Provision of gender-affirming hormones for trans and gender-diverse adults: a systematic review of health and quality of life outcomes, values and preferences, and costs

Summary

Background

Many transgender and gender-diverse individuals take gender-affirming hormones, but the evidence for their impact on health and wellbeing has not been comprehensively synthesized.

Methods

We conducted a systematic review of studies reporting health and quality of life outcomes, values and preferences, and costs associated with gender-affirming hormones. The protocol was prospectively registered in PROSPERO (CRD42024451558). We searched PubMed, Embase, CINAHL, Cochrane Central, LILACS, PsycINFO, and grey literature sources for relevant evidence published between 1 January 2013 and 31 March 2024, without language or geographical restrictions. The health and quality of life outcomes review included data from randomized controlled trials (RCT) and longitudinal studies that compared adults (age 18+) who received gender-affirming hormones within the health sector to those who did not but expressed a desire to receive such care. We synthesized outcomes data using GRADE evidence profiles and assessed risk of bias with Cochrane ROB2 and ROBINS-I. We qualitatively summarized findings on values and preferences and cost, and provided confidence ratings using GRADE CERQual.

Findings

Thirteen studies, including one RCT and 12 longitudinal studies, were included in the health and quality of life outcomes review. Data from the RCT was judged to be low-certainty evidence. Uncontrolled longitudinal studies had serious to critical risk of bias. Findings from these studies suggest gender-affirming hormones may improve outcomes including depression, quality of life, and suicidality. No studies identified substantive harms. We identified an additional 17 studies on values and preferences and one study presenting cost data; data suggest that gender-affirming hormones are acceptable and affordable.

Interpretation

Available evidence indicates gender-affirming hormones may improve critical outcomes, including mental health and quality of life, with no evidence of substantive harms. Further evidence from prospective, controlled studies could improve the confidence in these findings.

Funding

World Health Organization through a grant from the Elton John AIDS Foundation and the Bill and Melinda Gates Foundation.

Research in context.

Evidence before this study

Gender-affirming hormones have been provided to transgender (trans) and gender-diverse individuals as evidence-based interventions for several decades. Prior reviews have synthesized available data on the impact of gender-affirming hormones on various health outcomes including mental health, cardiovascular health, and quality of life, but no reviews have comprehensively reviewed and synthesized data on health and quality of life outcomes, values and preferences, and cost globally. We searched six databases (PubMed, Embase, CINAHL, Cochrane Central, LILACS, and PsycINFO) for articles published between 1 January 2013 and 31 March 2024 without geographic or language restriction. Search terms included two primary concepts: Concept 1 (trans and gender-diverse people) + Concept 2 (gender-affirming care). We identified 13 articles reporting health and quality of life outcome data from longitudinal studies, 17 reporting on values and preferences, and one reporting on cost. Findings were assessed to be low certainty in the outcomes review. We had moderate to high confidence in the synthesized themes from the values and preferences studies.

Added value of this study

This systematic review comprehensively synthesizes data for health and quality of life outcomes, values and preferences, and costs associated with gender-affirming hormones among trans and gender-diverse adults globally. We rigorously reviewed comparative data between trans and gender-diverse adults who received gender-affirming hormones through the health sector to those who desired but did not receive this care across 15 health and quality of life outcomes. Our values and preferences review identified nine key themes, including high sociocultural acceptability and many perceived benefits following hormone use. We also identified one cost study demonstrating relative affordability of gender-affirming hormones.

Implications of all the available evidence

Synthesizing all available evidence, gender-affirming hormones may improve health outcomes, including critical ones such as mental health, suicidality, and quality of life, among trans and gender-diverse adults. We found that trans and gender-diverse adults and health professionals who provide gender-affirming hormones generally perceive the benefits to outweigh the risks. We also found that perceived benefits included not just individual improvements in health and wellbeing but also positive societal impacts. We identified several outcomes for which there was no available comparative data including: access to health services, physical health, stigma related to gender identity, discrimination related to gender identity, utilization of health services, mortality, satisfaction with health services, or violence experienced by trans and gender-diverse people. Future research is needed to address these gaps, particularly data from prospective, controlled studies, where ethically feasible (e.g., quasi-experimental designs or standard-of-care waitlist designs). Given the finding that gender-affirming hormone therapy may improve mental health, suicidality, and quality of life, and the acceptability and affordability of this intervention, data support the continued provision of gender-affirming hormones to trans and gender-diverse adults who desire them, in accordance with evidence-based clinical practice recommendations.

