Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Sep 29.
Published before final editing as: Psychol Sex Orientat Gend Divers. 2026 Sep 28:10.1037/sgd0000968. doi: 10.1037/sgd0000968

Developmental Asset Profiles as Moderators of the Association Between Intersectional Discrimination and Mental Health Among Black Sexual Minority Men

Donte T Boyd 1, Addie Weaver 2, Xin Feng 3, Hans Oh 4
PMCID: PMC13618877  NIHMSID: NIHMS2206776  PMID: 42807736

Abstract

This cross-sectional study examined whether developmental assets moderate the relationship between intersectional discrimination and mental health among Black sexual minority men (BSMM) aged 14 to 24. BSMM (N = 541; M age = 17.00, SD = 2.85) residing in three Midwestern cities completed an online Qualtrics survey conducted between December 2023 and January 2024. The mental health outcomes assessed included depressive symptoms, anxiety, and suicidal behaviors. Intersectional discrimination was modeled as a latent factor characterized by three indicators: anticipated, day-to-day, and major discrimination. Eight developmental asset indicators (both external and internal) informed a latent profile analysis (LPA); moderation was examined using the Bolck–Croon–Hagenaars approach in Mplus, while controlling for age, household income, homelessness, and sexual identity disclosure. The LPA revealed a four-class solution: low assets (25.5%), moderate assets (30.5%), inconsistent assets (27.4%), and high assets (16.6%). Discrimination was positively associated with depressive symptoms across all profiles, with the exception of the high assets group; the strongest effect was noted in the inconsistent assets category, while the weakest was in the high assets group. Regarding anxiety, discrimination was positively correlated with low, moderate, and inconsistent assets but negatively associated with high assets. Additionally, discrimination was positively related to suicidal behavior across all profiles, with the strongest effect observed in the moderate assets group. Younger age and housing instability were consistently linked to poorer mental health outcomes. These findings suggest that developmental assets can buffer some—but not all—of the mental health impacts of discrimination, suggesting that uneven asset patterns may heighten risk.

Keywords: Black sexual minority men, developmental assets, intersectional discrimination, mental health, latent profile analysis

Black sexual minority men (BSMM) between the ages of 14 and 24 experience pervasive discrimination that adversely impacts their mental health and overall well-being, leading to issues such as depressive symptoms, anxiety, and suicidal behavior (Boyd et al., 2025a; Boyd et al., 2025b; Boyd et al., 2026). These individuals often encounter “double jeopardy,” where racism and homophobia intersect, contributing to stigmatization and exclusion from both racial and LGBTQ+ communities (Pachankis et al., 2014). This dual marginalization can negatively impact mental health and restrict access to essential mental health resources and support networks (Bostwick et al., 2014; Marciano & Antebi-Gruszka, 2022; Vargas et al., 2020). Existing research underscores the need for targeted approaches that address the unique stressors experienced by BSMM while promoting resilience (Boyd et al., 2025a). In particular, strengths-based approaches may cultivate internal, familial, and developmental assets that assist BSMM in managing stress related to discrimination and alleviating adverse mental health outcomes (Boyd et al., 2025b; Boyd et al., 2026; Boyd et al., 2021). However, there is a paucity of research empirically examining developmental assets as protective factors for BSMM. This study aims to address this gap by investigating whether latent developmental asset profiles moderate the association between intersectional discrimination and mental health outcomes in BSMM aged 14 to 24.

BSMM and Mental Health

Research documents a substantial burden of depressive symptoms, anxiety, suicidal ideation, and suicide attempts among BSMM, with risk varying across developmental and psychosocial profiles (Boyd et al., 2025a; Boyd et al., 2025b; Boyd et al., 2026). Comparative research further suggests that BSMM experience poorer mental health outcomes compared to their heterosexual counterparts and certain other sexual and gender minority groups (Bostwick et al., 2014; Lucassen et al., 2017). These disparities are compounded by discrimination based on race and sexual orientation, alongside structural racism and anti-LGBTQ+ policies (Jackson et al., 2023; Layland et al., 2025). The minority stress model elucidates these associations by proposing that marginalized individuals experience unique and chronic stressors that create hostile social environments and contribute to mental health disparities (Szymanski & Ikizler, 2013; Zelaya et al., 2021). For BSMM, the intersection of racial and sexual minority marginalization may intensify feelings of isolation and despair. Nevertheless, research also identifies important protective resources. Family support is associated with fewer depressive symptoms among Black and Latinx sexual minority men (Boyd et al., 2021), while internal and broader developmental assets may protect against depressive symptoms and suicidal behaviors among BSMM (Boyd et al., 2025b; Boyd et al., 2026).

Additionally, BSMM are susceptible to anxiety, with symptoms frequently exacerbated by racial discrimination, sexuality-related stigma, internalized homonegativity, and the ongoing demands of navigating intersecting racial and sexual identities (Amola & Grimmett, 2015; Choi et al., 2013; Graham et al., 2011; Quinn et al., 2015; Wade et al., 2022). Perceived discrimination during adolescence and young adulthood is linked to anxiety in Black men (Assari et al., 2017), and the interplay of racism and homophobia may further compound stress and uncertainty (Assari et al., 2018). As BSMM become increasingly aware of societal perceptions and expectations related to racial identity and sexual orientation, as well as pressures to conform to racial and gender norms, they may experience heightened levels of internalized anxiety (Assari et al., 2018; Boehmer et al., 2021; DeLaney et al., 2022). Addressing these anxiety-related challenges is essential for improving the mental health of BSMM.

Finally, BSMM face a substantial suicide-related burden, with studies documenting high levels of suicidal thoughts and behaviors shaped by structural oppression, depression, internalized homophobia, family-related factors, and limited access to protective developmental assets (Boyd et al., 2025a; Boyd et al., 2025b; Boyd et al., 2026; English et al., 2022). The increased risk for suicidal ideation and suicide attempts may stem from the cumulative effects of racism, homophobia, and internalized stigma (Boyd et al., 2025a; Lytle et al., 2015; Vargas et al., 2020). Additionally, exposure to violence, bullying, and anti-LGBTQ+ victimization may further amplify this risk (Barnett et al., 2019). Conversely, family support is associated with fewer depressive symptoms among young Black and Latinx sexual minority men (Boyd et al., 2021) and a decrease in the severity of suicidal behaviors among BSMM (Boyd et al., 2026). Related research on Black youth experiencing discrimination similarly identifies social connectedness as a protective factor against depression and suicidal ideation (Smith et al., 2026). Collectively, these findings clarify that the absence or insufficiency of family, social, and developmental assets—not their presence—may increase the vulnerability to suicidality among BSMM.

Structural Oppression and Dual Marginalization

Literature suggests that BSMM experience intersectional minority stress due to structural oppression rooted in both racism and homophobia (Bostwick et al., 2014; McConnell et al., 2018; Vargas et al., 2020). Structural racism, racial discrimination, community violence, and limited access to culturally responsive mental health services can compound the effects of homophobia and stigma (Assari & Caldwell, 2018; Jackson et al., 2023). Consequently, Black LGBTQ+ youth may experience elevated psychological distress and fewer pathways to affirming care.

Intersectional minority stress and dual marginalization can drive social stigmatization and exclusion (Pachankis et al., 2014), which in turn reinforces psychological distress among BSMM (Boehmer et al., 2021). Internalized stigma may heighten rejection sensitivity and vigilance for prejudice (McConnell et al., 2018), while pressure to conform to traditional masculinity norms may suppress the expression of sexual identity and deepen distress (Almazan, 2019). Factors such as nondisclosure, insufficient community support, and family stigma may further restrict access to affirming mental health care and elevate suicide risk (Friedman et al., 2018; McConnell et al., 2018). Collectively, these intersecting systems of oppression highlight the layered vulnerabilities faced by BSMM. The Developmental Assets Framework offers a strengths-based lens for examining whether supportive relationships, community engagement, and positive identity development can buffer the psychological effects of intersectional minority stress.

