Gender Minority Stress, Resilience, and Psychological Distress: The Role of Resilience Among Transgender and Gender Diverse Youth

Abstract

Purpose:

Gender minority (GM) stress, resulting from distal (i.e., external) and proximal (i.e., internal) stigma-based stressors, is thought to drive mental health disparities among transgender and gender diverse (TGD) youth. We tested the gender minority stress and resilience (GMSR) model hypotheses that distal GM stress effects on mental health are partially mediated by proximal GM stress and moderated by GM-specific resilience (i.e., community connectedness, identity pride) among a U.S. national sample of TGD youth.

Methods:

As part of an HIV prevention study (NCT03185975), 159 TGD youth (ages 15–24) completed an online survey that included the GMSR measure, assessing distal and proximal GM stress and GM resilience, and the 18-item Brief Symptom Inventory, assessing past-7-day psychological distress. Three models linking GMSR constructs to psychological distress were tested using PROCESS v4.0: (1) simple partial mediation, (2) moderated partial mediation, and (3) serial partial mediation.

Results:

A direct effect of distal GM stress was observed in all models. An indirect effect through proximal GM stress alone was observed in model 1, but not models 2 or 3. In model 2, resilience did not moderate the effects of distal or proximal GM stress. In model 3, indirect effects were observed through proximal GM stress and GM resilience serially as well as GM resilience alone.

Conclusion:

Larger prospective studies are needed to confirm the role of GM resilience as a mediator, rather than moderator, of GM stress effects on mental health and a critical, rather than supplementary, target for mental health intervention among TGD youth.

Introduction

Transgender and gender diverse (TGD) individuals—those whose gender differs from their sex assigned at birth—experience mental health disparities across the life span relative to their cisgender peers.^1–4 TGD adults in the United States report experiencing serious psychological distress in the past month at a rate (39%) nearly eight times higher than the cisgender adult population, with those ages 18 to 25 years reporting the highest rate (53%).³ Similarly, TGD youth (ages 15–24) have been identified as a high-risk population for mental health problems. In their 2016 review, Connolly et al. found consistent reports of elevated depression, suicidality, and self-harm among TGD youth across school, community, and clinical settings in the United States and globally.¹

Such disparities in mental health and well-being among TGD youth are thought to be shaped by stressors specific to the lives of TGD individuals that arise from societal norms and stigma regarding gender identity and expression.⁴ The resultant stress is referred to as gender minority (GM) stress.

Gender minority stress and resilience model

The minority stress model posits three categories of unique factors impacting the mental health of individuals who belong to a stigmatized minority group: (1) distal stressors, or stigma-related external events and conditions occurring at the structural and interpersonal levels; (2) proximal stressors, or stigma-related internal processes and conditions occurring at the individual level; and (3) community-specific resilience factors, or those that mitigate the negative impacts of stigma-related stressors.^5,6 While this model was originally conceived to reflect the experiences of sexual minority people, the gender minority stress and resilience (GMSR) model has been proposed to reflect factors experienced by TGD people.⁴

Distal stressors presented in the GMSR model include GM stigma-related discrimination, rejection, and victimization, as well as nonaffirmation of gender.⁴ TGD people in the United States face discrimination in many contexts, such as employment, housing, and public accommodations.^7–12 Many face rejection from their families and communities driven by stigmatization of their gender identity or expression,^13–16 in addition to high levels of victimization.^{7,13,15,17–19} TGD people are also subjected to experiences of nonaffirmation—those in which their correct gender, name, and/or pronouns are not respected—at both the structural and interpersonal levels.^3,4 Distal stressors are believed to threaten the health of stigmatized minorities through combinations of direct physical and/or mental harm and exclusion from resources essential to positive health outcomes.^4,6

Proximal stressors presented in the GMSR model include internalized stigma, anticipated stigma, and nondisclosure.⁴ Internalized stigma has been conceptualized as existing in two directions among TGD individuals: vertical, or toward the self, believed to induce stress through negative self-appraisals, and horizontal, or toward the community, believed to induce stress through attitudes and behaviors that alienate an individual from their community.²⁰ Anticipated stigma—resulting from socialization and/or past experiences—is believed to induce stress among sexual minorities through the chronic and effortful employment of vigilance that it necessitates,⁶ a process thought to be shared by TGD people.⁴

