Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Feb 1.
Published in final edited form as: Psychiatry Res. 2019 Oct 28;284:112664. doi: 10.1016/j.psychres.2019.112664

Sexual Orientation Differences in Non-Suicidal Self-Injury, Suicidality, and Psychosocial Factors among an Inpatient Psychiatric Sample of Adolescents

Jessica R Peters a,*, Ethan H Mereish b, Maya A Krek a, Adam Chuong a, Megan L Ranney c, Joel Solomon d, Anthony Spirito a, Shirley Yen a,e
PMCID: PMC7014818  NIHMSID: NIHMS1543878  PMID: 31732188

Abstract

Within broader community samples, sexual minority adolescents (SMA, e.g., lesbian, gay, bisexual, queer) are at greater risk than their heterosexual counterparts for nonsuicidal self-injury (NSSI) and suicidal thoughts and behaviors. The present study investigated whether sexual minority orientation continues to confer additional risk for these behaviors in an already higher-risk sample of youth. Frequency and function of NSSI, suicidal behavior, and psychosocial factors were assessed in a sample of 52 adolescents (aged 12-18 years) admitted to an inpatient psychiatric unit due to suicide risk; 27 of them identified as SMA, and 25 as heterosexual (HA). Greater proportions of SMA reported engaging in lifetime NSSI, compared to HA, with a greater variety and frequency of NSSI behaviors and greater endorsement of intrapersonal NSSI functions. SMA reported higher levels of suicide ideation than HA, but not suicidal behavior. Group differences in NSSI and SI persisted when controlling for the greater prevalence of abuse and levels of peer-victimization reported by SMA. In inpatient clinical settings, SMA may be more likely than heterosexual youth to engage in NSSI, including more severe forms, and to experience suicide ideation. Providing alternative coping mechanisms may serve as treatment goals for reducing NSSI in SMA.

Keywords: LGBT, LGBTQ, suicide, emotion regulation, deliberate self-harm, gay, bisexual

1. Introduction

Sexual minority adolescents (SMA) are at greater risk than their heterosexual counterparts for nonsuicidal self-injury (NSSI) and suicidal thoughts and behaviors. Specifically, meta-analyses indicate that SMA are three times more likely to endorse suicidal thoughts and behaviors (Marshal et al., 2011) and six times more likely to engage in self-harm behaviors compared to heterosexual adolescents (Batejan, Jarvi, & Swenson, 2014). These adolescents are additionally at greater risk of experiencing abuse, peer victimization (Friedman et al., 2011), depression, and other forms of clinical impairment (Russell & Fish, 2016), and community studies demonstrate mixed findings for these factors as potential contributors to the elevated rates of NSSI and suicidal thoughts and behavior (e.g., Baams, Grossman, & Russell, 2015; Bouris, Everett, Heath, Elsaesser, & Neilands, 2016; Mustanski & Liu, 2012; Puckett et al., 2017; Safren & Heimberg, 1999). As these studies typically use community or epidemiological samples, it is unclear whether these differences between SMA and heterosexual youth persist within higher risk contexts, such as inpatient clinical settings.

Understanding prevalence and correlates of NSSI in at-risk populations is particularly crucial for several reasons. A growing body of research demonstrates that engaging in NSSI, particularly in numerous types and with greater frequency, may increase risk of suicidal thoughts and behaviors (Burke, Hamilton, Cohen, Stange, & Alloy, 2016; Hamza, Stewart, & Willoughby, 2012; J. G. Stewart et al., 2017). By habituating individuals to the process of inflicting harm to the self, acts of NSSI may increase acquired capacity to cause more serious and potentially lethal harm when suicidal (Joiner, Ribeiro, & Silva, 2012). Support for this has been demonstrated specifically among SMA: a history of NSSI was associated with a 10-fold increase in risk for suicidal thoughts and behaviors amongst SMA in a large community sample (Reisner, Biello, Perry, Gamarel, & Mimiaga, 2014). Certain functions of NSSI, in addition to the presence of the self-injurious behavior, may also reflect heightened risk for suicide. For example, in a general population sample, youth who engaged in NSSI for the purposes of self-punishment, reducing dissociation, and/or managing suicidal ideation, were at greatest risk for suicidal behaviors (Paul, Tsypes, Eidlitz, Ernhout, & Whitlock, 2015).