Developmental Assets Framework (DAF)

The Developmental Assets Framework (DAF) is a strengths-based approach to positive youth development developed by the Search Institute in the 1990s. The framework aligns with resilience-based approaches to understanding adolescent development and identifies strengths that can support adaptation (Benson et al., 2011). The DAF defines developmental assets as interrelated experiences, relationships, skills, and values that enhance youth outcomes, including mental well-being. The framework organizes 40 developmental assets into internal and external domains: 20 internal assets represent the characteristics, beliefs, values, and skills of young individuals, while 20 external assets pertain to the experiences, relationships, opportunities, and support that families, schools, and communities provide (Benson et al., 2011).

Existing research has established a significant association between developmental assets and mental health, indicating that a greater number of assets is linked to improved outcomes for youth (Garcia et al., 2019; Rose et al., 2017; Wiium et al., 2021). Prior work has assessed developmental assets among LGBTQ+ youth and young adults more broadly (DelFerro et al., 2024; Fernandes et al., 2023; Russell & Fish, 2019; Toomey et al., 2019). A small but growing body of literature also indicates that developmental assets are associated with positive development and mental health among BSMM (Boyd et al., 2025a; Boyd et al., 2025b; Boyd et al., 2026).

External assets, particularly natural mentorship and family support, may be vital for LGBTQ+ youth as they navigate identity acceptance (Burningham & Weiler, 2021; DelFerro et al., 2024). Mentoring relationships can offer critical social support as LGBTQ+ youth address their identities and mental health needs (Burningham & Weiler, 2021). Familial acceptance and supportive adult relationships are also associated with fewer mental health disparities among LGBTQ+ youth, including BSMM (DelFerro et al., 2024). However, anti-LGBTQ+ policies and limited access to appropriate support services, particularly in schools, may exacerbate mental health challenges (Moran et al., 2024; Roulston et al., 2023).

An expanding body of literature suggests that a positive identity is a key internal asset for the mental health and well-being of LGBTQ+ youth. Those who report a more positive identity also tend to experience fewer mental health needs, exhibit increased help-seeking behaviors, and demonstrate stronger coping skills (Fernandes et al., 2023). Research centered on BSMM reveals that positive identity and other internal assets are associated with fewer depressive symptoms and suicidal behaviors (Boyd et al., 2025b; Boyd et al., 2026). Together, these findings support the examination of internal and external assets as potential protective resources for BSMM.

Although the understanding of the relationship between developmental assets and the mental health of LGBTQ+ youth has advanced, further research is needed to adapt and evaluate the DAF for BSMM, who face compounded vulnerabilities related to their intersecting identities (Bostwick et al., 2014). The present study addresses this gap by examining the patterns of internal and external assets and their potential to moderate discrimination-related mental health risk.

Study Purpose

This study examines whether developmental assets, assessed via latent developmental asset profiles, influence the relationship between intersectional discrimination and mental health outcomes—specifically depressive symptoms, anxiety symptoms, and suicidal behavior—among BSMM aged 14 to 24.

Research Question and Hypotheses

Research Question

Do latent asset profiles moderate the association between intersectional discrimination and mental health outcomes in BSMM aged 14 to 24?

Hypotheses

  1. 1. Hypothesis 1: Higher levels of intersectional discrimination will be associated with greater depressive symptoms, with the weakest relationship observed in individuals with high levels of developmental assets.

  2. 2. Hypothesis 2: Higher levels of intersectional discrimination will be associated with greater anxiety symptoms, with the weakest relationship observed in individuals with high levels of developmental assets.

  3. 3. Hypothesis 3: Higher levels of intersectional discrimination will be associated with increased suicidal behavior, with the weakest relationship observed in individuals with high levels of developmental assets.

Methods

Study Procedures and Recruitment

This cross-sectional study examined the influence of developmental assets on the sexual, physical, and mental health of BSMM aged 14 to 24 residing in Columbus, Ohio; Detroit, Michigan; and St. Louis, Missouri. Participants were recruited between December 1, 2023, and January 31, 2024, utilizing a multimodal, community-engaged approach. Community engagement was operationalized through active recruitment and survey-access partnerships: the research team collaborated with community-based organizations and schools serving Black youth to distribute tailored study materials; community health workers shared recruitment information with eligible clients; and partner organizations provided computers or tablets for on-site survey completion. Additionally, targeted social media advertisements on Facebook and X (formerly Twitter) offered an additional recruitment pathway.

All participants were required to confirm their residence in one of the three study cities, which was verified through ZIP code validation procedures. Participants recruited via social media completed the survey on personal devices, while those recruited through community-based settings used computers or tablets provided on-site. This targeted, convenience-based sampling strategy facilitated the efficient recruitment of a hard-to-reach population while maintaining alignment with the defined geographic scope. Although the sample is not intended to be nationally representative, the use of multiple recruitment channels and community-based partnerships enhances the diversity and validity of the sample for addressing the study aims.

Participants aged 18 to 24 provided informed consent, while those under 18 were required to give assent. The Institutional Review Board (IRB) granted a waiver of parental consent due to the minimal risk and sensitive nature of the study, thereby protecting the confidentiality of minors who may not have disclosed their sexual orientation to a parent or guardian. All study procedures were approved by The Ohio State University IRB.

Eligible respondents were those who self-identified as Black or African American, aged between 14 and 24, resided in Columbus, Ohio; Detroit, Michigan; or St. Louis, Missouri, were assigned male at birth and currently identified as male, reported fluency in English, and reported having engaged in sexual contact with a male partner within the past year. Individuals who did not meet these eligibility criteria were automatically excluded from the survey. Eligible participants completed measures assessing demographic characteristics, developmental assets, and mental health. To minimize the likelihood of missing data, forced-response options were used; however, participants had the option to select “Prefer not to answer” or withdraw from participation at any point. The survey required approximately 20 minutes to complete, and participants received a $35 electronic Visa gift card as compensation.

To ensure data integrity and minimize fraudulent or inattentive responding, multiple safeguards were implemented within Qualtrics. Survey protection features were used to verify U.S.-based IP addresses and prevent duplicate submissions (e.g., ballot box stuffing via cookies), and reCAPTCHA was employed to confirm human participation. Participants were restricted to completing the survey only once. Data quality procedures included embedded attention check items, with respondents failing these checks being excluded, as well as response time screening, which removed cases completed in less than one-third of the median duration. ZIP codes were cross-validated using Qualtrics geolocation tools to confirm U.S. residency, and cases with inconsistent or invalid geographic data were excluded. Additional checks for duplicate entries (e.g., IP address and device identifiers) and internal response inconsistencies were conducted. Cases failing one or more quality control criteria were removed prior to analysis. In total, 800 responses were collected, of which 259 were excluded for failing one or more data quality criteria, resulting in a final analytic sample of 541 participants included in all primary analyses.

Measures

Mental Health Outcomes

Depressive Symptoms.

The Center for Epidemiologic Studies Depression Scale-10 (CES-D-10; Andresen et al., 1994) assessed depressive symptoms over the past week. Respondents rated items such as “I felt depressed” on a scale from 0 (Rarely/None of the time) to 3 (All of the time), with two items being reverse coded. Scores ranged from 0 to 30, with higher scores indicating more severe symptoms. The CES-D-10 demonstrated acceptable reliability (α = .77).

Anxiety.

The Generalized Anxiety Disorder-7 (GAD-7; Spitzer et al., 2006) measured anxiety symptoms over the past two weeks. Items (e.g., “feeling nervous, anxious, or on edge”) were rated from 0 (Not at all) to 3 (Nearly every day), with total scores ranging from 0 to 21. The reliability of this measure was strong (α = .88).

Suicidality.

