Beyond a Single Risk Category: Within-Group Correlates of Bullying Victimization and Suicidality in Transgender Adolescents

Abstract

Aim: Transgender adolescents experience elevated rates of bullying and adverse mental health outcomes compared with cisgender peers, yet most research has treated this group as homogeneous. We aimed to identify sociodemographic and physical health correlates of bullying victimization and mental health outcomes within the transgender adolescent population. Methods: A secondary analysis of the 2023 national Youth Risk Behavior Survey (YRBS) was conducted with an analytic sample of 612 high school students who self-identified as transgender. Five outcomes were examined: school-based bullying, electronic bullying, persistent sadness or hopelessness, suicidal ideation, and suicide attempt. The complex sampling design was specified using sampling weights, strata, and primary sampling units together, with transgender students defined as a subpopulation (domain) of the full dataset. Survey-weighted multivariable logistic regression models were estimated with sex, grade, race/ethnicity, sexual identity, and body mass index (BMI) entered simultaneously; multivariable models were fitted on complete cases (analytic n = 429–439). Results: The weighted national prevalence of transgender identity was 3.3% (95% CI 2.7–3.9). Prevalence rates were substantially elevated: 42.6% reported school-based bullying, 33.8% electronic bullying, 71.0% persistent sadness or hopelessness, 53.8% suicidal ideation, and 25.7% suicide attempt. Sexual identity was associated with internalizing outcomes, with bisexual (aOR 6.23, p = 0.001), gay/lesbian (aOR 4.94, p = 0.017), and “some other way” (aOR 7.17, p = 0.001) identities showing higher odds of persistent sadness or hopelessness than heterosexual transgender students, corresponding to adjusted predicted probabilities of 75%, 71%, and 77%, respectively, versus 37% in the heterosexual group. Black transgender students had lower odds of reporting school-based bullying (aOR 0.17, p = 0.008); this estimate is based on a small subgroup (n = 31). Compared with 9th grade, only the 11th-grade contrast was significant for school-based bullying, and the 10th- and 11th-grade contrasts for electronic bullying. The a priori BMI hypothesis was not supported at the variable level (joint Wald test non-significant in all models); only the single obesity contrast for school-based bullying reached significance (aOR 2.08, p = 0.035). Conclusion: The transgender adolescent population is heterogeneous; sexual identity, race/ethnicity, and grade are differentially associated with outcomes, supporting individualized risk assessment in clinical and school health settings.