The present study sought to extend the literature on sexual orientation disparities in NSSI and suicidal thoughts and behaviors by examining these behaviors in adolescents on a psychiatric inpatient unit. Group differences observed in community samples, where these behaviors are relatively infrequent, may not generalize to high-risk settings, such as adolescents on a psychiatric inpatient unit, where rates of NSSI and suicidal behavior may be heightened across all groups. Additionally, given the relatively low occurrence of SI and NSSI in the population, most studies on community samples assess only the presence or absence of SI and NSSI; in contrast, an inpatient sample provides an opportunity to examine potential sexual orientation differences in higher severity of SI and NSSI. Determining whether sexual orientation continues to confer heightened risk for self-harm in high risk samples can help guide clinical decision-making and future intervention development and research. Therefore, the present study examined sexual orientation differences among youth in rates and characteristics of NSSI, suicidal behavior, and potentially relevant psychosocial factors, including depression, abuse or victimization, clinical severity, and life satisfaction. We hypothesized that although the sample as a whole would demonstrate elevated symptoms and rates of NSSI and suicidal behavior, SMA would demonstrate higher levels of NSSI, suicide ideation, and suicidal behavior, as well as greater psychopathology and interpersonal victimization, than their heterosexual counterparts. We also examined the forms and functions of NSSI for SMA, as this is pertinent to prevention and intervention efforts.

2. Methods

Data for these analyses were drawn from baseline interviews from a study examining the feasibility of a multi-modal (in-person and text messaging) intervention (Skills to Enhance Positivity in Suicidal Adolescents; STEP) to increase positive affect and decrease suicidality in adolescents recently discharged from a psychiatric inpatient unit. A description of the STEP intervention can be found in Yen et al (in press), which presents findings from the randomized controlled trial.

2.1. Participants

Participants were 52 patients recruited from an adolescent psychiatric inpatient unit, who were admitted due to concerns of suicide risk (e.g., attempt or ideation). Participants were between the ages of 12-18, living at home with a primary guardian, proficient in English, and with access to text-messaging. Exclusion criteria included any psychotic disorder, active psychosis, cognitive or intellectual deficits that hindered understanding of study materials, and being a ward of the state. Participants were screened using medical charts and consultation with the attending psychiatrist, and 76 patient/parent pairs were approached. Of those, 52 (68%) enrolled, signed consent/assent (consent only for 18-year-olds), and completed baseline assessments. Most common reasons patient/parent pairs approached were not subsequently enrolled including failure to complete self-report measures on unit (n=5), inability to contact parents (n=6), and declining to participate (n =9; see Yen et al, 2019, for further details about participant recruitment process).

2.2. Procedures

Each adolescent and participating parent were administered separate baseline assessments, consisting of structured clinical interviews and self-report questionnaires, and were compensated $40 upon completion. This study protocol was approved by the institutional review boards of the hospital recruitment site and the affiliated university.

2.3. Measures

All self-report assessments have demonstrated reliability and validity and were at an appropriate reading level for adolescents.

2.3.1. Demographics.

Gender, sexual orientation, and race were collected via self-report. Ethnicity, living arrangements, parents’ level of education and income, and abuse history were assessed within a structured interview based off the Schedule for Affective Disorders and Schizophrenia for School-Age Children– Present and Lifetime Version (K-SADS-PL; Kaufman et al., 1997). Abuse history was obtained from parent report only due to IRB concerns on direct assessment from adolescent participants; however, given the potential for lack of parental disclosure, this was supplemented with information from medical chart review which includes the intake assessment. For the present study, an “any abuse” variable was created indicating the presence of any abuse history (inclusive of physical, sexual, or psychological abuse and neglect) from either or both of these sources.