Nondisclosure among TGD people refers to the concealment of one's gender identity or transition history, which may vary drastically based on one's desire and ability to socially and/or physically transition, progress in doing so, and relevant biological factors.⁴ Despite such variation in how nondisclosure might be experienced, concealment of a stigmatized identity is believed to induce stress through the cognitive burden of preoccupation with concealment itself.^6,21

Resilience factors presented in the GMSR model include community connectedness and pride, which are thought to benefit TGD and sexual minority people in similar ways.⁴ Community connectedness—the building and maintenance of supportive relationships and/or social networks with members of one's stigmatized community—is believed to mitigate minority stress by fostering social environments in which the individual can engage without having to anticipate or endure stigma as well as by providing access to group resources and social support.^6,22 Furthermore, individuals who have a strong sense of community connectedness are thought to shift their frame of reference, allowing for more positive appraisals of the self and of stressful conditions.^6,23

Cass defines identity pride as a strong sense of both value for one's identity and loyalty to one's identity group, members of which are viewed as important and credible.²⁴ While Cass's definition was made in reference to sexual minority identity development, this construct has been extended to TGD people.²⁵

Psychological mediation and the role of resilience

The GMSR model posits the effect of distal GM stress on mental health outcomes to be partially mediated by proximal GM stress.⁴ This hypothesized relationship is based on the psychological mediation framework of minority stress among sexual minorities,²⁶ conceptualizing proximal stressors as community-specific psychological processes that work in conjunction with nonspecific psychological processes to mediate the effects of distal stressors on health outcomes. The GMSR model also proposes that the effects of both distal and proximal GM stress are moderated by GM resilience. A growing number of cross-sectional studies provide evidence supporting the psychological mediation hypothesis regarding GM stress and mental health among TGD adults.^27–32 However, those investigating the role of GM resilience did not find support for moderation as posited by the GMSR model.^29,33,34

Only one known study investigating the associations between GMSR factors in relation to mental health outcomes has included TGD individuals younger than 18 years. Katz-Wise et al. analyzed longitudinal data, collected in five waves over a 2-year period, from a U.S. sample of 30 TGD adolescents 13 to 17 years of age.³⁵ They found that internalized stigma (a proximal GM stressor) mediated the effects of distal GM stress on depressive and anxious symptomology as well as substance use. Surprisingly, they also found in a separate model that identity pride (a GM resilience factor) mediated the effects of distal GM stress on substance use.

The present study

While the number of studies testing the GMSR model among TGD adults is growing, research in this area among TGD youth is extremely limited with no known studies having tested all associations posited by the GMSR model simultaneously. Thorough empirical investigation of these relationships is crucial to addressing mental health disparities experienced by TGD youth, as findings have the potential to inform our understanding of GM stress processes and critical targets for relevant intervention. Thus, we conducted mediation and moderation analyses to examine the associations between distal and proximal GM stress and GM resilience simultaneously in relation to the outcome of psychological distress using cross-sectional data from a U.S. national, online sample of TGD youth.

Based on the GMSR model,⁴ two initial hypotheses were proposed: (1) the association of distal GM stress with psychological distress will be partially mediated by proximal GM stress, and (2) GM resilience will serve to moderate this association. Based on the findings of Katz-Wise et al.,³⁵ a third, alternative hypothesis was also proposed: proximal GM stress and GM resilience will serve as serial mediators of the association between distal GM stress and psychological distress.

Methods

Recruitment

The present data were obtained from baseline survey responses of participants in Project Moxie (NCT03185975), a pilot trial testing the pairing of a Health Insurance Portability and Accountability Act (HIPAA) secure, online video-counseling intervention with at-home HIV testing for TGD youth.³⁶ A complete study protocol was approved by the University of Michigan Institutional Review Board (IRB; HUM00123412) and is described in detail in a prior publication.³⁷ Eligibility criteria for Project Moxie included the following: (1) self-identification as noncisgender, indicated by a current gender identity differing from assigned sex at birth; (2) aged 15 to 24 years; (3) negative or unknown HIV status; (4) current U.S. residency; (5) willingness to receive an at-home HIV self-testing kit; and (6) access to a computer, smartphone, or tablet that supports VSee, a HIPAA-secure videoconferencing service used to implement the intervention.