Suicidal behaviors were measured using a self-report adaptation of the Columbia-Suicide Severity Rating Scale (C-SSRS; Posner et al., 2011). Four dichotomous items evaluated suicide attempts and related behaviors within the past three months and throughout the lifetime. Total scores reflected the number of behaviors, and attempts were recorded as a continuous variable when endorsed. The internal consistency was high (α = .87). All participants received crisis resource information, including the 988 Suicide and Crisis Lifeline.

Intersectional Discrimination

Intersectional discrimination was measured using the Intersectional Discrimination Index (Scheim & Bauer, 2019). This measure consists of three subscales: anticipated discrimination (nine items; e.g., “Because of who I am, I might have trouble finding or keeping a job”), day-to-day discrimination (10 items; e.g., “Have you been treated as if others are afraid of you?”), and major discrimination (13 items; e.g., “Because of who you are, have you ever been evicted or denied housing?”). The construct validity and test-retest reliability have been established (Scheim & Bauer, 2019). Our data also demonstrated high internal consistency for the three scales (α = .92, .88, and .91, respectively). Mean scores for the subscales were used as indicators for the latent discrimination variable.

Developmental Assets

Developmental assets were examined using four external and four internal subscales, which were derived from previously validated measures known for their reliability and construct validity (Syvertsen et al., 2021). These measures were selected due to their robust theoretical grounding and prior application across diverse adolescent populations. Although they have not been specifically validated among BSMM, they capture broadly relevant constructs (e.g., social support, identity, and internal strengths) that align with the study’s conceptual framework.

External Assets

Support.

The assessment of support involved examining the extent to which youth perceived consistent encouragement, care, and guidance from family, friends, and other significant adults. Four items included statements like “I have family or friends who really care about me” and “There is an adult I can turn to when I need help.” Items were rated from 1 (Strongly disagree) to 4 (Strongly agree). A mean composite score was calculated, with higher scores indicating stronger perceived support. Internal consistency was found to be excellent (Cronbach’s α = .91).

Extracurricular Activities.

This external asset measured the extent to which youth engaged in structured, prosocial activities, including sports, clubs, community programs, faith-based groups, arts, or service organizations. Participation was coded dichotomously as 1 (Yes) or 0 (No), and the items were summed, with higher scores reflecting greater engagement in structured extracurricular contexts.

Mattering and Belonging.

Mattering and belonging were assessed using the Caring School Climate (three items) and Community Values (seven items) subscales. Participants responded to items such as “My teachers really care about me” on a five-point Likert scale. Because both subscales assess perceptions of mattering and belonging in supportive environments, the items were combined into a 10-item composite. The internal consistency was strong (α = .85).

Boundaries.

Boundaries were assessed with seven items across family (three items), school (three items), and neighborhood (one item) contexts. Items included “In my family, there are clear rules about what I can and cannot do.” Responses were rated from 1 (Strongly disagree) to 5 (Strongly agree). The internal consistency of this measure was robust (α = .88).

Internal Assets

Positive Identity.

Positive identity was assessed using a six-item scale that included hope (personal power, sense of purpose, and a positive view of the future) and self-esteem subcategories. Sample items included “I feel good about myself” and “I have a sense of purpose in my life.” Items were rated on a scale from 1 (Strongly disagree) to 5 (Strongly agree), with higher scores indicating a more positive identity. The internal consistency was found to be adequate (α = .71).

Positive Values.

Positive values were evaluated using a 10-item scale with subcategories of caring (three items), social justice (two items), integrity (two items), and responsibility (three items). Example items included “Helping other people is important to me” and “I try to stand up for what is right.” Items were rated from 1 (Strongly disagree) to 5 (Strongly agree), with higher scores indicating stronger endorsement of positive values. The internal consistency was strong (α = .89).

Academic Engagement.

Academic engagement was measured with five items rated from 1 (Strongly disagree) to 5 (Strongly agree), with higher scores indicating greater engagement. Sample items included “It bothers me when I don’t do something well.” The internal consistency for this measure was adequate (α = .72).

Social Competencies.

Social competencies were assessed using eight items across social-emotional skills (four items) and planning/decision-making skills (four items). Sample items included “Being good at making and keeping friends” and “Thinking through the consequences of my decisions before acting.” Items were rated from 1 (Strongly disagree) to 5 (Strongly agree), with higher scores indicating stronger social competencies. The internal consistency was strong for this measure (α = .88).

Sociodemographic Characteristics

Covariates included in the analysis were age, household income, food insecurity, housing insecurity, homelessness, and disclosure of sexuality. Age and household income were treated as continuous variables. Food insecurity, housing insecurity, and homelessness were assessed using dichotomous items (0 = no, 1 = yes) indicating experiences over the past 12 months. The disclosure of sexuality was also measured with a dichotomous item (0 = no, 1 = yes).

Data Analysis

Descriptive statistics were calculated to examine sample characteristics and assess assumptions for the latent profile and moderation analyses. All analyses were conducted using Mplus version 8.11 (Muthén et al., 2017).

First, latent profile analysis (LPA) was employed to identify subgroups of BSMM based on developmental asset patterns. LPA is a person-centered analytical approach that assigns individuals to their most likely profiles based on posterior probabilities (Nagin & Odgers, 2010). By focusing on patterns of co-occurring and interrelated characteristics within individuals, LPA identifies meaningful configurations of traits that may help explain variations in outcomes (Henry et al., 2005). The goal of LPA is to determine the minimal number of classes that adequately capture the distinct patterns of association among the indicator variables (Muthén, 2001). Model selection was guided by a combination of statistical fit indexes, parameter estimates, and the theoretical interpretability of the solution (Nylund et al., 2007). The Bayesian Information Criterion (BIC) and sample-size-adjusted BIC (A-BIC) were examined, with lower values indicating better model fit. The Vuong–Lo–Mendell–Rubin likelihood ratio test (VLMR-LRT; Vuong, 1989) and the Lo–Mendell–Rubin likelihood ratio test (LMR-LRT; Lo et al., 2001) tested whether a model with k classes fit significantly better than a model with k –1 classes. Additionally, we considered entropy (Jedidi et al., 1993) to evaluate classification accuracy, the conceptual clarity of the identified profiles (Muthén, 2001), and the share of the sample in each class (Jung & Wickrama, 2008) when determining the final model.

Second, we examined whether latent asset profiles moderated the association between intersectional discrimination and mental health outcomes using the Bolck–Croon–Hagenaars (BCH) method (Vermunt, 2017). This method utilizes posterior probabilities to assign individuals into latent classes and computes class-specific weights for comparing regression coefficients across profiles. The BCH approach incorporates classification uncertainty, thereby reducing bias associated with misclassification. A graphical depiction of the analysis is provided in Figure 1. Three separate models were estimated, one for each outcome (depressive symptoms, anxiety symptoms, and suicidal behavior), with covariates included. Within each model, the association between intersectional discrimination and the outcome was estimated within each class, and the class-specific associations were compared. Notable difference between classes indicated moderation.

Figure 1.

Figure 1

Open in a new tab

Latent Class Analysis with Distal Outcomes and a Moderator

Note. ID-A = anticipated discrimination; ID-D = day-to-day discrimination; ID-M = major discrimination.

As depicted in Figure 1, intersectional discrimination was modeled as a latent variable with three indicators: anticipated discrimination, day-to-day discrimination, and major discrimination. A confirmatory factor analysis (CFA) was conducted to specify these three indicators as loading onto the latent construct of intersectional discrimination. Given that a three-indicator CFA is just identified, the global model fit cannot be evaluated empirically; the model precisely reproduces the observed covariance matrix. Therefore, we evaluated the magnitude, statistical significance, and theoretical coherence of the factor loadings.

The analyses encompassed the entire sample of 541 participants. Ten participants (1.8%) had missing data on covariates; thus, the analytic sample consisted of 531 participants. Considering the small proportion of missing data, the bias resulting from listwise deletion was deemed negligible (Allison, 2001; Schafer & Graham, 2002).