Introduction

Transgender (trans) and gender-diverse people globally experience disproportionate burdens of stigma, violence, and negative physical and mental health outcomes, which are primarily attributable to structural and social determinants of health.1, 2, 3 Many trans and gender-diverse individuals seek gender affirmation through social, psychological, medical, and legal processes and services as gender affirmation can mediate the relationships between stigma and discrimination and negative health and wellbeing outcomes.⁴ One common option available to trans and gender-diverse people seeking medical gender affirmation is gender-affirming hormone therapy.

Prior reviews have examined the associations between gender-affirming hormone therapy and quality of life, mental health, and cardiovascular outcomes.1, 2, 3, 4, 5 Gender affirming hormone therapy has been associated with better health and quality of life outcomes among trans and gender-diverse people who have received this care compared to those who have not.⁵ However, much of the data has been observational, and synthesis of the highest quality evidence is needed to inform global clinical and public health guidelines. Further, we did not identify any existing reviews that comprehensively and rigorously synthesized health and quality of life outcomes, values and preferences, and cost data.

We sought to systematically review the available evidence on the impact of gender-affirming hormone therapy on health and wellbeing among trans and gender-diverse individuals who desire this care. We also conducted complementary values and preferences and cost reviews, synthesizing available data from trans and gender-diverse adults and healthcare professionals that provide this care.

Methods

Search strategy and selection criteria

We systematically reviewed available evidence on gender-affirming hormones among trans and gender-diverse adults globally, including data on fifteen key health and quality of life outcomes, values and preferences, and cost, following PRISMA 2020 guidelines.⁶ The protocol for this review was prospectively registered in PROSPERO (CRD42024451558).

Studies published between 1 January 2013 and 31 March 2024 were eligible for inclusion. These dates were selected to capture data published after implementation of the World Professional Association of Transgender Health (WPATH) Guidelines version 7 through the search date. There were no geographic or language restrictions. Studies had to report primary data to be included. Our population of interest was trans and gender-diverse adults ages 18 years or older who sought gender-affirming hormones; therefore, we excluded studies that exclusively focused on children (below age 18). We requested disaggregated data for adult participants for studies that included both adults and children. If authors provided these data, we included data from adult participants; studies that did not respond to requests for disaggregated data were excluded. For the health and quality of life outcomes review, we included studies that presented quantitative data on at least one of fifteen health and quality of life outcomes selected a priori by the WHO Guideline Development Group, an external (i.e., not employed by WHO) group of international experts on trans and gender-diverse health. We considered the top eight outcomes as primary outcomes (summarized using GRADE) and the subsequent seven as secondary outcomes (summarized narratively). Final ranking and definitions of all fifteen outcomes are available in Appendix A. We included randomized controlled trial (RCT) or uncontrolled/controlled longitudinal (before/after) study designs because we were interested in comparative data between adults who received gender affirming hormones within the health sector to those who did not. For the values and preferences and cost review, we included studies that presented values and preferences of trans and gender-diverse individuals and/or health workers or data on cost, without restriction to study design. For the values and preferences review, we included studies that presented values and preferences for the use or provision of gender-affirming hormones within the health sector. For the cost review, we included studies if they presented primary data comparing cost, cost-effectiveness, cost-utility, or cost-benefit of the intervention and comparison, or if they presented cost-effectiveness of gender-affirming hormones as it related to the included health and quality of life outcomes.

We searched 6 databases (PubMed, EMBASE, CINAHL, PsycINFO, Cochrane Central, and LILACS) for relevant publications. Search terms were adapted from other recent reviews7, 8, 9 and included two primary concepts: gender-affirming care and trans and gender-diverse people (Appendix B). This review was nested within a larger review in which the search terms were developed to capture the broader concept of gender-affirming care, including hormones. References of included articles and four relevant systematic reviews^5,9, 10, 11 were secondarily searched to identify additional studies. We searched for grey literature by searching conference abstracts from health-related scientific consortia focused on trans and gender-diverse individuals (World Professional Association for Transgender Health (WPATH), US PATH, European PATH, Asia PATH) and websites of relevant organizations (Appendix C). For the values and preferences review, we focused on the acceptability of gender-affirming hormones, health and quality of life outcomes that individuals are seeking when accessing gender-affirming hormones, and the degree to which there is consensus from transgender individuals and stakeholders on these values and preferences. We excluded data on gender-inclusive health services broadly, discussion of barriers to accessing gender affirming hormones, and reactions to encountering these barriers.

A study team member conducted initial screening of all titles/abstracts to remove clearly irrelevant articles (e.g., pre-clinical studies) and duplicates. Titles/abstracts were then screened by two study team members independently. Full-text articles were assessed independently by two team members to determine eligibility for inclusion. Differences were resolved through consensus. While our protocol planned for inclusion of studies in all languages, all included studies were published in English.