A detailed report of the Project Moxie recruitment and enrollment process is available elsewhere.³⁸ In summary, TGD youth were recruited from June 2017 to June 2018 via advertisements and postings on Facebook, Instagram, Twitter, Tumblr, and Craigslist, and via peer referral. Advertisements featured photographs representing a spectrum of TGD persons, and directed interested individuals to the study website to earn up to $150 for participating in a study of transgender health. The website landing page provided basic study information, including a short description of activities, and a link to screen for eligibility. A waiver of need for parental consent was approved by the IRB for the screening and enrollment of those younger than 18 years. Eligible individuals underwent a comprehensive online consent before participation and were asked to take a baseline survey upon enrollment. In total, 202 baseline survey responses were recorded.

The sample for the present study was restricted to 159 individuals who completed relevant measures included in the analyses (i.e., had no missing data). There were no appreciable differences in demographic or other variables of interest between the present sample and full Project Moxie sample (Supplementary Table S1).

Measures

Demographics

Self-reported age, race/ethnicity, sexual orientation, state of residence, education level, employment status, and history of homelessness were assessed. Data on respondents' assigned sex at birth were collected in addition to their current gender identity, which included options for male or trans masculine/male and female or trans feminine/female, as well as multiple options that were categorized as nonbinary. These included genderqueer/gender nonconforming, agender/gender fluid, and an open-ended response option. Respondents were also asked whether and to what extent they currently live as the gender that most affirms them. Those who endorsed doing so most or all of the time were considered socially transitioned.

Gender minority stress and resilience

Minority stressors and resilience factors specific to TGD individuals were assessed using the GMSR measure.⁴ The GMSR measure consists of nine scales psychometrically validated in an initial sample of 1414 TGD adults, each measuring a different GM stress or resilience factor: discrimination, rejection, victimization, nonaffirmation, internalized transphobia (i.e., internalized stigma), negative expectations (i.e., anticipated stigma), nondisclosure, pride, and community connectedness. All scales of the GMSR measure have since been validated among TGD youth ages 12 to 18.³⁹ For the present sample, data regarding nondisclosure were not available.

Each of the eight GMSR scales for which data were available demonstrated adequate internal consistency (α > 0.70) in the present sample. For analyses herein, these eight scales were consolidated to create composite scales representing distal GM stress (discrimination, rejection, victimization, and nonaffirmation), proximal GM stress (internalized stigma and anticipated stigma), and GM resilience (pride and community connectedness).^29,33 Composite scales were scored from 0 to 41, 68, and 48 with Cronbach's α coefficients of 0.87, 0.91, and 0.84, respectively. Previous analyses of GMSR measure response patterns by gender identity across the full Project Moxie sample at baseline revealed no significant differences at the scale level.⁴⁰ As such, stratification by gender identity was not used in the present study.

Psychological distress

Levels of psychological distress were assessed using the Brief Symptom Inventory-18 (BSI-18).⁴¹ The 18-item scale consists of three 6-item subscales, each measuring a different construct associated with psychological distress: somatization, depression, and anxiety. Each item asks the extent to which the participant has been troubled by a given symptom or feeling in the last 7 days, including today, on a five-point scale from 0 (not at all) to 4 (extremely). Scores on each item were summed for a total score ranging from 0 to 72, with strong internal consistency in the present sample (α=0.96). The 53-item BSI, from which the shortened BSI-18 was developed, was validated in adult (N=914) and adolescent (N=2408) nonpatient samples.⁴²

Evidence suggests that the BSI-18 is best used as a measure of overall psychological distress, rather than to assess somatization, depression, or anxiety.⁴³ As such, data are reported and interpreted for total scores only.

