https://orcid.org/0000-0003-4825-6673
Figures
Abstract
Purpose
Gender minority (GM) adolescents, who have a different gender identity than their sex assigned at birth, may use substances as a coping strategy in response to GM-related stressors. This study examined longitudinal effects of gender minority stressors on substance use in GM adolescents, and related risk factors (internalized transphobia, depressive symptoms, anxious symptoms) and protective factors (resilience, gender-related pride, family functioning, social support, gender-related community connectedness).
Methods
Participants were 30 GM adolescents, ages 13–17 years, from the U.S. community-based longitudinal Trans Teen and Family Narratives Project. Participants completed an online survey every 6 months across 2 years (5 waves; data collected 2015–2019).
Results
Exposure to gender minority stressors was associated with higher odds of alcohol use. Across models, internalized transphobia (risk factor), resilience (protective factor), and gender-related pride (protective factor) were the most significant mediators of associations between gender minority stressors and substance use. Family functioning and social support (protective factors) significantly moderated the association between gender minority stressors and alcohol use, such that family functioning and social support were protective for alcohol use at lower levels of gender minority stress, but not at higher levels.
Conclusion
Results suggest that GM adolescents engage in substance use as a coping strategy in response to gender minority stressors. A number of hypothesized risk and protective factors mediated or moderated these associations. Future interventions with GM adolescents should focus efforts on addressing internalized transphobia as a risk factor and strengthening resilience, gender-related pride, and family functioning as protective factors for substance use.
Citation: Katz-Wise SL, Sarda V, Austin SB, Harris SK (2021) Longitudinal effects of gender minority stressors on substance use and related risk and protective factors among gender minority adolescents. PLoS ONE 16(6): e0250500. https://doi.org/10.1371/journal.pone.0250500
Editor: José C. Perales, Universidad de Granada, SPAIN
Received: July 17, 2020; Accepted: April 7, 2021; Published: June 2, 2021
Copyright: © 2021 Katz-Wise et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Researchers who are interested in accessing the data from this study will need to be formally added to the human subjects protocol for this study, which is held by the Boston Children’s Hospital Institutional Review Board. To be formally added to the protocol, researchers must either become affiliated with Boston Children’s Hospital or complete a data use agreement. Interested and qualified researchers should submit data access requests to Allegra Gordon, ScD at allegrag@bu.edu.
Funding: The Trans Teen and Family Narratives Project was funded by grants from the Eunice Kennedy Shriver National Institute of Child Health & Human Development (K99R00HD082340) and the Boston Children’s Hospital Aerosmith Endowment Fund for Prevention and Treatment of AIDS and HIV Infections, both awarded to SKW. SKW and SBA were also funded by the Maternal and Child Health Bureau, Health Resources and Services Administration (Leadership Education in Adolescent Health project 6T71-MC00009). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Increased substance use is common among adolescents, as youth experiment with strategies to cope with life stressors. For gender minority (GM) adolescents, who have a different gender identity than their sex assigned at birth, exposure to GM-related stressors, such as victimization, may result in substance use as a coping mechanism [1, 2], which may negatively affect GM adolescents’ health. Compared to cisgender adolescents, GM adolescents report higher rates of risk factors (e.g., emotional distress, bullying victimization) and lower rates of protective factors (e.g., internal assets, family connectedness) [2].
GM adolescents are at higher risk for substance use than cisgender adolescents [2, 3]. Previous research indicated the prevalence of substance use was 2.5–4 times higher for GM youth compared to cisgender youth.3 Substance use among GM adolescents may be linked to experiencing GM-related stressors, such as stigma-based prejudice [1, 2, 4]. A recent study of cisgender sexual minority and GM adolescents found that experiencing multiple types of victimization was more common among GM than cisgender sexual minority youth [1].
