)

AUB, abnormal uterine bleeding; CI, confidence interval.

Rate of bilateral salpingo-oophorectomy and surgical characteristics by age and indication

Bilateral salpingo-oophorectomy generally became more frequent as age increased for all indications, other than for the GAC group where bilateral salpingo-oophorectomy was likely to occur with hysterectomy even at young ages. Rates of bilateral salpingo-oophorectomy in the setting of hysterectomy were notably higher in the GAC group (86.4%) than in all other indication groups (22.7%–44.1% for all other benign indications, 77.4% for cancer) across all age ranges. A higher rate of hysterectomies performed for GAC was done laparoscopically or robotically (63.6%) than for other indications, and none was done vaginally for the GAC group, as opposed to a range of 0.7% to 9.8% for all other groups (Table 2).

Table 2.

Surgical Characteristics by Indication for Hysterectomy and Age Group (Weighted) Using the National Inpatient Sample, 2016–2017

Characteristic	Gender-affirming care	Fibroids	“Other benign”	AUB	Endometriosis	Cancer	Pelvic pain	Total
All ages (N = 263,190)
N (weighted)	110	124,915	68,490	27,445	18,330	16,735	7165	263,190
Surgical approach, n (%)
Abdominal	40 (36.4)	105,330 (84.3)	45,605 (66.6)	17,995 (65.6)	12,530 (68.4)	12,175 (72.8)	4440 (62.0)	198,115 (75.3)
Robotic/laparoscopic	70 (63.6)	16,850 (13.5)	16,145 (23.6)	7270 (26.5)	5115 (27.9)	4450 (26.6)	2340 (32.7)	52,240 (19.8)
Vaginal	0 (0.0)	2735 (2.2)	6740 (9.8)	2180 (7.9)	685 (3.7)	110 (0.7)	385 (5.4)	12,835 (4.9)
Bilateral salpingo-oophorectomy, n (%)	95 (86.4)	31,290 (25.0)	24,850 (36.3)	6225 (22.7)	8080 (44.1)	12,945 (77.4)	1985 (27.7)	85,470 (32.5)
18–24 years old (N = 1575)
N	10	30	1040	175	180	50	90	1575
Surgical approach, n (%)
Abdominal	5 (50.0)	30 (100.0)	980 (94.2)	135 (77.1)	120 (66.7)	25 (50.0)	65 (72.2)	1360 (86.3)
Robotic/laparoscopic	5 (50.0)	0 (0.0)	50 (4.8)	25 (14.3)	55 (30.6)	20 (40.0)	25 (27.8)	180 (11.4)
Vaginal	0 (0.0)	0 (0.0)	10 (1.0)	15 (8.6)	5 (2.8)	5 (10.0)	0 (0.0)	35 (2.2)
Bilateral salpingo-oophorectomy, n (%)	10 (100.0)	5 (16.7)	60 (5.8)	5 (2.9)	55 (30.6)	20 (40.0)	35 (38.9)	190 (12.1)
25–34 years old (N = 26,920)
N	60	4165	11,800	3950	3630	1450	1865	26,920
Surgical approach, n (%)
Abdominal	20 (33.3)	3435 (82.5)	9010 (76.4)	2595 (65.7)	2485 (68.5)	795 (54.8)	1145 (61.4)	19,485 (72.4)
Robotic/laparoscopic	40 (66.7)	620 (14.9)	1940 (16.4)	1010 (25.6)	1025 (28.2)	640 (44.1)	595 (31.9)	5870 (21.8)
Vaginal	0 (0.0)	110 (2.6)	850 (7.2)	345 (8.7)	120 (3.3)	15 (1.0)	125 (6.7)	1565 (5.8)
Bilateral salpingo-oophorectomy, n (%)	50 (83.3)	350 (8.4)	1525 (12.9)	485 (12.3)	1365 (37.6)	610 (42.1)	370 (19.8)	4755 (17.7)
35–44 years old (N = 106,960)
N	15	50,055	25,925	13,550	9040	4725	3650	106,960
Surgical approach, n (%)
Abdominal	0 (0.0)	42,000 (83.9)	17,180 (66.3)	8490 (62.7)	6190 (68.5)	3225 (68.3)	2230 (61.1)	79,315 (74.2)
Robotic/laparoscopic	15 (100.0)	6825 (13.6)	6335 (24.4)	3870 (28.6)	2515 (27.8)	1450 (30.7)	1240 (34.0)	22,250 (20.8)
Vaginal	0 (0.0)	1230 (2.5)	2410 (9.3)	1190 (8.8)	335 (3.7)	50 (1.1)	180 (4.9)	5395 (5.0)
Bilateral salpingo-oophorectomy, n (%)	10 (66.7)	6090 (12.2)	6845 (26.4)	2120 (15.6)	3650 (40.4)	3070 (65.0)	850 (23.3)	22,635 (21.2)
45–54 years old (N = 127,735)
N	25	70,665	29,725	9770	5480	10,510	1560	127,735
Surgical approach, n (%)
Abdominal	15 (60.0)	59,865 (84.7)	18,435 (62.0)	6775 (69.3)	3735 (68.2)	8130 (77.4)	1000 (64.1)	97,955 (76.7)
Robotic/laparoscopic	10 (40.0)	9405 (13.3)	7820 (26.3)	2365 (24.2)	1520 (27.7)	2340 (22.3)	480 (30.8)	23,940 (18.7)
Vaginal	0 (0.0)	1395 (2.0)	3470 (11.7)	630 (6.4)	225 (4.1)	40 (0.4)	80 (5.1)	5840 (4.6)
Bilateral salpingo-oophorectomy, n (%)	25 (100.0)	24,845 (35.2)	16,420 (55.2)	3615 (37.0)	3010 (54.9)	9245 (88.0)	730 (46.8)	57,890 (45.3)