Results

Sample Characteristics

A total of 541 participants completed the survey, yielding a mean age of 17.00 years (SD = 2.85). The majority of participants (85%; n = 460) self-identified as gay. Over one quarter (27%; n = 145) reported a household income between $40,000 and $74,999. A significant portion of participants (79%; n = 427) were enrolled in school, with 52% (n = 281) reporting they were in the 12th grade. In the previous 12 months, 51% (n = 275) experienced food insecurity and 40% (n = 216) experienced housing insecurity (see Table 1).

Table 1.

Participant Demographic Characteristics

Characteristic n % M SD
Age (years) — — 17.00 2.85
Sexual Identity
 Gay 460 85.0 — —
 Other (bisexual, queer, questioning, or other) 81 15.0 — —
Household Income
 $0–$19,999 15 3.00 — —
 $20,000–$39,999 55 10.22 — —
 $40,000–$74,999 145 27.00 — —
 $75,000–$99,999 62 11.52 — —
 $100,000–$124,999 196 36.43 — —
 $125,000–$149,999 30 5.58 — —
 $150,000 or more 35 6.51 — —
School Enrollment
 Enrolled 427 79.0 — —
 Not enrolled 114 21.0 — —
Current Grade Status
 12th grade 281 52.0 — —
 Other grade or not enrolled 260 48.0 — —
Food Insecurity,(Previous 12 Months) 275 51.0 — —
Housing Insecurity, (Previous 12 Months) 216 40.0 — —
Open in a new tab

Note. Household income data were available for 538 participants. Percentages are based on participants with available income data and may not total 100 because of rounding.

Latent Profiles of External and Internal Assets Among BSMM

An LPA was conducted using eight developmental asset indicators representing external assets (support, extracurricular activity participation, mattering and belonging, and boundaries) and internal assets (positive identity, positive values, academic engagement, and social competencies). Models were estimated for class numbers ranging from two to seven. As shown in Table 2, although the BIC and sample-size-adjusted BIC decreased with an increasing number of classes, the VLMR-LRT and LMR-LRT indicated that the four-class solution provided a significantly better fit compared to the three-class solution (ps = .016), while the five-class solution did not enhance the fit. Considering the fit indices and interpretability, the four-class solution was chosen.

Table 2.

Fit Indices for Latent Profile Analysis Models

BIC A-BIC VLMR-LRT LMR-LRT Entropy
2-class 10793.78 10695.37 p < .001 p < .001 0.93
3-class 10062.70 9929.37 p < .001 p < .001 0.92
4-class 9808.49 9640.25 p = .016 p = .016 0.90
5-class 9598.91 9395.74 p = .164 p = .168 0.92
6-class 9489.79 9251.71 p = .426 p = .431 0.89
7-class 9442.52 9169.52 p = .426 p = .428 0.90
Open in a new tab

Note. BIC = Bayesian Information Criterion; A-BIC = sample-size-adjusted BIC; VLMR-LRT = Vuong–Lo–Mendell–Rubin likelihood ratio test; LMR-LRT = Lo–Mendell–Rubin adjusted likelihood ratio test.

Figure 2 summarizes the four classes that represent likely groupings of individuals sharing similar patterns in the associations between internal and external assets. The low assets group (n = 138; 25.5%) reported the lowest levels of both external and internal assets, with the exception of extracurricular activity. The moderate assets group (n = 165; 30.5%) exhibited average levels across indicators. The inconsistent assets group (n = 148; 27.4%) reported generally higher levels of most external and internal assets but notably lower levels of positive identity and extracurricular activity—similar to or below those of the low assets group. The high assets group (n = 90; 16.6%) consistently exhibited elevated levels across all indicators. It is important to note that these profiles were probabilistic in nature, representing latent classes rather than fixed categories.

Figure 2.

Figure 2

Open in a new tab

Means of the Internal and External Assets Variables for Each Latent Profile

Note. Low = low asset class; Inconsistent = inconsistent asset class; Moderate = moderate asset class; High = high asset class.

Moderation Effect of Latent Asset Profiles

Building upon the LPA, we tested whether latent asset profiles moderated the associations between intersectional discrimination and depressive symptoms, anxiety symptoms, and suicidal behavior. Prior to conducting the moderation analyses, we estimated the measurement model for intersectional discrimination using anticipated, day-to-day, and major discrimination as indicators. The indicators demonstrated strong and statistically significant loadings, ranging from .69 to .88 (ps < .001), thereby providing evidence of the coherence and internal consistency of the latent construct of intersectional discrimination.

Depressive Symptoms.

To examine the moderation effect of latent profiles, we regressed depressive symptoms on intersectional discrimination and latent profiles of internal/external assets, comparing the strength of the association between intersectional discrimination and depressive symptoms across latent profiles. The results indicated that intersectional discrimination was positively associated with depressive symptoms in all profiles except the high assets group. Slope comparisons indicated that moderation by asset profile was present: the association was stronger in the inconsistent assets group compared to the low assets (b = 2.28, SE = 0.85, p = .007), moderate assets (b = 2.94, SE = 1.24, p = .020), or high assets groups (b = 4.95, SE = 0.86, p < .001). Furthermore, the association was weaker in the high assets group in comparison to the low assets group (b = -2.67, SE = 0.88, p = .002). There were no differences between the low and moderate assets groups. Additionally, age and household income were found to be negatively associated with depressive symptoms, while housing insecurity exhibited a positive correlation (Table 3).

Table 3.

Moderation of Latent Asset Profiles on the Relation Between Intersectional Discrimination and Depressive and Anxiety Symptoms

Predictors Depressive Symptoms
 Anxiety Symptoms
B SE p B SE p
Covariates
 Age -0.23 0.08 .005 -0.25 0.09 .004
 Household Income -0.35 0.15 .02 -0.06 0.15 .69
 Housing Insecurity 0.50 0.63 .43 -0.73 0.57 .20
 Homelessness 1.96 0.60 .001 2.28 0.56 < .001
 Disclosure of Sexuality -1.04 0.98 .29 -0.42 0.82 .61
Intersectional Discrimination
 Low Asset Class 3.03 0.64 < .001 3.79 0.69 < .001
 Moderate Asset Class 2.37 0.94 .01 3.17 0.77 < .001
 Inconsistent Asset Class 5.30 0.66 < .001 4.73 0.43 < .001
 High Asset Class 0.35 0.57 .53 -1.47 0.66 .03
Moderation Effects
 Moderate vs. Low Asset Class -0.66 1.21 .59 -0.62 1.07 .56
 Inconsistent vs. Low Asset Class 2.28 0.85 .007 0.94 0.75 .21
 High vs. Low Asset Class -2.67 0.88 .002 -5.25 1.03 < .001
 Inconsistent vs. Moderate Asset Class 2.94 1.24 .02 1.56 0.89 .08
 High vs. Moderate Asset Class -2.01 1.09 .06 -4.63 1.03 < .001
 High vs. Inconsistent Asset Class -4.95 0.86 < .001 -6.20 0.86 < .001
Open in a new tab

Anxiety Symptoms.

The same analysis was performed to investigate the moderation effect of latent profiles on the relationship between intersectional discrimination and anxiety symptoms. The findings revealed that intersectional discrimination was positively associated with anxiety symptoms in the low, moderate, and inconsistent assets groups, while it exhibited a negative correlation in the high assets group. The moderation effect of latent profile was evident: the discrimination–anxiety association was significantly weaker in the high assets group compared to the low (b = -5.25, SE = 1.03, p < .001), moderate (b = -4.63, SE = 1.03, p < .001), and inconsistent assets groups (b = -6.20, SE = 0.86, p < .001). No differences emerged among the low, moderate, and inconsistent groups. Moreover, younger age and housing insecurity were associated with increased anxiety symptoms (Table 3).

Suicidal Behavior.