Minority stress research has indicated GM individuals experience adverse mental health outcomes and use substances to cope with experiences of prejudice [3, 5–8]. The U.S. National Transgender Discrimination Survey reported that 28% of adult transgender men used substances to cope with GM-related stigma in healthcare [5]. Compared to cisgender adolescents, GM adolescents report higher rates of emotional distress [2], which may increase the likelihood of using substances to cope with minority stress [9]. A recent study found higher prevalence of substance use among GM youth compared to cisgender youth was partially explained by victimization and depressive symptoms [3]. Protective factors may buffer adverse effects of minority stress on GM adolescents’ mental health. A study conducted by our team found better family functioning was associated with less self-harm and depressive/anxious symptoms, and greater self-esteem and resiliency among GM adolescents [10], indicating family support is an important protective factor for GM youth. One of the few studies to examine protective factors related to substance use among GM adolescents found that family connectedness and social support may reduce the likelihood that GM adolescents will use substances to cope with minority stressors [9, 10]. Other GM-specific protective factors, such as gender-related pride and gender-related community connectedness, may also reduce substance use among GM adolescents, though to our knowledge, this has not yet been studied. Previous research examining associations between minority stressors and substance use among GM adolescents was cross-sectional; longitudinal methods are needed to establish patterns over time.
Study aims were to examine longitudinal effects of gender minority stressors on substance use among GM adolescents and to identify related risk and protective factors for intervention efforts. Notably, some risk and protective factors are relevant for all adolescents (depressive/anxious symptoms, resilience, family functioning, social support) and some risk and protective factors are relevant specifically for GM adolescents (internalized transphobia, gender-related pride, gender-related community connectedness). Thus, this study aimed to examine how risk and protective factors relevant to all adolescents specifically operate in GM adolescents related to their substance use, as well as examining some GM-specific risk and protective factors. We hypothesized that: H1) greater exposure to gender minority stressors would be longitudinally associated with greater substance use over time among GM adolescents, H2) risk factors (internalized transphobia, depressive/anxious symptoms) would mediate longitudinal associations between gender minority stressors and substance use, and H3) protective factors (mediators: resilience, gender-related pride; moderators: family functioning, social support, gender-related community connectedness) would either mediate or moderate (depending on the specific factor) longitudinal associations between gender minority stressors and substance use among GM adolescents.
Methods
Participants
Participants were 33 GM adolescents, ages 13–17 years (at Wave 1), from the longitudinal Trans Teen and Family Narratives (TTFN) Project. Participants were included in this analysis if they completed the relevant measures at Wave 1 and at least one additional wave (analytic N = 30). Participants identified as trans feminine (n = 11), trans masculine (n = 15), and nonbinary assigned female (n = 4). Most participants identified their race/ethnicity as White (73%) or Hispanic/Latinx (13%). See Table 1 for other sample characteristics.
Study design
Participants completed an online survey at five waves, every six months across two years (December 2015 to March 2019). Surveys were completed following a qualitative interview at the participants’ home, at the researchers’ offices, or via video conference for participants who lived beyond a 2.5-hour driving radius from researcher offices. GM adolescents participated in the study with 1–2 caregivers and a sibling. Survey data from GM adolescents only were analyzed for the current study. Prior to completing the interview and survey, each participant gave written informed assent or consent. Surveys were administered by researchers who were LGBTQ-identified or allies. Following study participation, each participant received a $20 gift card and a list of resources tailored for GM adolescents and their families. A comprehensive safety plan was used to follow-up with participants who reported intention to harm themselves or others in the interview or psychological distress on the survey (score of ≥11 on the CES-D) and connect them to mental health screening and resources. This study was approved by the Boston Children’s Hospital Institutional Review Board.
Measures
Predictor: Gender minority stressors.
Gender-minority stressor composite. Three subscales from the Gender Minority Stress and Resilience Measure (GMSR) [11]: gender-related rejection, non-affirmation of gender identity, and gender-related victimization, were combined into one composite scale. Gender-related rejection (5 items, sample item: “I have been rejected or distanced from friends because of my gender identity or expression”) and non-affirmation of gender identity (6 items, sample item: “I have to repeatedly explain my gender identity to people or correct the pronouns people use”) were both assessed at Waves 1–5. Wave 1 response options were modified to fit the age of the current sample: 0 (Never); 1 (Yes, before age 13); 1 (Yes, after age 13); 1 (Yes, in the past year). Gender-related victimization (6 items, sample item: “I have been threatened with physical harm because of my gender identity or expression”) was assessed at Waves 2–5. Only 2 of the 6 items, verbal harassment and threats of physical harm, were endorsed by participants in this sample, and therefore included in the composite score. Response options for all three subscales at Waves 2–5 assessed experiences from the past 6 months: 0 (Never), 1 (Yes, but not in the past 6 months), 1 (Yes, in the past 6 months). An overall composite scale score was calculated for each wave by summing responses across all items (range: 0–11); higher scores indicated greater exposure to gender-minority stressors. Cronbach’s alpha was 0.73 for the composite scale.