Patient characteristics and hospital region by indication

The patients in the GAC group tended to be younger (mean age of 34.5 years) than patients undergoing hysterectomy for another indication (Supplementary Table S2). The majority of GAC patients were under 35 (63.6%), compared to other noncancer indications where the majority of cases were 35 and older, ranging from 72.7% (pelvic pain) to 96.7% (fibroids). White patients made up the majority of patients undergoing inpatient hysterectomy for each indication other than fibroids (which had a higher prevalence of Black individuals [36.1%] than other groups) and consisted of half of the individuals undergoing hysterectomy. There were no uninsured patients in the GAC group, with the majority using private insurance (54.5%) and the remainder using public insurance (45.5%, Supplementary Table S2). Private insurance was more prevalent for other indications than for the GAC group (Supplementary Table S2).

Secondary gynecologic diagnoses were less common in patients with a primary indication of GAC. Endometriosis (18.2%), pelvic pain (13.6%), and fibroids (4.5%) were the only secondary diagnoses among patients with a primary GAC indication (compared to, for example, 43.3% of patients with a primary diagnosis of fibroids having a secondary diagnosis of AUB). A secondary diagnosis of GAC was extremely uncommon (<0.5%) for any other indication.

Medical comorbidity among those with GAC as their primary indication was more common (diabetes 9.1% and obesity 22.7%) than for other primary indications, with the exception of cancer. Comorbid hypertension occurred at a lower rate in the GAC group (13.6%) than all other groups (Supplementary Table S3).

Notably, comorbid affective disorders in those with the primary indication of GAC (18.2%) were more common compared to other indications. This trend holds with comorbid anxiety disorders as well, with a rate of 27.3% in the GAC group. Rates of comorbid mental and behavioral disorders due to psychoactive substance use appear to be similar across groups, with a slightly higher rate in the pelvic pain group and lower rate in the fibroids group (Supplementary Table S2). There were no individuals in the GAC group with comorbid psychotic disorders, behavioral syndromes, personality disorders, intellectual disabilities, developmental disorders, or premenstrual dysphoric disorder, and rates of these disorders were low among all other groups (ranging from 0.0% to 1.0%, Supplementary Table S2).

Hysterectomies for the GAC group were much more likely to occur in the Northeast (45.5%) and West (36.4%) as opposed to Midwest (4.5%) and South (13.6%). These regional rates are different for other primary indications. For all other indication groups, the majority of hysterectomies occurred in the South (42.1%). The Northeast had the lowest rates for all other groups (excluding fibroids, for which the lowest rates occurred in the Midwest) (Supplementary Table S2).

Discussion

The population-based rate for inpatient hysterectomy for GAC was low compared to other indications for hysterectomy in this hospital-based population; bilateral salpingo-oophorectomy was more common with GAC than other indications. This greater proportion of bilateral salpingo-oophorectomy was pronounced among young patients (ages 18–34 years), for whom bilateral salpingo-oophorectomy was typically uncommon for noncancerous gynecologic indications. In addition, hysterectomies for GAC were completed laparoscopically or robotically at a much higher rate than for other indications, and none was reported vaginally for GAC.