The moderation effects of latent profiles on the association between intersectional discrimination and suicidal behavior are presented in Table 4. Intersectional discrimination was positively associated with suicidal behavior across all profiles; however, the strength of the association varied by profile, indicating a moderating effect. The association was stronger in the moderate assets group compared to the low (b = 1.05, SE = 0.52, p = .040), inconsistent (b = 0.99, SE = 0.39, p = .010), or high assets groups (b = 0.91, SE = 0.31, p = .004). Younger participants and those experiencing housing insecurity exhibited more suicidal behavior than older participants and those with stable housing.

Table 4.

Moderation of Latent Asset Profiles on the Relation Between Intersectional Discrimination and Suicidal Behavior

Predictors Suicidal Behavior
B SE p
Covariates
 Age -0.21 0.03 < .001
 Household Income -0.05 0.07 .44
 Housing Insecurity 1.04 0.25 < .001
 Homelessness 0.93 0.28 .001
 Disclosure of Sexuality 0.35 0.36 .36
Intersectional Discrimination
 Low Asset Class 0.88 0.40 .03
 Moderate Asset Class 1.93 0.29 < .001
 Inconsistent Asset Class 0.94 0.27 < .001
 High Asset Class 1.02 0.18 < .001
Moderation Effects
 Moderate vs. Low Asset Class 1.05 0.52 .04
 Inconsistent vs. Low Asset Class 0.07 0.47 .89
 High vs. Low Asset Class 0.14 0.44 .75
 Inconsistent vs. Moderate Asset Class -0.99 0.39 .01
 High vs. Moderate Asset Class -0.91 0.31 .004
 High vs. Inconsistent Asset Class 0.08 0.31 .80
Open in a new tab

Discussion

The prevalence of depressive symptoms, anxiety, and suicidality remains alarmingly high among LGBTQ+ youth, with an increasing trend observed among Black youth and young adults. BSMM experience both racism and heterosexism, resulting in dual marginalization and intersectional discrimination that negatively impact mental health (Assari et al., 2017; Jackson et al., 2023; Layland et al., 2025; Zelaya et al., 2021). Prior research suggests that developmental assets can serve as protective factors for the mental health of youth (Burningham & Weiler, 2021; DelFerro et al., 2024; Russell & Fish, 2019); however, there has been limited attention given to how these assets influence mental health outcomes among BSMM (Boyd et al., 2025a; Boyd et al., 2025b; Boyd et al., 2026). This study represents the inaugural effort to identify latent asset profiles within BSMM and to examine these profiles as moderators in the association between intersectional discrimination and mental health outcomes. The results carry significant implications for strength-based prevention and intervention strategies aimed at supporting the mental health of BSMM.

Consistent with prior literature, intersectional discrimination was negatively associated with mental health (Boehmer et al., 2021; Pachankis et al., 2014), with greater discrimination linked to worse outcomes across depressive symptoms, anxiety, and suicidality. These findings underscore the need for interventions that help BSMM recognize and navigate the effects of dual marginalization. Addressing these challenges requires multi-level strategies that integrate individual, familial, community, and policy-based prevention and intervention efforts.

Although developmental assets are known to protect the mental health of LGBTQ+ youth (DelFerro et al., 2024; Russell & Fish, 2019), few studies have examined their role among BSMM (Boyd et al., 2025a; Boyd et al., 2025b; Boyd et al., 2026). Understanding this relationship can aid in identifying individuals who are most vulnerable and developing culturally responsive interventions that strengthen developmental assets and promote overall well-being.

This study identified four distinct latent asset profiles based on external and internal developmental assets reported by BSMM. The identified latent asset profiles were categorized as low assets, moderate assets, inconsistent assets, and high assets. Furthermore, the results indicated that latent asset profiles played a moderating role in the association between intersectional discrimination and mental health outcomes, suggesting that developmental assets may serve as a protective factor for the mental health of BSMM. These findings are consistent with existing literature indicating that developmental assets provide a safeguard for LGBTQ+ youths’ mental health (DelFerro et al., 2024; Russell & Fish, 2019). However, the moderating effects of latent asset profiles on the relationship between intersectional discrimination and mental health outcomes exhibited variability across the three mental health outcomes, thereby warranting further examination.

Intersectional discrimination was associated with depressive symptoms across all profiles except for high assets, with the strongest effect observed in the inconsistent assets group. This suggests that while high levels of developmental assets may protect BSMM from mental health issues, uneven or conflicting assets can increase their vulnerability. The inconsistent assets group, characterized by lower positive identity and limited extracurricular engagement, appears particularly at risk. Interventions aimed at fostering positive identity formation, social skills, and peer connectedness could mitigate these effects by enhancing both internal and external assets—especially for those who may have greater variability in their developmental assets.

For anxiety, developmental assets appeared protective, as discrimination was negatively associated with anxiety among those with high assets. This supports the role of strengths-based community programs that cultivate emotional coping and resilience (Fish, 2020). Conversely, all asset profiles revealed a positive association between discrimination and suicidal behavior, with the moderate assets group exhibiting the strongest effect. This finding indicates that while assets may offer some degree of protection, they may not fully offset the effects of persistent intersectional discrimination, emphasizing the need for asset-enhancing interventions tailored to BSMM with inconsistent or partial support networks.

Furthermore, age and housing insecurity emerged as significant covariates. Younger BSMM experienced higher levels of depressive symptoms, anxiety, and suicidal behavior, underscoring the importance of early interventions (Branje et al., 2021; Hall et al., 2021). Housing insecurity was similarly linked to poorer mental health, consistent with evidence that indicate elevated housing instability among LGBTQ+ youth (Goodyear et al., 2024). These findings highlight the broader role of material hardship in the mental health of BSMM and advocate for structural interventions addressing housing and resource inequities.

Limitations

This study presents several limitations. First, all variables were evaluated through self-report in a single session, which may introduce biases related to recall and social desirability, as well as common method variance. Second, the cross-sectional design precludes causal inference and limits conclusions regarding the temporal ordering of discrimination, developmental assets, and mental health outcomes. Third, certain operational decisions may have reduced construct nuance: measures were summarized as mean composites, and assessment time frames differed across instruments (e.g., CES-D: past week; GAD-7: past two weeks; C-SSRS: past three months/lifetime), potentially leading to temporal misalignment. Additionally, the self-report adaptation of the C-SSRS does not substitute for a clinician-administered assessment. Generalizability is limited by the targeted, convenience-based sample of BSMM aged 14 to 24 recruited from Columbus, Detroit, and St. Louis, as well as the reliance on online and community-based recruitment strategies, which may favor individuals with greater access to technology or those more comfortable engaging with BSMM-focused materials. Although forced-response settings were used to minimize missing data, the inclusion of a “prefer not to answer” option may not fully mitigate satisficing behavior. A variety of data-quality safeguards were implemented, but they may not fully eliminate invalid responses. Mixture modeling is also sensitive to model specification; classification uncertainty and unmeasured confounding could skew the observed differences between classes. Participation in extracurricular activities exhibited less distinction across profiles compared to most other indicators, suggesting that it played a relatively smaller role in class differentiation. It was retained because participation in structured activity is theoretically recognized as an external developmental asset; however, the dichotomous measure may have obscured meaningful variation in frequency, quality, and context. Future research should employ more sensitive measures of extracurricular participation and test whether this indicator improves profile classification. Finally, the developmental asset measures were not specifically validated among BSMM, which may limit construct validity and cultural specificity. Future research should prioritize the validation and adaptation of these measures to reflect the lived experiences of this population.

Implications for Practice and Future Research

In summary, the study findings suggest the need for a range of intervention approaches across various levels of practice to support the mental health of BSMM. First, the results indicate that tailored, culturally responsive intervention strategies are likely required to support the mental health needs of BSMM across different latent asset profiles. Additional research is necessary to explore latent asset profiles and the role of developmental assets in the mental health of BSMM, which will inform intervention development, implementation, and resource allocation. For instance, BSMM in the low and inconsistent asset groups are likely to require increased support and could benefit from interventions aimed at building and enhancing developmental assets as a means of fostering resilience and promoting mental health.