Data analysis

Statistical analyses were performed using IBM SPSS Statistics version 27.0. Correlation coefficients and their significance levels were measured for variables to be used as predictors in regression modeling. Mediation and moderation analyses were conducted using the macro program PROCESS v4.0.⁴⁴ Continuous variables included in moderation analyses were mean-centered. Percentile bootstrapping with 5000 repetitions was used to calculate standard errors (SEs) and 95% confidence intervals (CIs) for indices of mediation and moderated mediation. Estimates for which the 95% CI did not include zero (i.e., p<0.05) were considered statistically significant. Adjusted R² effect sizes were calculated and interpreted for each model as described by Cohen.⁴⁵ Partial eta-squared (η_p²) effect sizes were calculated for significant direct effects.

First, PROCESS model number 4 was used to test a simple partial mediation model in which proximal GM stress partially mediates the effect of distal GM stress on psychological distress. Next, PROCESS model number 15 was used to test a moderated partial mediation model in which GM resilience moderates the effects of distal GM stress, directly and indirectly through proximal GM stress, on psychological distress. Finally, PROCESS model number 6 was used to test a serial partial mediation model in which proximal GM stress and GM resilience serve as both serial and independent partial mediators of the effect of distal GM stress on psychological distress. Constitutive indirect effects of the serial mediation path were tested using PROCESS model 4. Assumptions of multivariate normality and homoscedasticity were assessed using Shapiro–Wilk and Breusch–Pagan tests, respectively.

Results

The present sample included 69 (43.4%) transgender men, 27 (17.0%) transgender women, and 63 (39.6%) nonbinary individuals, of whom 20 (12.6%) were assigned a male sex at birth and 43 (27.0%) were assigned a female sex at birth (Table 1). The sample ranged in age from 15 to 24 years, with a mean age of 19.2 years (standard deviation=2.65). Most had socially transitioned (98, 61.6%), identified as queer/pansexual (74, 46.5%) and non-Hispanic white (103, 64.8%), had a high school diploma or equivalent (113, 71.1%), were employed and/or a student (128, 80.5%), and did not report ever having experienced homelessness (132, 83.0%). Among the four Census-designated U.S. regions, sample residence was distributed as follows: 58 (36.5%) from the South, 45 (28.3%) from the Midwest, 33 (20.8%) from the West, and 23 (14.5%) from the Northeast.

Table 1.

Demographic Characteristics of the Sample

Characteristic	n	%
Gender
Transgender women	27	17.0
Transgender men	69	43.4
Nonbinary, assigned male at birtha	20	12.6
Nonbinary, assigned female at birthb	43	27.0
Socially transitioned	98	61.6
Sexual orientation
Homosexual/gay	24	15.1
Bisexual	36	22.6
Heterosexual/straight	9	5.7
Queer/pansexual	74	46.5
Otherc	16	10.1
Race/ethnicity
Non-Hispanic white	103	64.8
Hispanic white/Latinx	14	8.8
Black/African American	10	6.3
Asian/Pacific Islander	6	3.8
Middle Eastern	2	1.3
Native American	1	0.6
Mixed race/ethnicity	23	14.5
U.S. region
Northeast	23	14.5
Midwest	45	28.3
South	58	36.5
West	33	20.8
High school graduated	113	71.1
Unemployed, not a student	31	19.5
Ever been homeless	27	17.0

N=159. The mean age of participants was 19.2 years (range, 15–24).

^a Includes 16 genderqueer/gender nonconforming, 3 nonbinary, and 1 two-spirit.

^b Includes 23 genderqueer/gender nonconforming, 16 nonbinary, and 4 agender/genderfluid.

^c Includes 12 questioning/unsure, 2 asexual, 1 demisexual, and 1 sexually fluid.

^d Includes general educational development.

In preliminary analyses conducted to test the role of demographic covariates, their inclusion did not substantively impact results (Supplementary Table S2). As such, the final mediation models reported below did not include demographic covariates.

Means, standard deviations, and bivariate correlation coefficients for the three composite GMSR scales as well as BSI-18 scores of psychological distress are presented in Table 2. A significant positive correlation was found between distal and proximal GM stress scores, r=0.51, p<0.001, and a significant negative correlation was observed between proximal GM stress and GM resilience scores, r=−0.40, p<0.001. No statistically significant correlation was observed between distal GM stress and GM resilience scores. Significant bivariate correlations with BSI-18 scores were observed for distal GM stress, r=0.45, p<0.001, proximal GM stress, r=0.38, p<0.001, and GM resilience, r=−0.27, p<0.001.