Data analysis

Data were extracted independently by two study team members using standardized forms in Microsoft Excel. Differences were resolved through consensus and referral to a third team member. Extracted information included: citation information, location (country, region, urban/rural, World Bank income classification), population description (e.g., age, gender identities), study methods (e.g., sample size, study design), intervention description, comparator, key findings, study outcomes, limitations, conclusions, and ethical considerations.

We assessed risk of bias (ROB) using the Cochrane Collaboration’s ROB 2.0 tool for RCTs.¹² This tool assesses random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data addressed (attrition bias), and selective reporting (reporting bias). Methodological components of the studies were assessed and classified as high, unclear, or low risk of bias. We used the ROBINS-I tool to assess the risk of bias for longitudinal studies.¹⁰ For the “bias due to missing data” domain, we judged loss to follow-up at and after 6 months as “low” if <5% missing, “moderate” if 5–10% missing, “serious” if 10–30% missing, and “critical” if 30% or more were missing. If baseline characteristics of those lost to follow-up and those retained for follow up were compared by study investigators and found comparable, we moved the risk down a level (e.g., from serious to moderate). We used the robvis tool¹³ to create risk of bias traffic light and summary plots.

Findings for the health and quality of life outcomes review were summarized in a GRADE Evidence Profile using GRADEPro, prioritizing RCT data over observational data where available and rigorously conducted according to appropriate ethical standards. The certainty for each outcome was judged as either high, moderate, low, or very low based on the risk of bias, consistency, directness and precision of the pooled estimate of effect per outcome. Imprecision was operationalized as a risk ratio with a confidence interval including both clinically relevant benefit and harm (i.e., greater than 1.25 or less than 0.80); we downgraded once for imprecision. Inconsistency was downgraded once if there was only a single study, since consistency could not be assessed. We downgraded once for indirectness if a study included a sample that may not represent the wider trans and gender-diverse population (e.g., trans men only). Statements of confidence throughout the manuscript reflect standard GRADE language.¹¹ Our protocol planned for meta-analysis, but the outcome measures and study designs across studies were not sufficiently similar.

We summarized the values and preferences data qualitatively and used GRADE CERQual to synthesize evidence and provide a confidence rating.¹² Findings for the cost review were summarized descriptively.

Role of the funding source

This review was commissioned to inform a WHO guideline development process around the health of trans and gender-diverse people. Members of the review team were not involved in decision making regarding recommendations. Conversely, those involved in developing the recommendations had no engagement in the screening, data extraction, data synthesis, quality assessment, manuscript development and/or decision to submit for publication.

Results

A PRISMA diagram representing the yield of citations through each stage of the search, screening, and extraction process is presented in Fig. 1. The database search identified 17,532 unique citations, of which 94 were selected for full-text review to further assess eligibility. We identified an additional 146 citations through additional search methods. Across the two search strategy streams, 208 articles were excluded (reasons for exclusion are reported in Appendix D). The health and quality of life outcomes studies were all conducted in Europe and Australia, while the values and preferences studies were conducted in Asia, Europe, North America, Oceania, and South America. The single cost study was conducted in the United States (Fig. 2).

Fig. 1.

PRISMA flowchart showing yield of citations through the search and screening process.

Fig. 2.

Map of studies included in systematic review of health and quality of life outcomes, values and preferences, and costs associated with gender-affirming hormones.

For the health and quality of life outcomes review, one RCT^13,14 and twelve uncontrolled longitudinal (before/after) studies met our inclusion criteria (Table 1).15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26 The RCT was conducted in Melbourne, Australia. The authors randomized trans and gender-diverse adults who were seeking testosterone therapy for masculinization (N = 64) to standard care (3-month waiting list before commencing testosterone therapy) or intervention (immediate initiation of testosterone therapy); as there was a standard care waiting list of 3 months, this design ensured no individuals would be waiting longer than the standard of care. The twelve observational studies included a variety of hormone regimens and follow-up times ranged from 3 months to 2 years following hormone initiation.

Table 1.

Studies included in the outcomes review (k = 13).