Second, the findings advocate for prevention and intervention programs that strengthen internal assets, such as self-efficacy and identity development (Min et al., 2025), as well as external assets like community engagement and supportive adult relationships (Burningham & Weiler, 2021; Fish, 2020). Practical strategies may include mentorship programs, family education focused on acceptance and support, positive identity development activities, social-emotional skill building, and the creation of safe, affirming spaces. These approaches may hold particular relevance for younger BSMM, as strengths-based coping strategies developed during adolescence can carry into emerging adulthood (Branje et al., 2021; Hall et al., 2021).

Third, the results underscore the importance of community-level interventions (Russell & Fish, 2019). Community-based organizations, schools, families, and mentoring programs can collaborate to build and maintain developmental assets among BSMM. Programs that integrate family involvement, natural mentorship, and inclusive school environments may enhance external support while promoting positive identity and help-seeking behaviors. Partnerships should actively involve BSMM in program planning and evaluation to ensure that services reflect local needs, reduce barriers to affirming mental health care, and avoid treating the youth as passive recipients of support.

Fourth, the consistent negative association between intersectional discrimination and mental health highlights the structural oppression embedded within U.S. social systems (Boehmer et al., 2021; Pachankis et al., 2014). Asset-building interventions alone cannot eliminate the harms caused by racism and heterosexism. Local, state, and federal policy responses should strengthen and enforce nondiscrimination protections across schools, healthcare institutions, housing, and employment; allocate funding for culturally responsive and LGBTQ+-affirming mental health and suicide prevention services; and address issues related to housing and resource insecurity. Organizations should partner with BSMM and community-based groups to audit access and outcomes while establishing accountability mechanisms. These structural changes can mitigate exposure to discrimination while ensuring that neither individual nor community assets are relied upon to compensate for oppressive conditions.

Future research should examine the unique roles of specific external and internal assets, determining which assets most strongly predict positive outcomes. Studies should also investigate how these assets interact with experiences of intersectional discrimination through mechanisms such as identity development, social connectedness, and exposure to structural barriers. Longitudinal and mixed-methods designs could clarify how asset profiles evolve and influence mental health trajectories over time, providing richer, contextual insights into the lived experiences of BSMM.

Conclusion

This study makes a significant contribution to understanding developmental asset profiles among BSMM, highlighting how these profiles may moderate the association between intersectional discrimination and mental health outcomes. Guided by the DAF and a resilience-based, positive youth development perspective, findings revealed four latent asset profiles—low, moderate, inconsistent, and high—with elevated asset levels appearing protective against depressive symptoms and anxiety. Moreover, latent asset profiles moderated the association between intersectional discrimination and mental health outcomes, although the effects varied across outcomes. These results reinforce the framework’s central premise that access to internal and external developmental resources can buffer the psychological effects of social adversity. Prevention and intervention efforts should prioritize tailored, strengths-based, and culturally responsive strategies that equip families, schools, and communities with the tools to help BSMM cultivate supportive relationships, positive identities, and community engagement. Future longitudinal research is imperative to examine how developmental assets evolve and sustain their protective effects as BSMM transition through adolescence and young adulthood.

Public Significance Statement.

BMSM often face overlapping racism and homophobia that adversely affect their mental health. This study demonstrates that stronger developmental assets—such as supportive relationships, opportunities for engagement, and a positive sense of identity—can mitigate some of these detrimental effects, highlighting important areas for intervention by schools, families, and communities.

Funding.

This research was supported by the National Institute of Mental Health under Award No. 1R21MH130302–01A1, awarded to Donte T. Boyd.