Table 2.

Descriptive Statistics and Bivariate Correlations for Gender Minority Stress and Resilience Measure and 18-Item Brief Symptom Inventory Scales

Measure	Mean	Standard deviation	1	2	3	4
1. Distal GM stress	25.5	8.9	—
2. Proximal GM stress	36.7	14.9	0.51***	—
3. GM resilience	27.6	9.0	–0.02	–0.40***	—
4. Psychological distress	29.3	18.5	0.45***	0.38***	–0.27***	—

N=159.

^*** p<0.001.

GM, gender minority.

Results regarding the simple partial mediation model are presented in Figure 1. Distal GM stress explained a large portion of the variance in proximal GM stress, adjusted R²=0.26, F(1, 157)=56.54, p<0.001. A significant positive association between distal and proximal GM stress was observed, b=0.87, SE=0.12, 95% CI [0.64–1.09], p<0.001, η_p²=0.26.

FIG. 1.

Simple partial mediation model of gender minority stress and psychological distress. Solids arrows represent statistically significant paths. *p<0.05, ***p<0.001.

Distal and proximal GM stress jointly explained a medium-to-large portion of the variance in psychological distress, adjusted R²=0.22, F(2, 156)=23.61, p<0.001. Significant effects were observed for distal GM stress directly, b=0.72, SE=0.17, 95% CI [0.39–1.06], p<0.001, η_p²=0.10, proximal GM stress directly, b=0.25, SE=0.10, 95% CI [0.05–0.45], p=0.016, η_p²=0.04, and distal GM stress indirectly via proximal GM stress, b=0.21, bootstrap SE=0.10, 95% CI [0.04–0.43].

Results regarding the moderated partial mediation model are presented in Figure 2. Distal and proximal GM stress, GM resilience, and their hypothesized interaction terms jointly explained a large portion of the variance in psychological distress, adjusted R²=0.26, F(5, 153)=11.90, p<0.001. Moderation effects were not statistically significant for interactions involving GM resilience and distal, b=−0.01, SE=0.02, 95% CI [−0.05 to 0.02], p=0.528, or proximal GM stress, b=−0.01, SE=0.01, 95% CI [−0.03 to 0.01], p=0.552. However, a significant main effect was observed for GM resilience directly, b=−0.48, SE=0.16, 95% CI [−0.79 to −0.16], p=0.003, η_p²=0.06.

FIG. 2.

Moderated partial mediation model of gender minority stress and resilience and psychological distress. Solid arrows represent statistically significant paths, and dashed arrows represent nonsignificant paths. **p<0.01, ***p<0.001.

A significant main effect was also observed for distal GM stress directly, b=0.83, SE=0.17, 95% CI [0.48–1.17], p<0.001, η_p²=0.13, but not proximal GM stress directly, b=0.09, SE=0.11, 95% CI [−0.13 to 0.31], p=0.437. The index of moderated mediation was not statistically significant in this model, b=−0.01, bootstrap SE=0.01, 95% CI [−0.02 to 0.01].

Results regarding the serial partial mediation model are presented in Figure 3. Distal and proximal GM stress jointly explained a medium-to-large portion of the variance in GM resilience, adjusted R²=0.20, F(2, 156)=20.46, p<0.001. Significant associations were observed between GM resilience and distal, b=0.25, SE=0.08, 95% CI [0.09–0.42], p=0.003, η_p²=0.06, as well as proximal GM stress, b=−0.32, SE=0.05, 95% CI [−0.42 to −0.22], p<0.001, η_p²=0.21. A significant indirect effect of distal GM stress via proximal GM stress was observed, b=−0.28, bootstrap SE=0.06, 95% CI [−0.40 to −0.17].

FIG. 3.

Serial partial mediation model of gender minority stress and resilience and psychological distress. Solid arrows represent statistically significant paths, and dashed arrows represent nonsignificant paths. *p<0.05, **p<0.01, ***p<0.001.

