Trajectories of Gender Identity and Depressive Symptoms in Youths

Importance: Concerns about the mental health of youths going through gender identity transitions have received increased attention. There is a need for empirical evidence to understand how transitions in self-reported gender identity are associated with mental health. Objective: To examine whether and how often youths changed self-reported gender identities in a longitudinal sample of sexual and gender minority (SGM) youths, and whether trajectories of gender identity were associated with depressive symptoms. Design, Setting, and Participants: This cohort study used data from 4 waves (every 9 months) of a longitudinal community-based study collected in 2 large cities in the US (1 in the Northeast and 1 in the Southwest) between November 2011 and June 2015. Eligible participants included youths who self-identified as SGM from community-based agencies and college groups for SGM youths. Data analysis occurred from September 2022 to June 2023. Exposure: Gender identity trajectories and gender identity variability. Main Outcomes and Measures: The Beck Depression Inventory for Youth (BDI-Y) assessed depressive symptoms. Gender identity variability was measured as the number of times participants' gender identity changed. Hierarchical linear models investigated gender identity trajectories and whether gender identity variability was associated with depressive symptoms over time. Results: Among the 366 SGM youths included in the study (mean [SD] age, 18.61 [1.71] years; 181 [49.4%] assigned male at birth and 185 [50.6%] assigned female at birth), 4 gender identity trajectory groups were identified: (1) cisgender across all waves (274 participants ), (2) transgender or gender diverse (TGD) across all waves (32 participants), (3) initially cisgender but TGD by wave 4 (ie, cisgender to TGD [28 participants]), and (4) initially TGD but cisgender by wave 4 (ie, TGD to cisgender [32 participants]). One in 5 youths (18.3%) reported a different gender identity over a period of approximately 3.5 years; 28 youths varied gender identity more than twice. The cisgender to TGD group reported higher levels of depression compared with the cisgender group at baseline (Β = 4.66; SE = 2.10; P = .03), but there was no statistical difference once exposure to lesbian, gay, bisexual, and transgender violence was taken into account (Β = 3.31; SE = 2.36; P = .16). Gender identity variability was not associated with within-person change in depressive symptoms (Β = 0.23; SE = 0.74; P = .75) or the level of depressive symptoms (Β = 2.43; SE = 2.51; P = .33). Conclusions: These findings suggest that gender identity can evolve among SGM youths across time and that changes in gender identity are not associated with changes in depressive symptoms. Further longitudinal work should explore gender identity variability and adolescent and adult health.