Outcomes: Substance use.
Tobacco, alcohol, and marijuana use were assessed at Waves 1–5 with items from the Youth Risk Behavior Surveillance System Survey (YRBS) [12]. Participants ages 13–14 years completed items from the middle school survey; participants ages 15–17 years completed items from the high school survey. When younger participants became age 15–17 years in later waves, they completed items from the high school survey.
Each substance was assessed with a yes/no question from the YRBS [12] (“Have you ever tried cigarette smoking, even one or two puffs?”; “Have you ever had a drink of alcohol, other than a few sips?”; “Have you ever used marijuana?”). For alcohol use assessed in Waves 2–5, this question was changed to “In the past 6 months, have you ever had a drink of alcohol, other than a few sips?” (yes/no). If participants responded yes to any substance at any wave, they answered 2 additional items assessing age of first use (both age groups) and past 30-day use (both age groups for tobacco; age 15–17 only for alcohol and marijuana).
Risk mediators.
Internalized transphobia. Assessed at Waves 1–5 with an 8-item subscale from the GMSR [11]. Sample item: “When I think of my gender identity or expression, I feel depressed.” Response options were on a 5-point Likert scale: 0 (strongly disagree) to 4 (strongly agree). We calculated a subscale score for each wave by summing item responses (range: 0–24); higher scores indicated greater internalized transphobia. Cronbach’s alpha was 0.90.
Depressive symptoms. Assessed for the past week with the 10-item Center for Epidemiologic Studies Depression Scale (CES-D)–Short Form [13]. Sample item: “I was bothered by things that usually don’t bother me.” Response options: 0 (rarely or none of the time–less than 1 day), 1 (some or a little of the time– 1–2 days), 2 (occasionally or a moderate amount of time– 3–4 days), 3 (most or all of the time– 5–7 days). Items worded in the opposite valence were reverse-coded and a scale score was created by summing item responses (range: 0–30); higher scores indicated greater depressive symptoms. Cronbach’s alpha was 0.88.
Anxious symptoms. Assessed with the 6-item Spence Children’s Anxiety Scale (SCAS) [14, 15]. Sample item: “I worry about things.” Response options: 1 (never), 2 (sometimes), 3 (often), 4 (always). A scale score was created by summing item responses (range: 6–24); higher scores indicated greater anxious symptoms. Cronbach’s alpha was 0.87.
Protective mediators.
Resilience. Assessed with the 22-item Resilience Scale for Adolescents [16]. Sample item: “I reach my goals if I work hard.” Response options were on a 5-point Likert scale: 1 (totally disagree) to 5 (totally agree). A scale score was calculated by computing a mean of the items (range: 1–5); higher scores indicated greater resiliency. Cronbach’s alpha was 0.91.
Gender-related pride. Assessed at Waves 1–5 with an 8-item subscale from the GMSR [11]. Sample item: “My gender identity or expression makes me feel special and unique.” Response options were on a 5-point Likert scale: 0 (strongly disagree) to 4 (strongly agree). A subscale score was calculated for each wave by summing item responses (range: 0–32); higher scores indicate greater gender-related pride. Cronbach’s alpha was 0.83.
Moderators.
Family functioning. Assessed with two subscales from the Family Adaptability and Cohesion Evaluation Scales (FACES IV) [17]. Family communication subscale (10 items): Sample item: “Family members are very good listeners.” Response options were on a 5-point Likert scale: 1 (strongly disagree) to 5 (strongly agree). Family satisfaction subscale (10 items): Sample item: “The degree of closeness between family members.” Response options were on a 5-point Likert scale: 1 (very dissatisfied) to 5 (extremely satisfied). The two subscales were combined into one scale. An overall scale score was calculated for each wave by summing responses across items from both subscales (range: 20–100); higher scores indicated better family functioning. Cronbach’s alpha was 0.95 for the combined subscales.