2.3.2. Suicidal behavior.

Suicidal behavior characteristics were assessed using the Columbia Suicide Severity Rating Scale (C-SSRS; Posner et al., 2011), a semi-structured interview assessing frequency and intensity of behaviors, including suicide ideation, attempts, aborted and interrupted attempts, and preparatory acts, in the week prior and over lifetime. The C-SSRS was administered to both adolescent and parent; where there were discrepancies between child and parent ratings, consensus ratings were made by the research team based on all available data, including hospital admission notes.

2.3.3. Suicide ideation.

Suicide ideation was assessed using three measures, including the C-SSRS described above. The Suicide Ideation Questionnaire (SIQ; Reynolds, 1988) is a 30-item self-report instrument designed to assess thoughts about suicide experienced by adolescents during the prior month. Internal consistency was excellent in the present sample (Cronbach’s ∝=0.97). SI was also assessed with the Adolescent–Longitudinal Interval Follow-Up Evaluation (A-LIFE; Keller et al., 1987), a semi-structured interview that assesses psychopathology and functioning. Psychiatric status ratings are assigned to SI at each assessment on a 6-point scale (1 = not present, 2 = thoughts of death, 3 = SI without method, 4 = SI with method, 5 = SI with plan, 6 = SI with plan and preparations). The present manuscript utilized the single timepoint A-LIFE rating of SI severity conducted at baseline, referring to the past week.

2.3.4. NSSI.

Lifetime presence of any NSSI behaviors was assessed using the C-SSRS (see above). Lifetime presence of specific NSSI behaviors, NSSI frequency, and functions of NSSI were also assessed on the Inventory of Statements About Self-injury (ISAS; Klonsky & Glenn, 2008). The ISAS assesses 13 functions of NSSI, as well as the frequency of 12 NSSI behaviors (e.g., banging/hitting self, biting, burning, cutting, wound picking, etc.). Participants who endorse one or more of the 12 NSSI behaviors complete the second section assessing 13 potential functions of NSSI, that have been empirically grouped into two factors (Klonsky & Glenn, 2008; Klonsky, Glenn, Styer, Olino, & Washburn, 2015). Intrapersonal functions include affect-regulation, feeling generation (anti-dissociation), anti-suicide, self-care, self-punishment, and marking distress. Interpersonal functions include autonomy, interpersonal boundaries, interpersonal influence, peer-bonding, revenge, sensation seeking, and toughness. Each function is assessed by three items, rated on a 3-point Likert-style scale (0 “not relevant,” to 2 “very relevant” to the individual’s experience of NSSI), and function scores are then summed to create the two factor scores.

2.3.5. Peer victimization.

Peer victimization over the last month was assessed using a 4-item peer victimization scale assessing verbal, relational, and physical victimization (Poteat, Scheer, DiGiovanni, & Mereish, 2014). The scale was designed to assess general victimization, and items reflect common themes in self-report and peer nomination measures of peer victimization. Response options were zero times, 1 or 2 times, 3 or 4 times, 5 or 6 times, and 7 or more times (scaled 0 to 4). Internal consistency for the total scale was good in the present sample (Cronbach’s ∝ = 0.83).

2.3.6. Depression.

The Beck Depression Inventory-II (BDI-II; Beck, Steer, & Brown, 1996) assessed self-reported depression symptoms over the past two weeks. Scores over 20 represent clinically elevated levels of depression. Internal consistency was excellent in the present sample (Cronbach’s ∝ = 0.90).

2.3.7. Clinical impairment.

The Columbia Impairment Scale (CIS; Bird, Shaffer, Fisher, & Gould, 1993) is a 13-item scale that assesses functioning in children and adolescents over “the past week or two” in four areas: interpersonal, psychopathology, school/work, and leisure. Internal consistency was adequate in the present sample (Cronbach’s ∝ = 0.74).