Distal and proximal GM stress and GM resilience jointly explained a large portion of the variance in psychological distress, adjusted R²=0.26, F(3, 155)=19.39, p<0.001. Significant direct effects were observed for distal GM stress, b=0.84, SE=0.17, 95% CI [0.50–1.18], p<0.001, η_p²=0.14, and GM resilience, b=−0.47, SE=0.16, 95% CI [−0.78 to −0.16], p=0.004, η_p²=0.05, but not proximal GM stress, b=0.10, SE=0.11, 95% CI [−0.12 to 0.32], p=0.384. Significant indirect effects were observed for distal GM stress via proximal GM stress and GM resilience serially, b=0.13, bootstrap SE=0.06, 95% CI [0.04–0.25], and GM resilience alone, b=−0.12, bootstrap SE=0.05, 95% CI [−0.23 to −0.02], but not proximal GM stress alone, b=0.08, bootstrap SE=0.10, 95% CI [−0.12 to 0.29].

The indirect effect of proximal GM stress via GM resilience, a portion of the observed serial mediation path, was itself statistically significant, b=0.15, bootstrap SE=0.06, 95% CI [0.04–0.27].

Discussion

The GMSR model posits that the effect of distal GM stress on mental health outcomes is partially mediated by proximal GM stress, and that both effects are moderated by GM resilience.⁴ However, our results suggest that GM resilience may be best conceptualized as a partial mediator of the effect of distal GM stress on psychological distress among TGD youth. Furthermore, our results suggest that proximal GM stress is not associated with psychological distress directly but indirectly through GM resilience. These findings are consistent with those of Katz-Wise et al., who found that internalized stigma (a proximal GM stressor) and identity pride (a GM resilience factor) separately mediated the effects of distal stress on mental health outcomes longitudinally among TGD adolescents ages 13 to 17.³⁵

In addition, our results are consistent with those of Jäggi et al. who did not find the contemporaneous effects of distal GM stress on depressive symptoms, directly or indirectly through proximal GM stress, to be moderated by GM resilience among TGD adults.²⁹

Our findings suggest that GM resilience may serve to suppress the effects of distal GM stress—that is, the stress-buffering effects of community connectedness and identity pride may be activated by distal GM stress.^22,46 As the oppressive social construction of binary gender forms the normative reference from which TGD individuals are perceived to deviate and thus forced to differentiate their identity,²⁵ it is plausible that such circumstances too create the foundation for connection and pride around these identities. Furthermore, TGD youth who are subjected to higher levels of distal GM stress may reside in social ecologies (e.g., families, neighborhoods) where GM stigma is more prevalent, and therefore, connection to community is more necessary for support and protection. In addition, those who have been most impacted by distal GM stigma may be more motivated to participate in social and political action through which they build community connections and foster identity pride.

Our findings also suggest that greater levels of proximal GM stress may serve to inhibit the accumulation of GM resilience. In this way, proximal GM stress appears to contribute indirectly to psychological distress by rendering GM resilience factors less available to suppress the deleterious effects of distal GM stress. This process could be driven by social isolation, resulting from anticipated and horizontal internalized stigma, as well as low levels of self-esteem, resulting from vertical internalized stigma.²⁰ This process may also serve to obscure the relationship between distal GM stress and GM resilience at the bivariate level, as distal GM stress demonstrated both a positive association with GM resilience directly, independent of proximal GM stress, and a negative association with GM resilience indirectly via proximal GM stress.

Limitations, strengths, and future research

Interpretation of the present findings may be limited by the study design. First, due to the cross-sectional nature of the present study, causality cannot be established for any of the relationships discussed. Second, given the small effect sizes typically observed in moderation, this study may not have been adequately powered to detect interaction effects of GM resilience.

Third, our operationalization of GMSR constructs may not have achieved full coverage of important GMSR processes. Although the associations posited by the GMSR model are proposed as consistent across variables in each domain (i.e., distal GM stressors, proximal GM stressors, GM resilience factors),⁴ the omission of the nondisclosure scale of the GMSR measure from this study may constitute an influential lapse in data collection on proximal GM stress. In addition, the included community connectedness scale of the GMSR measure assesses emotional connectedness to the TGD community but not behavioral participation, both of which have been identified as important forms of GM resilience.⁴⁷ Fourth, our study focused on constructs included in the GMSR model and did not include general psychological (e.g., coping) or resilience (e.g., social support) constructs also hypothesized to influence minority stress processes.²⁶