Social support. Assessed with the 12-item Multidimensional MSPSS [18]. Sample item: “There is a special person who is around when I am in need.” Response options were on a 7-point Likert scale: 1 (very strongly disagree) to 7 (very strongly agree). Scale scores were calculated by summing item responses (range: 12–84); higher scores indicated more perceived support. Cronbach’s alpha was 0.92.
Gender-related community connectedness. Assessed at Waves 1–5 with a 5-item subscale from the GMSR [11]. Sample item: “I feel part of a community of people who share my gender identity.” Response options were on a 5-point Likert scale: 0 (strongly disagree) to 4 (strongly agree). A subscale score was calculated for each wave by summing item responses (range: 0–20); higher scores indicate greater gender-related community connectedness. Cronbach’s alpha was 0.81.
Statistical analyses
We computed univariate statistics (frequency for categorical variables, M/SD for continuous variables) for all predictors, outcomes, and hypothesized mediators/moderators across Waves 1–5 (i.e., across two years). We assessed change over time in gender minority stressors and substance use, using generalized linear mixed effects modeling with data nested within participant, and wave as a continuous predictor. We ran these models accounting for the multiple observations per participant by including a random intercept; we examined wave as a fixed effect. To evaluate potential mediation/moderation of the association between gender minority stressors (fixed predictor) and reported use of each type of substance (outcomes) at the subsequent wave by our hypothesized risk and protective factors, we first examined bivariate associations between gender minority stressors and the hypothesized mediator/moderator variables. Bivariate associations were examined using regression analysis with generalized estimating equations (GEE), with data nested within subjects, to account for intra-subject correlation of data across Waves 1–5. We then used Hayes’ approach to test three mediation and moderation models [19]. Two mediation models tested for mediation of the associations between gender minority stressors and substance use by risk factors (H1) and protective factors (H2). The first mediation model (Fig 1A) tested a risk pathway by which gender minority stressors were associated with greater substance use across two years via sequential mediating internal risk factors (internalized transphobia leading to depressive/anxious symptoms). The second mediation model (Fig 1B) tested a protective pathway by which gender minority stressors were associated with less substance use across two years via mediating internal protective factors (resilience, gender-related pride). Finally, models tested for moderation of associations between gender minority stressors and substance use by social factors (family functioning, social support, gender-related community connectedness) (H3). Models predicting alcohol and marijuana use were adjusted for wave and sex assigned at birth (both treated as fixed effects); models predicting tobacco use were adjusted only for sex assigned at birth due to the low rate of tobacco use across waves. Wave was used as a covariate rather than age because age data were missing and the two variables were highly correlated. Participants were excluded from analyses if they were missing data across Waves 2–5 (n = 3).
(a) Multiple Mediation Statistical Model with Mediation by Risk Factors of the Longitudinal Effect of Gender Minority Stressors on Substance Use Across Waves 1–5. (b) Mediation Statistical Model with Mediation by Protective Factors of the Longitudinal Effect of Gender Minority Stressors on Substance Use Across Waves 1–5.
Results
At Wave 1, 17% of participants reported any substance use and 4% reported multiple substance use (Table 2). Regarding specific substances, at Wave 1, 13% of participants had ever used tobacco, 27% had ever used alcohol, and 23% had ever used marijuana (Table 1). Among participants who had ever used substances at Wave 1, the average age of first use was 13.7 years for tobacco, 13.4 years for alcohol, and 14.4 years for marijuana (Table 1). By Wave 5 (two years after Wave 1), 56% of participants reported any substance use and 32% reported multiple substance use (Table 2).
Gender minority stressors and substance use
Results indicated that rates of recent tobacco, alcohol, and marijuana use all significantly increased across two years (Table 2). Higher exposure to gender minority stressors significantly increased the odds of alcohol use in later waves (OR = 1.59, 95% CI = 1.02, 2.49), but not of tobacco or marijuana use; therefore, H1 was partially supported.