Study	Country	Intervention	Population	Sample size	Follow-up time(s)	Outcomes
Castellini et al., 202316	Florence, Italy	Testosterone for AFAB individuals; both antiandrogens and estrogens for AMAB	Transgender and gender-diverse individuals	200 (122 AFAB including 36 nonbinary, 78 AMAB including 20 nonbinary)	1 year	Mental health (critical) Perception of wellbeing (important) Gender incongruence (important)
Colizzi et al., 201415	Bari, Italy	Transdermal estradiol gel and oral cyproterone acetate for AMAB; Intramuscular testosterone injections for AFAB	Transgender and gender-diverse individuals, excluding “unstable psychiatric comorbidity”	107 (78 AMAB, 29 AFAB)	1 year	Mental health (critical)
Pavanello Decaro et al., 202119	Ghent, Belgium	Hormone therapy (testosterone for AFAB; Estrogens and possible additional antiandrogenic therapy for AMAB)	Transgender and gender-diverse individuals	119 (57 AMAB, 62 AFAB)	6 months	Quality of life (critical)
Fisher et al., 201617	Italy	AMAB: oral cyproterone acetate combined with oral estradiol valerate or transdermal estradiol; AFAB: intramuscular testosterone. All patients received standardized professional mental health support every 3 months.	Transgender and gender-diverse individuals	54 (28 AMAB, 26 AFAB)	2 years	Mental health (critical) Perception of wellbeing (important) Gender incongruence (important)
Foster Skewis et al., 202120	Melbourne, Australia	Masculinizing and femininizing hormone therapy	Transgender and gender-diverse individuals	77 (35 AMAB, 42 AFAB)	3 and 6 months	Quality of life (critical) Gender incongruence (important)
Heylens et al., 201421	Ghent, Belgium	Hormone therapy (unspecified); most participants also received sex reassignment surgery	Transgender and gender-diverse individuals	57 (46 AMAB and 11 AFAB)	3–6 months	Mental health (critical) Quality of life (critical) Suicidal behaviors (critical) Substance use (important)
Manieri et al., 201422	Turin, Italy	Hormone therapy (unspecified); all participants had weekly psychodynamic psychotherapy	Transgender and gender-diverse individuals	83 (56 AMAB, 27 AFAB)	1 year	Quality of life (critical) Physical health (important) Perception of wellbeing (important) Substance use (important)
Matthys et al., 202123	Amsterdam, the Netherlands; Ghent, Belgium; Florence, Italy; Oslo, Norway	Testosterone undecanoate, testosterone gel, and/or testosterone esters for AFAB; estrogens plus antiandrogens for AMAB	Transgender and gender-diverse individuals	873 (451 AMAB, 422 AFAB)	3, 6, 9, 12, 18, 24, and 36 months	Quality of life (critical)
Metzger et al., 202124	Germany & Switzerland	Testosterone therapy	Transgender and gender-diverse individuals AFAB	26	3 and 6 months	Mental health (critical)
Morssinkhof et al., 202425	Amsterdam and Groningen, The Netherlands	Testosterone and progestin/combined hormonal contraceptives for AFAB; estradiol and/or anti-androgens for AMAB	Transgender and gender-diverse individuals	199 (89 AMAB including 5 non-binary; 110 AFAB including 19 non-binary)	3 and 12 months	Mental health (critical)
Nolan et al., 202313 Nolan et al., 202414	Melbourne, Australia	Testosterone therapy offered immediately (as opposed to 3-month waitlist per standard of care)	Transgender and gender-diverse adults seeking masculinization	64	3 months	Mental health (critical) Quality of life (critical) Suicidal behaviors (critical)
Ristori et al., 202018	Italy	Gender-affirming hormone therapy (unspecified)	Transgender and gender-diverse Individuals	72 (36 AMAB, 36 AFAB)	2 years	Quality of life (critical)
Tebbens et al., 201926	Amsterdam, The Netherlands	Cyproterone acetate and estradiol for AMAB, testosterone for AFAB	Transgender and gender-diverse individuals	23 (14 AMAB, 9 AFAB)	1 year	Quality of life (critical) Perception of well-being (important)

Summary and study-level risk of bias assessments are presented by study design in Appendices E and F. The RCT was judged to be low risk of bias on all categories except allocation concealment, where it was judged unclear risk of bias as few details were provided, and blinding, where it was judged high risk of bias because blinding was not possible due to the nature of the intervention. While inherently limited compared to RCTs, the uncontrolled before/after studies tended to score moderately across risk of bias domains. The primary weaknesses across studies were bias due to missing data, bias in measurement of outcomes, and bias in selection of the reported result.

Several pre-specified outcomes of interest were not measured in any study: access to health services, physical health, stigma related to gender identity, discrimination related to gender identity, utilization of health services, mortality, satisfaction with health services, or violence experienced by trans and gender-diverse people. Data for the remaining outcomes are presented by study design.

