Disclosure of Self-Injurious Thoughts and Behaviors Across Sexual and Gender Identities

We recruited participants through paid Instagram ads, which directed individuals to a Qualtrics-based eligibility screening survey. Eligibility was determined on the basis of age (13–17 years old), English language comprehension, living in the United States, having a SITB history, and having a history of mental health treatment. Qualified individuals who provided assent received a link to the full study on Qualtrics; parental consent was waived for this study. All individuals who completed the screener and/or full study were provided with mental health resources; all participants who completed the full study were entered into a lottery for $25 gift cards. Full details about recruitment, screening, and procedures are provided by Fox et al (K. Fox, A. H. Bettis, T. A. Burke, E. A. Hart, S. B. Wang. Exploring adolescent experiences with and impacts of disclosing self-injurious thoughts and behaviors across settings, unpublished observations). All procedures were approved by the Harvard University Institutional Review Board.

Participants were included in the analytic sample if they responded to questions about sexual orientation and gender identity. Because demographics were assessed at the end of the study, this yielded a total of 931 (mean age = 15.7, SD = 1.11) of the original 1706 participants who started the study. Participants reported substantial diversity across sexual orientation and gender, with moderate racial and ethnic diversity (see Table 1). At the intersection of sexual orientation and gender identity, 117 (12.6%) participants identified as cisgender heterosexual, 338 (36.3%) as GM, and 476 (51.1%) as SM. Participants’ average socioeconomic status was 5.82 (SD = 1.68, range = 1–10), as assessed with the MacArthur Scale of Subjective Social Status – Youth Version.¹⁶  Higher scores indicate higher social standing.

We assessed eligibility using items from the Self-Injurious Thoughts and Behavior Interview-Revised,¹⁷  including items assessing lifetime history of NSSI, suicidal ideation, and suicide attempts. We used single items to assess lifetime history of mental health treatment.

We assessed age, race, ethnicity, sexual orientation, and socioeconomic status. We assessed gender using a measure with a wide range of gender identities.¹⁸  Participants were able to select multiple genders and sexual orientations.

Several items were used to assess SITB disclosure. SITB history was first determined with the question “Have you ever told anyone about times where you [purposely hurt yourself without wanting to die/had thoughts of killing yourself/tried to kill yourself]?” Second, participants were asked to think of times they had engaged in a given SITB in their life. Third, participants were instructed to report who they had told about each SITB engagement using a Likert scale ranging from 0 (never) to 4 (every time); options included parent or guardian, therapist, psychiatrist or doctor (who prescribes mental health medication), doctor you see for check-ups and physicals (ie, pediatrician), another adult you trust, sibling, friend, acquaintance, someone you met online, and other. To maximize statistical power for paired analyses, we focused analysis on disclosure to friends and parents or guardians.

We asked participants to report their likelihood of disclosing each SITB with a therapist in the future on a scale from 0 (not at all) to 4 (extremely).

We asked participants if providers (including pediatricians and/or medical providers and mental health professionals and/or therapists) had ever asked them about each SITB, separately, with response options of “yes,” “no,” and “don’t remember”; current analyses only considered yes and no responses. We also asked participants how honest they were in disclosing each SITB to pediatricians and/or medical providers and mental health professionals and/or therapists using a scale ranging from 0 (not at all honest) to 4 (completely honest).

We adapted items from Hom et al⁸  to assess barriers to disclosure to mental health professionals and/or therapists. We added items relevant to adolescents (eg, telling parents or guardians) and aimed to reflect additional forms of stigma (eg, shame).

The Expanded Everyday Discrimination Scale measures the extent to which participants experience chronic, everyday forms of discrimination (eg, “You receive poorer service than other people at restaurants or stores”). Participants were asked to rate each experience from 1 (never) to 5 (almost every day). The original scale has been validated and has strong psychometric properties in an adolescent sample.¹⁹ 

First, we examined whether therapists’ and pediatricians’ and/or medical providers’ assessments of specific SITBs differed across identities, including participants identifying as SM or GM (ie, SGM) and cisgender heterosexual (ie, non-SGM). Results were mixed. Pearson χ² tests revealed a significant main effect of SGM status on therapist assessment of NSSI (⁠$χ22$ = 9.84, P = .007) but no main effect of therapists’ assessments of suicide ideation or attempts. For the purpose of SGM subgroup analyses, the GM group included individuals identifying as GM, some of whom also identified as SM; the SM group only included those identifying as SM (but not GM). Holm-corrected post hoc tests revealed that therapists were more likely to directly ask participants who identified as GM compared with SM about their NSSI histories, with no other differences across groups emerging. Pearson χ² tests revealed no main effect of pediatricians’ and/or medical providers’ assessments of any SITB across identities (P > .40).