Risk pathway models: Mediation by risk factors
Bivariate coefficients indicated strong positive associations between gender minority stressor exposure and hypothesized risk mediators (Table 3; see Supplemental S1 Table for QIC). In all analyzed models, higher gender minority stressor exposure was associated with higher odds of internalized transphobia, and depressive and anxious symptoms across two years (paths a11 and a12, Table 4). Internalized transphobia, in turn, was associated with increased odds of depressive and anxious symptoms over time. Internalized transphobia also had a significant independent effect on subsequent substance use risk in all models, except in the tobacco use mediation model with depressive symptoms as the second mediator. Consequently, internalized transphobia was a significant mediator of the effect of gender minority stressor exposure on subsequent substance use across all models, except for the tobacco use model in which depressive symptoms was the second mediator (Table 4); in that model, depressive symptoms was the significant mediator. In all other models, however, depressive/anxious symptoms did not show a significant independent effect on substance use risk. As a result, tests of the indirect effect through both internalized transphobia and depressive/anxious symptoms (i.e., multiple sequential mediators) were largely non-significant. Across models, internalized transphobia was more consistently a significant mediator than either depressive or anxious symptoms. In sum, H2 was partially supported.
Protective pathway models: Mediation by protective factors
Across models, gender minority stressor exposure was significantly negatively associated with the protective factors of resilience and gender-related pride (Table 5). Increased resilience and gender-related pride, in turn, tended to predict lower odds of subsequent substance use (all adjusted odds ratios below 1), although the effects were only statistically significant for tobacco (resilience and gender-related pride) and marijuana use (resilience only). This protective mediation pathway was significant for tobacco use (through gender-related pride) and for marijuana use (through resilience). In sum, H3 was partially supported.
Moderation effects
We found that family functioning and social support both significantly moderated the effect of gender minority stressor exposure on alcohol use across two years (Table 6), with family functioning and social support protective for alcohol use at lower levels of gender minority stress, but not at higher levels of gender minority stress (Fig 2A and 2B). We found no other significant moderation effects. In sum, H3 was partially supported.
(a) Significant Interaction Between Gender Minority Stressor Composite and Family Functioning Predicting Alcohol Use (-1 = lower family functioning; 1 = higher family functioning). (b) Significant Interaction Between Gender Minority Stressor Composite and Social Support Predicting Alcohol Use (-1 = lower social support, 1 = higher social support).
Discussion
This longitudinal study examined prospective associations between exposure to gender minority stressors and substance use across two years among GM adolescents, with nearly 60% of participants reporting recent substance use at any wave by the final time point. At the final time point, 20% of the GM adolescents in the current study reported recent tobacco use, and 40% reported recent alcohol use and marijuana use. For comparison, data from the Youth Risk Behavior Survey collected during 2019 (data for the current study were collected in December 2015 to March 2019) indicated that 37% of high school students reported recent tobacco use, 29% reported recent alcohol use, and 22% reported recent marijuana use [20]. Thus, GM adolescents the in current study reported higher rates of alcohol and marijuana use, but not tobacco use. In the current study, exposure to gender minority stressors was associated with higher odds of alcohol use, but not higher odds of tobacco or marijuana use. These findings support previous minority stress research indicating that GM individuals engage in substance use (namely alcohol) to cope with prejudice [3, 5–8].
In testing hypothesized mediation pathways between gender minority stressor exposure and substance use, we found that internalized transphobia, resilience, and gender-related pride significantly mediated risk or protective pathways between exposure to gender minority stressors and substance use. Notably, two of these three factors are specific to GM: internalized transphobia and gender-related pride. Internalized transphobia is comprised of four separate dimensions: lack of pride in GM identity, desire to pass as a cisgender person, isolation from other GM people, and shame [21]. GM adolescents may need support to reduce internalized transphobia and bolster gender-related pride, ultimately reducing substance use in this population.