Generalizability of the present findings may be limited given the study sample. Our study included TGD youth in the United States recruited online who had access to a smartphone, tablet, or computer as well as were eligible and willing to participate in a paid pilot trial (Project Moxie) of a video-counseling-based HIV testing intervention and receive an at-home HIV testing kit. Furthermore, Project Moxie participants identified predominantly as non-Hispanic white and 21% were excluded from the complete case analysis. Although the present sample showed no appreciable differences from the full Project Moxie sample, it may not be fully representative of TGD youth in the United States, particularly racial/ethnic minority TGD youth and those with limited internet access or who are otherwise unwilling, unable, or ineligible to participate in paid HIV prevention research.

Our study also has notable strengths. It is among the first to test all relationships posited by the GMSR model simultaneously in a sizeable sample of TGD youth in the United States. We used data on psychometrically validated measures of GMSR constructs and psychological distress, analyzed with well-established cross-sectional methods. Our observed effect sizes were not only statistically but also practically significant, explaining large portions of variance in psychological distress. Given the dearth of research testing the GMSR model among TGD youth, our study provides important preliminary findings to be tested in further research using prospective designs with larger sample sizes.

We recommend that those conducting such research explicitly test GM resilience as both a moderator and mediator of GM stress effects on mental health outcomes among TGD youth, including racial/ethnic minority TGD youth and those with limited internet access. We also recommend that such investigations include all constructs of the GMSR model, as well as general psychological and resilience constructs whenever possible.

Clinical and policy implications

Our findings, although preliminary, have implications for clinical and policy decisions regarding TGD youth. Interventions aiming to improve psychological well-being among TGD youth may be best positioned to do so by emphasizing adaptive and feasible strategies for minimizing exposure to distal GM stressors. While clinicians may find continued success in addressing proximal GM stressors among TGD youth for whom this is indicated (e.g., cognitive restructuring to address internalized stigma), such interventions may only prove beneficial in so much as they also serve to foster youth's pride and connection to their community. At the same time, resilience in these forms can serve only so well in protecting TGD youth from the deleterious effects of oppression.

Structural changes must be made to adequately address the disparities in mental well-being faced by TGD youth in the United States. Policies prohibiting the oppression of TGD people need to be enacted by institutions and governments at the local, regional, and national levels, while those placing unnecessary and unjust burden on TGD individuals of all ages need to be repealed. In not making such changes, our society will continue to directly and indirectly subject TGD youth to stressors of the very nature presently identified as significantly related to their psychological distress.

Conclusion

This study aimed to test the associations between the three major constructs of the GMSR model simultaneously in relation to the outcome of psychological distress in a U.S. national, online sample of TGD youth. We found that GM resilience mediates, rather than moderates, the effects of distal and proximal GM stress on psychological distress, with results suggesting that GM resilience may serve to suppress the effects of distal GM stress on mental health while proximal GM stress may serve to inhibit the accumulation of GM resilience. Although requiring replication in future research, our findings have important implications for clinical and policy decisions: TGD youth could benefit from interventions incorporating GM resilience as a critical mental health treatment target, though structural changes must be made to fully address the distal GM stressors that appear to ultimately drive their psychological distress.

Abbreviations Used

BSI-18

Brief Symptom Inventory-18

CIs

confidence intervals

GM

gender minority

GMSR

gender minority stress and resilience

HIPAA

Health Insurance Portability and Accountability Act

IRB

institutional review board

SE

standard error

TGD

transgender and gender diverse

Disclaimer

The content of this publication is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Authors' Contributions

M.M.-P.: Writing—original draft, conceptualization, methodology, formal analysis, and investigation. K.J.H.: Writing—review and editing and supervision. E.K.: Writing—review and editing and supervision. R.S.: Writing—review and editing, supervision, resources, and funding acquisition.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health under award number R01HD078131.

Cite this article as: Miller-Perusse M, Horvath KJ, Kahle E, Stephenson R (2023) Gender minority stress, resilience, and psychological distress: the role of resilience among transgender and gender diverse youth, Transgender Health 9:4, 307–316, DOI: 10.1089/trgh.2022.0117.