Second, we tested for differences in the frequency of SITB disclosure to friends and parents or guardians across groups. Across SM, GM, and non-SGM participants, identical disclosure patterns emerged across SITBs (see Fig 1). Third, we examined whether SGM identity impacted self-reported honesty in disclosure to both therapists and pediatricians and/or medical providers. Analysis of variance tests revealed that SGM identity did not significantly impact disclosure honesty across any SITB to either provider (see Fig 2).

Fourth, we explored whether barriers to SITB disclosure to therapists differed across SGM and non-SGM youth. See Table 2 for the top 5 barriers endorsed among participants identifying as SGM and non-SGM. χ² tests revealed that only 2 of these barriers received significantly different endorsements across SGM identity groups. Specifically, main effects were observed for the belief that the therapist would tell a parent or guardian (⁠$χ22$ = 8.11, P = .017) and for fear of worrying the parent or guardian (⁠$χ22$ = 7.25, P = .027). Holm-corrected post hoc tests revealed that non-SGM participants were significantly less likely to endorse the belief that a therapist would share with a parent or guardian compared with GM participants (P = .0346), and relatedly, non-SGM participants were significantly less concerned that this information would worry a parent or guardian compared with both SM (P = .024) and GM participants (P = .037).

Fifth, we examined whether the frequency of discrimination experiences differed across groups and whether there were group differences in the frequency of discrimination impacting willingness and/or desire to disclose SITBs to therapists across SGM identities. As shown in Fig 3, results of Welch’s one-way analysis of variance were significant. Holm-corrected post hoc tests revealed that GM participants reported significantly greater discrimination experiences than those identifying as SM (no other significant differences emerged). Regarding the impact of such experiences on future disclosure willingness, Holm-corrected post hoc tests revealed significantly greater impact for GM-identifying participants compared with both SM and non-SGM participants. See Supplemental Table 3 for the reasons for discrimination reported within the study sample.

Sixth, we conducted Pearson correlations to examine the impact of discrimination experiences on previous and future willingness to disclose SITBs both in the full sample and in only SGM participants. Among the full sample, small negative correlations were observed between discrimination experiences and past SITB disclosure; significant associations were not observed with future willingness to disclose (Fig 4). Our findings partially supported our hypotheses; when considering only participants identifying as SGM, these associations became largely insignificant, with one exception: odds of disclosing suicide attempts in the future.

The present study investigated the differences in SITB disclosure patterns between cisgender heterosexual, SM, and GM youth. Results suggested few differences in disclosure of SITBs to friends and parents or guardians and in disclosure honesty to therapists, as well as with related primary barriers to disclosure. However, our findings provide some evidence supporting the minority stress model, revealing that minority stress experiences may interfere with SITB disclosure, particularly among GM youth.

Although patterns of disclosure were similar across SM, GM, and non-SGM youth and across provider types, some notable differences emerged. GM participants were more likely to express the concern that a therapist would share with a parent or guardian as a barrier to disclosure. Both SM and GM participants were more likely to report the related barrier to disclosure that this information would worry a parent or guardian compared with non-SGM participants. Given that family support is tied to improved mental health outcomes in SGM youth,^20,21  SGM youth who experience SITBs may be particularly worried about maintaining their relationship with their parents. Fear of disrupting their caregiver relationship(s) further may contribute to their SITB disclosure decision. Although no differences in rates of SITB disclosure to parents emerged, additional research examining family-related factors associated with SITB disclosure decision-making is warranted. Results highlight the need to provide psychoeducational resources for parents both to mitigate such barriers and to respond to initial disclosure in a manner that will facilitate ongoing disclosure.