References

  1. Al-Halabí S, Sáiz PA, Burón P, Garrido M, Benabarre A, Jiménez E, Cervilla J, Navarrete MI, Díaz-Mesa EM, García-Álvarez L, Muñiz J, Posner K, Oquendo MA, García-Portilla MP, & Bobes J (2016). Validation of a Spanish version of the Columbia-Suicide Severity Rating Scale (C-SSRS). Revista De Psiquiatría Y Salud Mental (English Edition), 9(3), 134–142. 10.1016/j.rpsmen.2016.06.004 [DOI] [PubMed] [Google Scholar]
  2. Allison PD (2001). Missing data. SAGE Publications. [Google Scholar]
  3. Almazan EP (2019). Are Black sexual minority adults more likely to report higher levels of psychological distress than white sexual minority adults? Findings from the 2013–2017 National Health Interview Survey. Social Sciences, 8(1), 14. 10.3390/socsci8010014 [DOI] [Google Scholar]
  4. Amola O, & Grimmett MA (2015). Sexual identity, mental health, HIV risk behaviors, and internalized homophobia among black men who have sex with men. Journal of Counseling & Development, 93(2), 236–246. 10.1002/j.1556-6676.2015.00199.x [DOI] [Google Scholar]
  5. Andresen EM, Malmgren JA, Carter WB, & Patrick DL (1994). Screening for depression in well older adults: Evaluation of a short form of the CES-D. American Journal of Preventive Medicine, 10(2), 77–84. 10.1016/s0749-3797(18)30622-6 [DOI] [PubMed] [Google Scholar]
  6. Assari S, & Caldwell CH (2018). Social determinants of perceived discrimination among Black youth: Intersection of ethnicity and gender. Children, 5(2), 24. 10.3390/children5020024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Assari S, Gibbons FX, & Simons RL (2018). Perceived discrimination among Black youth: An 18-year longitudinal study. Behavioral Sciences, 8(5), 44. 10.3390/bs8050044 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Assari S, Moazen-Zadeh E, Caldwell CH, & Zimmerman MA (2017). Racial discrimination during adolescence predicts mental health deterioration in adulthood: Gender differences among Blacks. Frontiers in Public Health, 5. 10.3389/fpubh.2017.00104 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Barnett AP, Molock SD, Nieves-Lugo K, & Zea MC (2019). Anti-LGBT victimization, fear of violence at school, and suicide risk among adolescents. Psychology of Sexual Orientation and Gender Diversity, 6(1), 88–95. 10.1037/sgd0000309 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Benson PL, Scales PC, & Syvertsen AK (2011). The contribution of the developmental assets framework to positive youth development theory and practice. Advances in Child Development and Behavior, 41, 197–230. 10.1016/B978-0-12-386492-5.00008-7 [DOI] [PubMed] [Google Scholar]
  11. Boehmer U, Jesdale BM, Streed CG, & Agénor M (2021). Intersectionality and cancer survivorship: Sexual orientation and racial/ethnic differences in physical and mental health outcomes among female and male cancer survivors. Cancer, 128(2), 284–291. 10.1002/cncr.33915 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Bostwick WB, Meyer I, Aranda F, Russell S, Hughes T, Birkett M, & Mustanski B (2014). Mental health and suicidality among racially/ethnically diverse sexual minority youths. American Journal of Public Health, 104(6), 1129–1136. 10.2105/AJPH.2013.301749 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Boyd DT, Feng X, Quinn CR, Kirklewski S, Smith A, Slesnick N, & Hill SV (2025). Understanding the developmental profiles of young Black sexual minority males ages 14 to 24: Insights from latent profile analysis. Journal of Racial and Ethnic Health Disparities. 10.1007/s40615-025-02677-2 [DOI] [PubMed] [Google Scholar]
  14. Boyd DT, Kirklewski S, Benavides JL, Quinn CR, Saigal A, & Hill SV (2025b). Fostering internal assets to mitigate suicidal behaviors among young Black sexual minority males aged 14–24. BMC Public Health, 25(1), Article 2858. 10.1186/s12889-025-24131-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Boyd DT, Oh H, Patterson A, Weaver A, Durkee M, & Quinn CR (2026). Developmental assets and mental health among Black sexual minority male adolescents: A dominance analysis of depressive symptoms and suicidal behaviors. Journal of Research on Adolescence, 36(2), e70185. 10.1111/jora.70185 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Boyd DT, Ramos SR, Quinn CR, Jones KV, Wilton L, & Nelson LE (2021). Family support and sociocultural factors on depression among Black and Latinx sexual minority men. International Journal of Environmental Research and Public Health, 18(13), 6759. 10.3390/ijerph18136759 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Branje S, De Moor EL, Spitzer J, & Becht AI (2021). Dynamics of identity development in adolescence: A decade in Review. Journal of Research on Adolescence, 31(4), 908–927. 10.1111/jora.12678 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Burningham KL, & Weiler LM (2021). Natural mentoring and LGBTQ youth: A systematic review. Adolescent Research Review, 6(4), 391–407. 10.1007/s40894-020-00146-x [DOI] [Google Scholar]
  19. Ching THW, Finkelstein-Fox L, Lee SY, & Watson RJ (2024). Effects of sexual and gender minority stress on depressive symptoms among adolescents of color in the United States. Cultural Diversity & Ethnic Minority Psychology, 30(2), 309–318. 10.1037/cdp0000562 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Choi KH, Paul J, Ayala G, Boylan R, & Gregorich SE (2013). Experiences of discrimination and their impact on the mental health among African American, Asian and Pacific Islander, and Latino men who have sex with men. American Journal of Public Health, 103(5), 868–874. 10.2105/AJPH.2012.301052 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. DeLaney EN, Williams CD, Jones SCT, Corley NA, Lozada FT, Walker CJ, & Dick DM (2022). Black college students’ ethnic identity and academic achievement: Examining mental health and racial discrimination as moderators. Journal of Black Psychology, 48(1), 100–129. 10.1177/00957984211034268 [DOI] [Google Scholar]
  22. DelFerro J, Whelihan J, Min J, Powell M, DiFiore G, Gzesh A, Jelinek S, Schwartz KTG, Davis M, Jones JD, Fiks AG, Jenssen BP, & Wood S (2024). The role of family support in moderating mental health outcomes for LGBTQ+ youth in primary care. JAMA Pediatrics, 178(9), 914–922. 10.1001/jamapediatrics.2024.1956 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. English D, Boone CA, Carter JA, Talan AJ, Busby DR, Moody RL, Cunningham DJ, Bowleg L, & Rendina HJ (2022). Intersecting structural oppression and suicidality among Black sexual minority male adolescents and emerging adults. Journal of Research on Adolescence, 32(1), 226–243. 10.1111/jora.12726 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Fernandes T, Alves B, & Gato J (2023, July). Between resilience and agency: A systematic review of protective factors and positive experiences of LGBTQ+ students. Healthcare, 11(14), 2098. 10.3390/healthcare11142098 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Fish JN (2020). Future directions in understanding and addressing mental health among LGBTQ youth. Journal of Clinical Child & Adolescent Psychology, 49(6), 943–956. 10.1080/15374416.2020.1815207 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Friedman MR, Bukowski L, Eaton LA, Matthews DD, Dyer TV, Siconolfi D, & Stall R (2018). Psychosocial health disparities among Black bisexual men in the U.S.: Effects of sexuality nondisclosure and gay community support. Archives of Sexual Behavior, 48(1), 213–224. 10.1007/s10508-018-1162-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Garcia GM, Hedwig T, Hanson BL, Rivera M, & Smith CA (2019). The relationship between mixed race/ethnicity, developmental assets, and mental health among youth. Journal of Racial and Ethnic Health Disparities, 6(1), 77–85. 10.1007/s40615-018-0501-2 [DOI] [PubMed] [Google Scholar]
  28. Goodyear T, Chayama KL, Oliffe JL, Kia H, Fast D, Mniszak C, Knight R, & Jenkins E (2024). Intersecting transitions among 2S/LGBTQ+ youth experiencing homelessness: A scoping review. Children and Youth Services Review, 156, 107355. 10.1016/j.childyouth.2023.107355 [DOI] [Google Scholar]
  29. Graham LF, Aronson RE, Nichols T, Stephens CF, & Rhodes SD (2011). Factors influencing depression and anxiety among Black sexual minority men. Depression Research and Treatment, 2011, 587984. 10.1155/2011/587984 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Hall WJ, Dawes HC, & Plocek N (2021). Sexual orientation identity development milestones among lesbian, gay, bisexual, and queer individuals: A systematic review and meta-analysis. Frontiers in Psychology, 12. 10.3389/fpsyg.2021.753954 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Henry DB, Tolan PH, & Gorman-Smith D (2005). Cluster analysis in family psychology research. Journal of Family Psychology, 19(1), 121–132. 10.1037/0893-3200.19.1.121 [DOI] [PubMed] [Google Scholar]
  32. Jackson SD, Harvey TD, Watson RJ, Pereira K, & Clark KA (2023). Multilevel stigma and depression among a national sample of Black and Latinx LGBTQ+ adolescents in the United States. Journal of Psychopathology and Clinical Science, 132(5), 577–589. 10.1037/abn0000841 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Jedidi K, Ramaswamy V, & Desarbo WS (1993). A maximum likelihood method for latent class regression involving a censored dependent variable. Psychometrika, 58(3), 375–394. 10.1007/bf02294647 [DOI] [Google Scholar]
  34. Jung T, & Wickrama KAS (2008). An introduction to latent class growth analysis and growth mixture modeling. Social and Personality Psychology Compass, 2(1), 302–317. 10.1111/j.1751-9004.2007.00054.x [DOI] [Google Scholar]