Table 2 presents the GRADE evidence profile. The RCT started as high certainty of evidence due to the study design, and was downgraded to low certainty evidence due to risk of bias (blinding was not possible given the nature of the intervention and the outcomes may have been affected by lack of blinding) and imprecision (given the small sample size for outcome evaluation (n = 62), a small number of individuals have the potential to substantially change the statistical significance of the findings). The analysis followed an intention-to-treat approach. Results suggested that gender-affirming hormone therapy may decrease mental health conditions. The trial found that those who initiated testosterone therapy immediately were more likely to report improvements in depression symptomology at 3-months as measured by the Patient Health Questionnaire-9 (PHQ-9; mean difference: −5.6, 95% CI: (−6.8, −4.4), p < 0.001). A clinically significant improvement greater than or equal to 5 points in the PHQ-9 score was reported in 19 individuals (61%) in the intervention group and 4 individuals (13%) in the comparison group (p < 0.001). Results also suggested that gender-affirming hormone therapy may decrease suicidal behaviors. The trial found that those who initiated testosterone therapy immediately were more likely to report improvements in suicidal behaviors as measured by the Suicidal Ideation Attributes Scale (SIDAS; mean difference: −6.5, 95% CI: (−8.2, −4.8), p < 0.001). In individuals with suicidality at baseline as assessed by PHQ-9 item 9, resolution of suicidality (no longer endorsing PHQ-9 item 9) occurred in 11 individuals (52%) in the intervention group and 1 individual (5%) in the comparison group (p = 0.002). Results also suggested that gender-affirming hormone therapy may increase or have no effect on quality of life. The trial found that those who initiated testosterone therapy immediately reported increased quality of life as measured by the EQ-VAS score (mean difference: +11.6, 95% CI: (4.9, 18.3), p = 0.02), while there was a positive but non-significant trend towards benefit in the EQ-5D-5L score (mean difference: +0.07, 95% CI: (−0.07, 0.21), p = 0.06). This mean difference in EQ-VAS score represents a clinically significant improvement, which has been previously defined as an improvement of 7–10 points. Similarly, the mean difference in EQ-5D-5L represents a clinically significant improvement, which has previously defined improvements of 0.03–0.07 as clinically significant. However, it is important to note that EuroQoL Group recommends context-specific interpretation given variability in minimally important differences across populations, and clinically significant improvements have not been established for trans and gender-diverse populations.²⁷ While the trial reported adverse events in the intervention group, these were not measured in the control group. Seven individuals reported pain/discomfort at the injection site, and one individual reported a transient headache 24 h after intramuscular administration of testosterone undecanoate. No individuals developed polycythemia. This study also measured the secondary outcome of gender dysphoria using the GPSQ score. The study found that those who initiated testosterone therapy immediately were more likely to report improvements in gender dysphoria compared to those who received standard of care (mean difference: −7.2, 95% CI: (−8.3, −6.1), p < 0.001). Clinically meaningful improvements were observed in 39% of intervention participants compared to 0% of control participants. In addition to the GRADE evidence profile, Table 3 summarizes the complete results from the RCT.^13,14

Table 2.

GRADE evidence profile for gender-affirming hormone therapy for trans and diverse people.

Certainty assessment							No of patients		Effect		Certainty	Importance
No of studies	Study design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	Gender-affirming hormone therapy	No provision of hormones for gender transition	Relative (95% CI)	Absolute (95% CI)
Critical: mental health: depression (follow-up: mean 3 months; assessed with: Patient Health Questionnaire-9 (PHQ-9) (higher scores indicate greater depression severity); Scale from: 0 to 27)
113,a	randomised trials	seriousb	not seriousc	not seriousd	seriouse	none	31	31	–	MD 5.6 points lower (6.8 lower to 4.4 lower)	⨁⨁◯◯ Low	CRITICAL
Critical: quality of Life (follow-up: mean 3 months; assessed with: EQ-Visual Analogue Scale (EQ-VAS) (higher scores indicate best imaginable health); Scale from: 0 to 100)
114,f,g	randomised trials	seriousb	not seriousc	not seriousd	seriouse	none	31	31	–	MD 11.6 points higher (4.9 higher to 18.3 higher)	⨁⨁◯◯ Low	CRITICAL
Critical: quality of Life (follow-up: mean 3 months; assessed with: EQ-5D-5L (measures 5 dimensions of mobility, self-care, usual activities, pain/discomfort, and anxiety/depression) (higher scores indicate full health); Scale from: 0 to 1)
114,f,h	randomised trials	seriousb	not seriousc	not seriousd	seriouse	none	31	31	–	MD 0.07 points higher (0.07 lower to 0.21 higher)	⨁⨁◯◯ Low	CRITICAL
Critical: access to health services
0									not estimable		–	CRITICAL
Critical: suicidal behaviors (i.e., suicidal ideation, attempted suicide, suicide mortality) (follow-up: mean 3 months; assessed with: Suicidal Ideation Attributes Scale (SIDAS) (higher scores indicate greater suicidality); scale from: 0 to 50)
113,i	randomised trials	seriousb	not seriousc	not seriousd	seriouse	none	31	31	–	MD 6.5 points lower (8.2 lower to 4.8 lower)	⨁⨁◯◯ Low	CRITICAL
Critical: stigma related to gender identity
0									not estimable		–	CRITICAL
Critical: discrimination related to gender identity
0									not estimable		–	CRITICAL
Important: physical health
0									not estimable		–	IMPORTANT
Important: perception of well-being
0									not estimable		–	IMPORTANT

CI: confidence interval; MD: mean difference.