GM youth also reported significantly greater discrimination experiences than those identifying as SM, and these discrimination experiences impacted their willingness to disclose SITBs in the future more so than their SM and non-SGM peers. Results are somewhat consistent with previous work that found minority stress was associated with lower likelihood of SITB disclosure.¹⁵  Prospective research is needed to determine if discrimination experiences predict disclosures over time, particularly in GM youth, and to explore specific mechanisms that may play a role in this relationship. In addition, although both SM and GM youth report elevated rates of SITBs, some research suggests that GM youth may be at particularly high risk because of higher levels of discrimination and lower levels of acceptance across family, peers, and society.⁴  Additional research is necessary to elucidate how GM youth’s specific experiences with discrimination may impact SITB disclosure to inform clinical guidelines and intervention efforts for these youth.

Across the full sample, previous discrimination experiences were associated with lower rates of previous SITB disclosure honesty but not future disclosure willingness. Reasons for this discrepancy necessitate future prospective research. For example, it is possible that positive disclosure experiences may influence individuals’ willingness to disclose in the future, despite previous experiences of discrimination. Alternatively, the relationship between minority stress exposure and disclosure honesty may depend on their temporal connection or may differ on the basis of the specific domain of minority stress assessed (eg, family rejection, health care discrimination, peer victimization). Disentangling discrimination factors, both specific to LGBTQ+ youth and experienced across other minoritized identities, that contribute to how youth make decisions to disclose SITBs will be an important area of future investigation.

Finally, although most associations among discrimination experiences, disclosure honesty, and future disclosure willingness were not statistically significant in this sample of SGM youth, one important association emerged. Greater discrimination experiences in SGM youth were associated with lower intention of disclosing suicide attempts, but not suicidal ideation or NSSI, in the future. Results did not replicate those of Chang et al,¹⁵  who found that greater minority stress was associated with lower intention to disclose suicidal thoughts in the future. Lack of replication here may relate to differences in measurement of minority stress; Chang et al¹⁵  employed a specific measure of SGM stress, whereas the measure used in this study was not specific to SGM experiences.

Strengths include the recruitment of a large sample of adolescents with a history of SITBs and the successful employment of targeted recruitment efforts to increase the proportion of LGBTQ+ youth in our sample. We also assessed disclosure and related barriers across a range of SITBs because both rates of disclosures and barriers to disclosures may differ across SITB types.

That said, several issues limit generalizability of results. Most notably, we did not assess whether participants were “out” to close others, including parents, friends, or providers; results may differ as a function of outness and may not generalize to older LGBTQ+ populations. Additionally, this study relied on adolescent self-report. Observational and/or medical record data from health care providers (eg, assessing rates of provider SITB assessment) may be useful in future research. Future researchers should also consider incorporating qualitative methods to identify additional LGBTQ+-specific factors, as well as individual differences, that may influence SITB disclosure. The cross-sectional nature of this study precludes causal inferences; prospective data are needed to better understand disclosure likelihood as well as barriers and their association with SITBs over time. General psychopathology and distress may have impacted retrospective recall related to disclosure. However, we did not measure current levels of psychopathology or distress in the current study and thus cannot rule out their impact on findings. Although recruiting a large proportion of LGBTQ+ youth was a strength of this study, there remain limitations regarding our assessment of multiple minority identities and minority stress experiences. Because of the sample’s racial and ethnic homogeneity, we were unable to explore whether differences also emerged for racial and ethnic minority groups. We were also unable to examine differences within subgroups of LGBTQ+ identities. Research suggests bisexual individuals are at elevated risk for SITBs compared with both those who identify as lesbian or gay and those who identify as heterosexual²² ; whether similar differences emerge in rates of or barriers to disclosure in this subpopulation is an important area for investigation.

Intervention targeting SITBs in youth relies, in part, on adolescents’ willingness to share their SITB experiences. Results from the current study highlight the importance of considering the parent-child relationship when inquiring about and treating youth SITBs. Parents may benefit from direct guidance around how to support their child in discussing SITB experiences, particularly parents of GM youth. In addition, clinical interventions that address effective ways to cope specifically with discrimination experiences are needed.

Disclosure honesty is a critical component of SITB assessment and intervention in youth. In the current study, we explored whether SM, GM, and cisgender heterosexual youth report differences in SITB disclosures, and few differences emerged. Findings highlight the need for prospective research examining SITB disclosures over time as well as mechanisms that may contribute to disclosure likelihood and honesty in youth.

GM

gender minority

LGBTQ+

lesbian, gay, bisexual, transgender, gender nonconforming, queer, and questioning

NSSI

nonsuicidal self-injury

SGM

sexual and gender minority

SITB

self-injurious thought and behavior

SM

sexual minority