  35. Layland EK, Diaz JE, Parra LA, Berglund P, Kipke MD, & Bray BC (2025). How do anxiety and depression trajectories vary among Black, Latinx, and Afro-Latinx sexual minority young men? Uncovering variation in development with intersectional subgroups. Developmental Psychology, 61(11), 2119–2136. 10.1037/dev0001968 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lindsay BL, Bernier E, Boman J, & Boyce MA (2022). Understanding the connection between student wellbeing and teaching and learning at a Canadian research university: A qualitative student perspective. Pedagogy in Health Promotion, 9(1), 5–16. 10.1177/23733799221089578 [DOI] [Google Scholar]
  37. Liu F, Wang N, & Chong ESK (2025). Effect of sexual orientation victimization on suicidality: Perceived burdensomeness and thwarted belongingness as mediators, and independent self-construal as moderator. Psychology of Sexual Orientation and Gender Diversity, 12(4), 646–656. 10.1037/sgd0000694 [DOI] [Google Scholar]
  38. Liu Q, Nestor BA, Eckstrand KL, & Cole DA (2024). Stress proliferation in ethnoracial disparities of mental health among U.S. sexual minority adults. Cultural Diversity & Ethnic Minority Psychology, 30(3), 553–565. 10.1037/cdp0000598 [DOI] [PubMed] [Google Scholar]
  39. Liu RT, Walsh RFL, Sheehan AE, Cheek SM, & Carter SM (2020). Suicidal ideation and behavior among sexual minority and heterosexual youth: 1995–2017. Pediatrics, 145(3), e20192221. 10.1542/peds.2019-2221 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Lo Y, Mendell NR, & Rubin DB (2001). Testing the number of components in a normal mixture. Biometrika, 88(3), 767–778. 10.1093/biomet/88.3.767 [DOI] [Google Scholar]
  41. Lucassen MFG, Stasiak K, Samra R, Frampton CMA, & Merry SN (2017). Sexual minority youth and depressive symptoms or depressive disorder: A systematic review and meta-analysis of population-based studies. Australian & New Zealand Journal of Psychiatry, 51(8), 774–787. 10.1177/0004867417713664 [DOI] [PubMed] [Google Scholar]
  42. Lytle MC, De Luca SM, Blosnich JR, & Brownson C (2015). Associations of racial/ethnic identities and religious affiliation with suicidal ideation among lesbian, gay, bisexual, and questioning individuals. Journal of Affective Disorders, 178, 39–45. 10.1016/j.jad.2014.07.039 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Marciano A, & Antebi-Gruszka N (2022). Offline and online discrimination and mental distress among lesbian, gay, and bisexual individuals: The moderating effect of LGBTQ Facebook use. Media Psychology, 25(1), 27–50. 10.1080/15213269.2020.1850295 [DOI] [Google Scholar]
  44. McConnell EA, Janulis P, Phillips G II, Truong R, & Birkett M (2018). Multiple minority stress and LGBT community resilience among sexual minority men. Psychology of Sexual Orientation and Gender Diversity, 5(1), 1–12. 10.1037/sgd0000265 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Min MO, Yoon M, Minnes S, & Singer LT (2025). Internal assets in low‐income urban adolescents from 12 to 17 years: Gender variation. Journal of Research on Adolescence, 35(1), e70018. 10.1111/jora.70018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Moran SA, Bishop MD, Watson RJ, & Fish JN (2024). LGBTQ+ youth policy and mental health: Indirect effects through school experiences. Journal of Research on Adolescence, 35(1), e13052. 10.1111/jora.13052 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Muthén BO (2001). Latent variable mixture modeling. In Marcoulides G & Schumacker R (Eds.), New developments and techniques in structural equation modeling (pp. 1–33). Lawrence Erlbaum Associates. [Google Scholar]
  48. Muthén BO, Muthén LK, & Asparouhov T (2017). Regression and mediation analysis using Mplus. Muthén & Muthén. [Google Scholar]
  49. Nagin DS, & Odgers CL (2010). Group-based trajectory modeling in clinical research. Annual Review of Clinical Psychology, 6(1), 109–138. 10.1146/annurev.clinpsy.121208.131413 [DOI] [PubMed] [Google Scholar]
  50. Nylund KL, Asparouhov T, & Muthén BO (2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling: A Multidisciplinary Journal, 14(4), 535–569. 10.1080/10705510701575396 [DOI] [Google Scholar]
  51. Pachankis JE, Hatzenbuehler ML, & Starks TJ (2014). The influence of structural stigma and rejection sensitivity on young sexual minority men’s daily tobacco and alcohol use. Social Science & Medicine, 103, 67–75. 10.1016/j.socscimed.2013.10.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Posner K, Brown GK, Stanley B, Brent DA, Yershova KV, Oquendo MA, Currier GW, Melvin GA, Greenhill L, Shen S, & Mann JJ (2011). The Columbia–Suicide Severity Rating Scale: Initial validity and internal consistency findings from three multisite studies with adolescents and adults. American Journal of Psychiatry, 168(12), 1266–1277. 10.1176/appi.ajp.2011.10111704 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Quinn K, Dickson-Gomez J, DiFranceisco W, Kelly JA, St. Lawrence JS, Amirkhanian YA, & Broaddus M (2015). Correlates of internalized homonegativity among Black men who have sex with men. AIDS Education and Prevention, 27(3), 212–226. 10.1521/aeap.2015.27.3.212 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Rose T, Lindsey MA, Xiao Y, Finigan-Carr NM, & Joe S (2017). Mental health and educational experiences among Black youth: A latent class analysis. Journal of Youth and Adolescence, 46(11), 2321–2340. 10.1007/s10964-017-0723-3 [DOI] [PubMed] [Google Scholar]
  55. Roulston C, McKetta S, Price M, Fox KR, & Schleider JL (2023). Structural correlates of mental health support access among sexual minority youth of color during COVID-19. Journal of Clinical Child & Adolescent Psychology, 52(5), 649–658. 10.1080/15374416.2022.2034633 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Russell ST, & Fish JN (2016). Mental health in lesbian, gay, bisexual, and transgender (LGBT) youth. Annual Review of Clinical Psychology, 12(1), 465–487. 10.1146/annurev-clinpsy-021815-093153 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Russell ST, & Fish JN (2019). Sexual minority youth, social change, and health: A developmental collision. Research in Human Development, 16(1), 5–20. 10.1080/15427609.2018.1537772 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Schafer JL, & Graham JW (2002). Missing data: Our view of the state of the art. Psychological Methods, 7(2), 147–177. 10.1037/1082-989x.7.2.147 [DOI] [PubMed] [Google Scholar]
  59. Scheim AI, & Bauer GR (2019). The Intersectional Discrimination Index: Development and validation of measures of self-reported enacted and anticipated discrimination for intercategorical analysis. Social Science & Medicine, 226, 225–235. 10.1016/j.socscimed.2018.12.016 [DOI] [PubMed] [Google Scholar]
  60. Smith BL, Boyd DT, & Quinn CR (2026). Examining the impact of social connectedness on depression and suicide ideation among Black youth experiencing discrimination: A path analysis. Journal of Racial and Ethnic Health Disparities, 13(3), 2237–2247. 10.1007/s40615-025-02414-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Spitzer RL, Kroenke K, Williams JBW, & Löwe B (2006). Generalized Anxiety Disorder 7 (GAD-7) [Database record]. APA PsycTESTS. 10.1037/t02591-000 [DOI] [Google Scholar]
  62. Syvertsen AK, Scales PC, & Toomey RB (2021). Developmental Assets framework revisited: Confirmatory analysis and invariance testing to create a new generation of asset measures for applied research. Applied Developmental Science, 25(4), 291–306. 10.1080/10888691.2019.1613155 [DOI] [Google Scholar]
  63. Szymanski DM, & Ikizler AS (2013). Internalized heterosexism as a mediator in the relationship between gender role conflict, heterosexist discrimination, and depression among sexual minority men. Psychology of Men & Masculinity, 14(2), 211–219. 10.1037/a0027787 [DOI] [Google Scholar]
  64. Toomey RB, Syvertsen AK, & Flores M (2019). Are developmental assets protective against suicidal behavior? Differential associations by sexual orientation. Journal of Youth and Adolescence, 48(4), 788–801. 10.1007/s10964-018-0954-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Vargas SM, Huey SJ Jr., & Miranda J (2020). A critical review of current evidence on multiple types of discrimination and mental health. American Journal of Orthopsychiatry, 90(3), 374–390. 10.1037/ort0000441 [DOI] [PubMed] [Google Scholar]
  66. Vermunt JK (2017). Latent class modeling with covariates: Two improved three-step approaches. Political Analysis, 18(4), 450–469. 10.1093/pan/mpq025 [DOI] [Google Scholar]
  67. Vuong QH (1989). Likelihood ratio tests for model selection and non-nested hypotheses. Econometrica, 57(2), 307–333. 10.2307/1912557 [DOI] [Google Scholar]
  68. Wade RM, Bouris AM, Neilands TB, & Harper GW (2022). Racialized sexual discrimination (RSD) and psychological wellbeing among young sexual minority Black men (YSMBM) who seek intimate partners online. Sexuality Research and Social Policy, 19(3), 1341–1356. 10.1007/s13178-021-00676-6 [DOI] [Google Scholar]
  69. Wiium N, Beck M, & Ferrer-Wreder L (2021). The importance of developmental assets to mental health in Norwegian youth. Frontiers in Psychology, 12, 687537. 10.3389/fpsyg.2021.687537 [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Williams KA, & Chapman MV (2012). Unmet health and mental health needs among adolescents: The roles of sexual minority status and child-parent connectedness. American Journal of Orthopsychiatry, 82(4), 473–481. 10.1111/j.1939-0025.2012.01182.x [DOI] [PubMed] [Google Scholar]
  71. Zelaya DG, DeBlaere C, & Velez BL (2023). Psychometric validation and extension of the LGBT People of Color Microaggressions Scale with a sample of sexual minority BIPOC college students. Psychology of Sexual Orientation and Gender Diversity, 10(3), 405–412. 10.1037/sgd0000553 [DOI] [Google Scholar]

RESOURCES