^aOutcomes: PHQ-9 at follow-up: Intervention Group: Decline of 6.6 from BL to FU (15.2–8.6); Comparison Group: Decline of 0.7 from BL to FU (14.1–13.4). Mean difference at follow-up (between-group difference over 3 months adjusted for the corresponding measure at baseline): −5.6 (95% CI: −6.8 to −4.4). A clinically significant improvement greater than or equal to 5 points in the PHQ-9 score was reported in 19 individuals (61%) in the intervention group and 4 individuals (13%) in the comparison group (p < 0.001).

^bRisk of bias: Study protocol was prospectively registered. Three of four study outcomes included in the protocol were reported in the published article; fourth outcome (quality of life) data was provided by authors via email, and later published in the peer-reviewed literature (after the search date of this review). While the protocol listed one quality of life outcome (EQ-5D-5L), the authors provided results on two quality of life outcomes (EQ-5D-5L and EQ-VAS). Downgraded by one as blinding of participants and personnel was not possible due to the nature of the intervention (delayed control group), and the outcome may have been affected by lack of blinding (self-report).

^cInconsistency: Single study, so evaluation of inconsistency is not possible.

^dIndirectness: Population is transgender and gender-diverse adults seeking masculinization, a subset of all trans and gender-diverse people who are seeking gender-affirming care.

^eImprecision: 64 individuals were randomized; 62 were included in the analysis at follow-up (31 in each arm). Downgraded once for small sample size, in which a small number of individuals could change the effect substantially and which may not meet the optimum information size.

^fData provided by authors via email.

^gOutcomes: EQ-VAS score: Overall cohort at baseline: mean: 56.8 (SD: 20.6). Mean difference at follow-up in intervention vs. comparison group: +11.6 points (4.9, 18.3), p = 0.02.

^hOutcomes: EQ-5D-5L score: Overall cohort at baseline: mean: 0.72 (SD: 0.19). Mean difference at follow-up in intervention vs. comparison group: +0.07 points (−0.07, 0.21), p = 0.06.

ⁱOutcomes: SIDAS at follow-up: Intervention Group: Decline of 6.4 from BL to FU (12.0–5.6); Comparison Group: Increase of 0.2 from BL to FU (11.6–11.8). Mean difference at follow-up (between-group difference over 3 months adjusted for the corresponding measure at baseline): −6.5 (95% CI: −8.2 to −4.8). In individuals with suicidality at baseline as assessed by PHQ-9 item 9, resolution of suicidality (no longer endorsing PHQ-9 item 9) occurred in 11 individuals (52%) in the intervention group and 1 individual (5%) in the comparison group (p = 0.002).

Table 3.

Change from baseline to month 3 in individuals receiving immediate commencement testosterone compared with those receiving standard care from the single randomized trial reported in Nolan et al. (2023) and Nolan et al. (2024).

Parameter	Time point	Mean (Standard deviation)		Mean difference (95% confidence interval (CI))a	p-value
		Intervention (n = 31)	Control (n = 31)
PHQ-9	0 months	15.2 (4.7)	14.1 (6.0)	−5.6 (−6.8, −4.4)	<0.001
	3 months	8.6 (5.2)	13.4 (6.3)
SIDAS	0 months	12.0 (8.5)	11.6 (10.0)	−6.5 (−8.2,−4.8)	<0.001
	3 months	5.6 (7.3)	11.8 (11.9)
EQ-VAS	0 months	56.6 (19.8)	57.0 (21.7)	+11.6 (4.9, 18.3)	0.02
	3 months	69.1 (20.6)	57.6 (21.9)
EQ-5D-5L	0 months	0.70 (0.21)	0.74 (0.17)	+0.07 (−0.07, 0.21)	0.06
	3 months	0.79 (0.16)	0.74 (0.19)
Adverse events	N/A	Pain/discomfort at the injection site: 7 Transient headache 24 h after intramuscular administration of testosterone undecanoate: 1	N/A	N/A	N/A
GPSQ	0 months	46.2 (6.0)	44.0 (6.8)	−7.2 (−8.3, −6.1)	<0.001
	3 months	38.0 (6.2)	43.4 (7.9)

PHQ-9: Patient Health Questionnaire 9; SIDAS: Suicidal Ideation Attributes Scale; EQ-VAS: EuroQOL Visual Analogue Scale; EQ-5D-5L: EuroQol 5-Dimension 5-Level; GPSQ: Gender Preoccupation and Stability Questionnaire.