The younger age at hysterectomy in the GAC group was consistent with previous findings.¹⁴ Individuals undergoing hysterectomy for gender-affirming surgery (which as demonstrated here was often accompanied by bilateral salpingo-oophorectomy) may be more inclined to do so at a younger age due to potential factors, including differing considerations of fertility and impact on endogenous hormone levels that aligns with their gender. Furthermore, other patients may be older because symptoms may be worse as they get older. Although our findings show that the GAC group has lower prevalence of pelvic pain than other indications (except cancer), past research demonstrated that those taking testosterone medication may experience abdominopelvic pain which may then be resolved by hysterectomy.¹⁵

Gender minority stress, added stressors to one's life from discrimination, violence, expectations of rejection against their gender identity, particularly from increased anti-transgender sentiment during this time, may put transgender individuals at an increased risk of comorbidities such as diabetes and obesity.¹⁶ In addition, there is evidence from previous research of greater than expected prevalence of polycystic ovarian syndrome (and thus possibly an insulin-resistant state)^17,18 among transgender men, regardless of exposure to exogenous testosterone. It is difficult to draw inferences regarding the lower rate of comorbid hypertension in the GAC group, but this may simply be due to the younger age in this group and thus decreased likelihood of hypertension.¹⁹ The higher prevalence of comorbid affective and anxiety disorders in those with GAC could be explained, in part, by the overall higher incidence of affective and anxiety disorders in transgender and gender nonconforming individuals.²⁰

The low prevalence of GAC compared to other indications for hysterectomy reflects general population estimates of transgender individuals (0.5%) compared to cisgender individuals eligible for hysterectomy.²¹ Given the low number of individuals in the GAC group, conclusions cannot be drawn regarding the lack of other psychiatric secondary diagnoses, but these diagnoses appear to occur at similarly low rates as in the general population. If relatively few secondary diagnoses were coded for those with the primary indication of GAC, this may imply that some insurance payers were covering hysterectomy for GAC. This is notable because lack of insurance coverage for gender-affirming surgery remains an important concern.⁷

With increased insurance coverage for gender-affirming procedures, it is even more important to measure hysterectomy procedures for GAC over time. In addition, if relatively few secondary diagnoses were coded for those with the primary indication of GAC, a primary indication of GAC is adequately sensitive for GAC, which is similar to past literature using ICD codes for GAC alone to identify transgender patients within Veterans Affairs data.¹¹

Although the feasibility of vaginal hysterectomy for transgender and gender-nonconforming individuals has been supported by a number of studies,^14,17,22,23 this surgical route was rare for GAC surgeries in this national sample. Vaginal hysterectomy could be performed in outpatient centers; outpatient surgeries were not included in this national dataset of inpatient surgeries. The lack of vaginal hysterectomy in the GAC group may also be due, in part, to decreases in rates of vaginal hysterectomy in general.²⁴ In addition, younger gynecologists, who may be more likely to perform gender-affirming surgery, are more likely to be trained in laparoscopic than vaginal hysterectomy.²⁵

As patients seek concomitant bilateral salpingo-oophorectomy at the time of hysterectomy, a laparoscopic approach is also reasonable for that indication. Other reasons to avoid vaginal hysterectomy for GAC may include greater prevalence of nulliparity, “no uterine descent,” and vaginal atrophy due to testosterone therapy.²³ The high prevalence of abdominal cases across all indications is not surprising given that we used an inpatient sample. The lower proportion of abdominal hysterectomies in the GAC group in contrast to all other groups may be partially due to the younger age, lower prevalence of secondary disorders or complex gynecologic conditions that would call for abdominal surgery, and potential decreased degree of difficulty of hysterectomy (large uterus or scar tissue after other surgical procedures) in this group.

The higher rates of bilateral salpingo-oophorectomy in the GAC group may be because the removal of ovaries may address gender dysphoria and incongruence of the presence of ovaries with gender identity among transgender and gender nonconforming patients.²⁶ Transgender patients may also want to avoid future gynecological appointments and utilize the procedure as a means of prevention if they have prior medical problems.²⁷ Furthermore, the procedure may be required for legal gender change documents in their state.^27,28 Conversely, transgender patients may want to retain their ovaries for fertility preservation or future use of sex steroids if they were to stop taking testosterone.²⁸ The above are just a few factors that are a part of the complicated discussion regarding the decision on route of surgery and whether or not to undergo bilateral salpingo-oophorectomy.

Clear communication of potential negative outcomes such as all-cause mortality from early oophorectomy with discontinuation of testosterone needs to be discussed with transgender patients.²⁹ Recent articles document the need for more informed oophorectomy counseling and provide suggestions of discussion points to consider.^28,30 The need for clear and informed oophorectomy counseling highlights the importance of shared decision-making, especially concerning oophorectomy in younger individuals given the high rate of bilateral salpingo-oophorectomy in the GAC group.

The low rates of hysterectomy for GAC in the South and Midwest are interesting considering that notably higher proportions of the respondents in the 2015 U.S. Transgender Survey were from the South (29%) and Midwest (19%).⁷ There are many possible explanations for the regional differences seen by indication, including more stigma regarding gender-affirming surgery and thus fewer providers who conduct GAC procedures in the South and Midwest, and greater acceptance and knowledge of transgender and gender nonconforming individuals and their surgical needs in parts of the Northeast and West. Finally, lower rates of insurance coverage for nonelderly adults are especially a barrier to gender-affirming surgery in the South⁷; the regional disparity of hysterectomy for GAC presented here highlights the need for universal health care insurance coverage and providers able to provide GAC and surgeries.