Results also indicated family functioning and social support moderated associations between gender minority stressor exposure and alcohol use, such that family functioning and social support were protective for alcohol use at lower levels of gender minority stress, but not at higher levels of gender minority stress. This is consistent with previous research demonstrating the protective role of social support and family connectedness for GM adolescents’ substance use [8]. Interestingly, family functioning and social support were not protective for alcohol use at higher levels of gender minority stress, highlighting the need for more research to identify protective factors. Still, family functioning and social support were protective at lower levels of gender minority stress and family functioning was independently associated with lower odds of marijuana use among GM youth in this sample. Substance use interventions for this population should focus on improving family functioning and increasing social support within and outside of the family.
A number of limitations should be mentioned; the sample was small, and disproportionately White and non-Hispanic and higher socioeconomic status. Participants were also geographically limited to the New England region of the U.S. Findings may not be generalizable to GM groups underrepresented in this sample. The measures used at Wave 1 assessed a different timeframe than the measures used at Waves 2–5, which limits comparability of the data across waves. The sample size was also too small to allow for modeling wave as a multinomial variable; thus, wave was modeled as a linear variable. An additional limitation is that the Hayes approach to modeling mediation/moderation did not allow for specification of longitudinal correlation structure of data, resulting in greater risk for Type 1 error. Despite limitations, the TTFN Project provides a unique opportunity to examine longitudinal effects of gender minority stressor exposure on substance use across two years among GM adolescents, as a first step for future research in this area.
Conclusion
Findings from this research have implications for intervention efforts to reduce substance use among GM adolescents. Efforts should focus on addressing internalized transphobia as a risk factor and strengthening resilience, gender-related pride, family functioning, and social support as protective factors for substance use. Since GM adolescents appear to be using substances to cope with exposure to gender minority stressors, programs could assist adolescents in identifying adaptive coping strategies in response to such stressors. GM adolescents should also be connected to resources where they can connect with other GM adolescents. Efforts on a macro-level to increase anti-discrimination policies and laws and decrease stigma toward GM individuals may ultimately improve the lives of GM adolescents by reducing exposure to gender minority stressors.
Supporting information
S1 Table. Goodness-of-Fit (QIC) for bivariate associations among gender minority stressors and hypothesized risk and protective factors.
https://doi.org/10.1371/journal.pone.0250500.s001
(DOCX)
Acknowledgments
We would like to thank the many people and organizations who have contributed to this project, including: Laura Bogart, Diane Ehrensaft, Roberta Goldman, Milagros Rosal, Annie Pullen Sansfaçon, Melissa MacNish, Ralph Vetters, Michelle Forcier, the Harvard CFAR Community Advisory Board, the Fenway Youth Community Advisory Board, Kyle Megrath and Athena Edmonds at Greater Boston PFLAG, Shannon Sennott at TransLate Gender, Nick Teich at Camp Aranu’tiq, Julian Dormitzer at the Massachusetts Commission on LGBTQ Youth, Mason Dunn at the Massachuetts Transgender Political Coalition, Effie Molina and the LGBT Liaisons at the Massachusetts Department of Children and Families, Lori Davison at PFLAG Hartford, the Parents of Transgender* Kids Facebook page, Susan Maasch at the Trans Youth Equality Foundation, Eli Godwin, Meghan Doherty, Jax Gonzalez, Til Kus, Julia Medzhitova, LB Moore, Kate Oeser, David Pletta, Elena Rossen, Killian Ruck, Carl Streed Jr., Rachel Kahn, Kate Thomson, Courtney Brown, Lou Arcudi, Edgar Garcia Saiz, Adam Hunter, Jess Kant, Shea Nagle, Neeki Parsa, and the Boston Children’s Hospital Research Informatics Coding Shop. We would also like to thank the Division of Adolescent/Young Adult Medicine at Boston Children’s Hospital and the Harvard Sexual Orientation Gender Identity and Expression Health Equity Research Collaborative (Harvard SOGIE). Finally, we would especially like to thank the families who have participated in the Trans Teen and Family Narratives Project.