^aThe mean difference and 95% CI refer to the between-group difference over 3 months adjusted for the corresponding measure at baseline.

Additional evidence from twelve uncontrolled before/after studies reported on measures corresponding to the following outcome categories: mental health conditions, quality of life, suicidal behaviors, perception of wellbeing/body image, gender incongruence/dysphoria, and substance use. In general, the evidence from these uncontrolled before/after studies is uncertain. Nearly all studies found improvements or no change over time across outcomes. Detailed summaries by outcome are reported in Appendix G. Briefly, six studies reported improvements or no change in mental health outcomes.15, 16, 17^,21,24,25 Six studies reported improvements or no change in quality of life.18, 19, 20^,22,23,26 One study found no difference in suicidal behaviors.²¹ Four studies reported improvements in perception of wellbeing (e.g., satisfaction with body image).^16,17,22,26 Four studies reported improvements or no change in gender incongruence/dysphoria.^16,17,19,20 Two studies reported reductions or no change in substance use.^21,22

For the values and preferences review, 17 articles met the inclusion criteria (Appendix H). Included articles primarily utilized qualitative study designs (n = 14), though one study utilized both qualitative and quantitative study designs, and two studies qualitatively analyzed open-ended survey data. Findings are summarized in a CerQUAL table (Appendix I) and narratively below.

Nine studies reported on the acceptability and benefits of gender-affirming hormone therapy.28, 29, 30, 31, 32, 33, 34, 35, 36 Overall, participants found gender-affirming hormone therapy to be acceptable and that the benefits outweighed the risks. Eight studies reported findings related to the outcomes prioritized in this review.28, 29, 30, 31, 32, 33^,35,37,38 The largest motivator for using gender-affirming hormones across studies included anatomical and physiological changes to the body. Six studies found that values and preferences may change over time, based on experiences using gender-affirming hormones.^31,35,36,38, 39, 40 A common theme across seven studies was related to hormone therapy’s role in individuals achieving social gender affirmation.^28,29,31, 32, 33^,37,40 Aligning one’s social presentation with their gender was an important motivator for many transgender individuals.^32,33,37,40 Participants described the importance of hormone therapy for aligning their physical characteristics with social expectations of gender.^32,33 One study from the U.S. found that participants recognized important benefits of mental health care, but did not feel that requiring mental health screening before hormone therapy initiation promoted those benefits or quality mental health care.¹⁶ Participants in two studies expressed their desire for enhanced healthcare coordination between hormone providers and those providing other services.^12,17 Limited access to gender-affirming hormone therapy and related services led many participants to rely on non-prescribed alternatives (e.g., purchasing hormones from unregulated sources).¹² Two studies mentioned hormone therapy impacting trans and gender-diverse people’s social status and participation in society.^5,7 Hormone therapy enabled trans and gender-diverse individuals to be viewed and affirmed as their gender by others, which can be especially important or desired when attempting to access social and health care services.

For the cost review, one study was included.⁴¹ This study used insurance claims data from over 200 million Americans to calculate the costs (from a payer perspective) associated with provision of gender-affirming hormones. They estimated the total costs per year of coverage for gender-affirming hormone therapy (inclusive of care provided to adolescents and adults) at US$157. The average total costs per year were US$161 for testosterone and US$248 for estrogen and antiandrogen therapy (Appendix J).

Discussion

This review identified 13 studies that presented comparative outcomes data for trans and gender-diverse adults who received gender-affirming hormones compared to those who did not. These data come exclusively from Europe and Australia, and interventions included a variety of different hormone regimens. The single RCT provided low certainty evidence, while uncontrolled before/after studies all had serious to critical risk of bias. It is important to note that the risk of bias determinations were primarily attributable to inherent limitations in studying these outcomes when gender-affirming hormones are the intervention of interest (i.e., it is not possible to mask intervention assignment due to the obvious physical changes associated with hormones and the lack of feasible alternatives to self-reported measures for most of these outcomes). For the seven outcomes for which data were available (mental health conditions, quality of life, suicidal behaviors, physical health, perception of wellbeing/body image, gender incongruence/dysphoria, and substance use), no studies identified substantive harms. Evidence from the one RCT found that gender-affirming hormones may decrease mental health conditions, suicidality, and gender dysphoria and may increase or have no effect on quality of life, depending on the measure used. Data from the uncontrolled longitudinal studies resulted in lower certainty evidence and found that it is: uncertain whether hormone therapy increases (or has no difference) in quality of life, uncertain whether hormone therapy has no difference in suicidal behaviors, uncertain whether hormone therapy increases perception of wellbeing (e.g., satisfaction with body image), uncertain whether hormone therapy decreases or has no change in gender incongruence/dysphoria, and uncertain whether hormone therapy reduces or has no difference in substance use. Additional data from prospectively registered, longitudinal, controlled study designs would meaningfully contribute to the evidence base by providing comparative data for these health and well-being outcomes. Although equipoise would not allow for placebo-controlled trials in settings where hormones are immediately available as standard-of-care, innovative study designs including quasi-experimental designs and waitlist-controlled trials (in settings where waitlists are standard of care) could ethically produce prospective comparative data.