Limitations, strengths, and future directions

There are several limitations of this analysis. The most important limitation is the lack of data on procedures performed at ambulatory surgical centers. Because inpatient surgeries represent a minority of hysterectomies performed, these numbers are likely an underestimation of gender-afffirming hysterectomy. A statistical brief found that in 2013, 53% of hysterectomies with oophorectomy were in the inpatient setting, while 30.7% of hysterectomies alone were in the inpatient setting in five states in the South, the Northeast, and the Midwest.³¹ Since then, there has been a large shift from inpatient hysterectomies to outpatient hospital settings, particularly for benign conditions.^32,33

The total number of individuals with a primary indication of GAC is relatively low, limiting our ability to do more analyses because of low precision. The National Inpatient Sample contains no variable on self-reported gender identity, so it is difficult to identify with certainty all patients who are transgender and gender nonconforming, and another reason our estimates are likely underestimated. In addition, the American Community Survey in the U.S. census only records self-identified sex with a binary selection (male or female), so we are not able to separate population based rates of inpatient hysterectomies by gender identity or know whether respondents are transgender women or men.³⁴

The diagnoses coded in the National Inpatient Sample do not entirely encompass all transgender and gender nonconforming individuals who have received hysterectomy, as not all will have GAC recorded as an indication for their surgeries. Diagnosis codes are primarily for billing purposes, and the coding may not reveal gender identity diagnoses due to privacy or insurance coverage reasons. For example, health care providers have used endocrine disorder codes to avoid outing patients as transgender to their health insurance, leading to underreporting of gender-affirming hysterectomies.³⁵

The primary indication was assumed to be the first ICD-10 diagnosis code listed for each encounter, but the primary indication cannot be known with absolute certainty, which may lead to misclassification. However, in this analysis, GAC was rarely listed as a secondary diagnosis, which leads us to believe that we captured most hysterectomies coded this way.

Our study years are close to when Medicare began approving gender-affirming surgeries in 2014, and private insurance had recently begun to catch up in coverage.³⁶ In addition, the 2016 Presidential Election had a negative impact on the health of transgender and gender-diverse people because of advocacy and implementation of antitransgender policies.³⁷ Thus, this study provides data on health care utilization among transgender patients during a critical transitional period.

The major strength of this work is the nationally representative sample. The National Inpatient Sample's well-validated weighted sample allowed us to calculate absolute population rates of surgery and identify pronounced geographic variation in gender-affirming hysterectomy. Furthermore, we were able to compare across indications in the same U.S. population. Other strengths include relatively more recent data than previous literature in a fast-evolving area of practice.

The regional disparities described may reflect transgender and gender nonconforming individuals who are not able to receive the care they need for good health and well-being; further investigation must evaluate the impacts of not receiving this care in underserved regions. Given the likely increase in rate of hysterectomies for GAC in recent years and higher prevalence in outpatient settings, future research could update our findings by determining the trends in rates of hysterectomies for GAC in the outpatient setting.

Conclusion

With further advocacy to use the validated two-step method to document gender identity among patients, future research can include this measure to validate GAC ICD-codes.⁵ Other studies could also include longitudinal data, with mapping of ICD-10 to ICD-9 codes to compare data. Our findings reflect that the transgender and gender nonconforming population was achieved greater acceptance of GAC in the years immediately after insurance coverage began expanding in the United States. However, this group remained underserved in some parts of the United States. We hope that more research characterizing this population improves efforts to deliver high-quality care to them equitably, regardless of socioeconomic and residential circumstances.

Authors' Contributions

D.E.C.: Conceptualization, methodology, data curation, formal analysis, writing—original draft preparation. T.G.B.: Visualization, writing—reviewing and editing. J.M.W.: Software, data curation, formal analysis, writing—reviewing and editing. A.E.: Validation, writing—reviewing and editing. W.R.R.: Project administration, funding acquisition, writing—reviewing and editing. E.R.M.: Methodology, investigation, formal analysis, writing—reviewing and editing. B.A.G.: Supervision, resources, project administration, funding acquisition, writing—reviewing and editing.

Disclaimer

The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

B.A.G. was supported by an IUD trial by Sebela. T.G.B. and W.R.R. were supported by the National Institute on Minority Health and Health Disparities of the National Institutes of Health (R01MD011680, PI: W.R.R.). Sponsors had no such involvement in the writing or submission of this article.

Supplementary Material

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3