References
- 1. Sterzing PR, Ratliff GA, Gartner RE, McGeough BL, Johnson KC. Social Ecological Correlates of Polyvictimization among a National Sample of Transgender, Genderqueer, and Cisgender Sexual Minority Adolescents. Child Abuse Negl. 2017;67:1–12. pmid:28226283
- 2. Eisenberg ME, Gower AL, McMorris BJ, Rider GN, Shea G, Coleman E. Risk and Protective Factors in the Lives of Transgender/Gender Nonconforming Adolescents. J Adolesc Health. 2017;61(4):521–6. pmid:28736148
- 3. Day JK, Fish JN, Perez-Brumer A, Hatzenbuehler ML, Russell ST. Transgender Youth Substance Use Disparities: Results From a Population-Based Sample. J Adolesc Health. 2017;61(6):729–35. pmid:28942238
- 4.
James SE, Herman JL, Rankin S. The Report of the 2015 U.S. Transgender Survey. Washington, DC: National Center for Transgender Equality; 2016.
- 5. Reisner S, Pardo ST, Gamarel K, White Hughto JM, Pardee DJ, Keo-Meier CL. Substance Use to Cope with Stigma in Healthcare Among U.S. Female-to-Male Trans Masculine Adults. LGBT Health. 2015;2(4):324–32. pmid:26788773
- 6. Hatzenbuehler ML, Pachankis JE. Stigma and Minority Stress as Social Determinants of Health Among Lesbian, Gay, Bisexual, and Transgender Youth. Pediatr Clin North Am. 2016;63(6):985–97. pmid:27865340
- 7. Bockting W, Miner MH, Swinburne Romine RE. Stigma, mental health, and resilience in an online sample of the US transgender population. 2016;103(5):943–51.
- 8. Kidd JD, Jackman KB, Wolff M, Veldhuis CB, Hughes TL. Risk and Protective Factors for Substance Use among Sexual and Gender Minority Youth: A Scoping Review. Curr Addict Rep. 2018 Jun;5(2):158–173. Epub 2018 May 26. pmid:30393591; PMCID: PMC6214200.
- 9. Watson RJ, Veale JF, Gordon AR, Clark BA, Saewyc EM. Risk and protective factors for transgender youths’ substance use. Prev Med Rep. 2019;15:100905. pmid:31193847
- 10. Katz-Wise SL, Ehrensaft D, Vetters R. Family Functioning and Mental Health of Transgender and Gender-Nonconforming Youth in the Trans Teen and Family Narratives Project. J Sex Res. 2018; pmid:29336604
- 11. Testa R, Habarth J, Peta J, Balsam K, Bockting W. Development of the Gender Minority Stress and Resilience Measure. Psychology of Sexual Orientation and Gender Diversity. 2(1):65–77.
- 12.
Youth Risk Behavior Survey. Centers for Disease Control and Prevention; 2013.
- 13. Radloff LS. The CES-D Scale: A self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1(3):385–401.
- 14. Spence SH. A measure of anxiety symptoms among children. Behav Res Ther. 1998;36(5):545–66. pmid:9648330
- 15. Essau C a., Sasagawa S, Anastassiou-Hadjicharalambous X. Psychometric properties of the Spence Child Anxiety Scale with adolescents from five European countries. J Anxiety Disord. 2011;25(1):19–27. pmid:20685072
- 16. Hjemdal O, Friborg O, Stiles T. A new scale for adolescent resilience: Grasping the central protective resources behind healthy development. Meas Eval Couns Dev. 2006;39(2):84–96.
- 17. Olson D. FACES IV and the circumplex model: Validation study. J Marital Fam Ther. 2011;37(1):64–80. pmid:21198689
- 18. Zimet GD, Dahlem NW, Zimet SG, Farley GK. The Multidimensional Scale of Perceived Social Support. J Pers Assess. 1988;52(1):30.
- 19.
Hayes AF. Introduction to Mediation, Moderation, and Conditional Process Analysis, Second Edition: A Regression-Based Approach. New York, NY: Guilford Press; 2017.
- 20.
Centers for Disease Control and Prevention. 2019. Youth Risk Behavior Survey Data. Available at: www.cdc.gov/yrbs. Accessed on December 19, 2020.
- 21. Bockting W, Miner MH, Swinburne Romine RE. The Transgender Identity Survey: A Measure of Internalized Transphobia. LGBT Health. 2020;7(1):15–27. pmid:31880493