We identified 17 studies that reported values and preferences of trans and gender-diverse individuals, 1 of which also included health workers, which resulted in moderate to high confidence themes related to the individual and societal benefits of providing gender-affirming hormones to trans and gender-diverse adults who seek this care. Gender-affirming hormones were broadly acceptable to trans and gender-diverse adults and health providers, and the benefits were perceived to outweigh potential harms. Gender-affirming hormones were perceived as a critical medical intervention for gender affirmation, which can support trans and gender-diverse individuals to more fully engage in society by aligning their gender presentation with their identity. Evidence suggests this may improve likelihood of seeking other types of healthcare (e.g., primary care), increase feelings of safety, and reduce the impact of stigma, violence, and discrimination, which is critical given the disproportionate burden of negative health and violence outcomes that trans and gender-diverse individuals experience globally.1, 2, 3

We identified one study estimating the annual costs of gender-affirming hormones, which found that hormones, including masculinizing and feminizing therapies, cost approximately US$200 per year in the US (total cost). Due to differences in health systems globally, including differences in infrastructure and resource availability, these costs are not likely to be generalizable and the relative affordability for health systems may vary depending on the setting.

This review had several strengths. We included data published in any language and any region and covered complementary data from outcomes, values and preferences, and cost studies. We conducted a comprehensive search, which included six databases and an extensive search of gray literature sources. We were able to include and prioritize RCT data, which was supplemented by uncontrolled longitudinal data for outcomes that were not measured in the RCT. Limitations of this review included the lack of data for eight of our pre-specified outcomes, inability to perform meta-analyses due to lack of comparable data, and reliance on significance testing rather than established minimal clinically important difference thresholds in some of the included studies.^42,43 It is possible that some of the heterogeneity in the observed results across studies is due to heterogeneity in measures and/or reliance in measures that have not been validated in this population; to facilitate meta-analyses in the future, use of standardized outcome measures that have been specifically designed for trans and gender-diverse individuals is needed.44, 45, 46 It is also possible that we have excluded relevant data from methodologically rigorous studies published prior to 2013 or after March 2024, as these would not have been captured in our search; thus, these findings may be subject to publication bias. Further, we were unable to verify the extent to which the informed consent model of care was implemented in each study, which has previously been shown to be correlated with patient reported outcomes.^47,48 We also excluded articles that included both children and adults in instances where authors did not provide data for adults only; this may have led to the exclusion of some relevant data. Although our search spanned all regions, the included articles were conducted in just 16 countries and no included articles were from Africa; future research in countries and regions not represented in this review would further enhance the evidence base.

In conclusion, this review found that gender-affirming hormones are acceptable, relatively affordable, and may improve the health of trans and gender-diverse adults globally, including critical outcomes such as mental health, suicidality, and quality of life. These data support the continued provision of gender-affirming hormones in accordance with evidence-based clinical care guidelines.^49,50

Contributors

EEC, PTY, and CEK conceptualized the study and designed the protocol. PTY ran the search. EEC, PTY, and CEK oversaw screening and data extraction, assessment of inclusion, and quality of reporting. RPK, KSK, and BW reviewed articles for eligibility, and conducted data extraction and risk of bias assessment. EEC, PTY, and CEK synthesized the data, including generation of key figures and tables. EEC, PTY, and CEK drafted the manuscript. All authors reviewed the drafts, provided review, and approved the final manuscript. EEC, PTY, and CEK directly accessed and verified the underlying data reported in the manuscript.

Data sharing statement

Aggregated data relevant to the study are included in the article in Tables. Details on search terms and databases are available in Appendices. Data extraction sheets are available on reasonable request.

Editor note

eClinicalMedicine takes a neutral position with respect to territorial claims in published maps.

Declaration of interests

We declare no competing interests.