2.3.8. Life satisfaction.

The 5-item Satisfaction with Life Scale (SLS; Diener, Emmons, Larsen, & Griffin, 1985) was administered as a measure of global life satisfaction rather than a measure of any specific life domains. Content included items such as, “The conditions of my life are excellent” and was in reference to the past month. Internal consistency was good in the present sample (Cronbach’s ∝ = 0.86).

3. Results

3.1. Demographics

The mean age of our sample was 15.6 years (SD = 1.47), and 32 (62%) identified as female, 19 (37%) as male, and 1 (1.9%) as other. Gender identity was not further assessed, so we cannot determine gender minority status across the sample. Our sample was mostly White (77%), with 5.8% Black or African American, 3.8% Asian, 3.8% American Indian/Alaskan Native, and 9.6% Other. Of the full sample, 25% identified as Hispanic/Latinx. For household income, 47% of the sample reported under $50,000.

When asked about sexual orientation, 25 (48%) of the adolescents identified as straight (heterosexual adolescents referred to as the HA group). Of the remaining 27 (52%), 2 (3.8% of the full sample) identified as Lesbian or Gay; 16 (31%) as Bisexual; 3 (5.8%) as Other (e.g., pansexual); 2 (3.8%) endorsed “Don’t know/Not sure”; and 4 (7.7%) endorsed “Rather not say.” A sensitivity analyses excluding the participants endorsing “Don’t know” or “Rather not say” produced results consistent with results including these participants; accordingly, the decision was made to include these youth in the SMA category.1

T-tests and chi-square tests were conducted to compare SMA and HA on variables of interest. The SMA and HA did not differ on age, gender, or race; however, significantly more SMA were of Hispanic ethnicity (see Table 1). Race was collapsed into White compared to Non-White, given low rates of non-White participants; no significant differences between SMA and HA were found for this analysis nor when examining race as individual categories.

Table 1.

Sample demographics, for total sample and by sexual orientation.

Overall (n=52)
N (%) /M (SD)		 SMA (n=27)
N (%) /M (SD)		 HA (n=25)
N (%) /M (SD)		 t/χ2
Age 15.6 (1.47) 15.3 (1.51) 16 (1.35) −1.76
Gender: Female 32 (62%) 17 (63%) 15 (60%) 5.06
Male 16 (31%) 6 (22%) 10 (40%)
Transgender/non-binary 4 (8%) 4 (15%) 0 (0%)
Race: White 40 (77%) 20 (74%) 20 (80%) 2.40
Black 3 (6%) 2 (7%) 1 (4%)
Asian 2 (4%) 1 (4%) 1 (4%)
American Indian/Alaskan 2 (4%) 1 (4%) 1 (4%)
Other 5 (10%) 3 (11%) 2 (8%)
Hispanic Ethnicity 13 (25%) 10 (37%) 3 (12%) 4.34*
Open in a new tab
*

= p <.05

Notes: SMA = sexual minority adolescents; HA = heterosexual adolescents. χ2 value for race is presented for White vs Non-White given the low frequencies of other races and thus low power to test across all categories; results remained non-significant if tested across all groups.

3.2. Suicide Ideation and Behavior

Rates of suicide ideation and attempts were examined across the full sample and by SMA status (see Table 2A; effect sizes for group comparisons also reported in table). When assessed with the SIQ, no group differences were found for suicide ideation in the past month; however, SMA reported significantly higher severity ratings of suicide ideation in the past week based on the LIFE interview, with a large effect size. These results converged with the C-SSRS SI severity ratings, which were higher in the SMA for the week prior to hospitalization, but not significantly different between groups for most severe rating over lifetime. Approximately half of the sample had a lifetime suicide attempt history and one quarter had a suicide attempt in the week prior to hospitalization; this did not significantly differ by SMA status.

Table 2.

Life history, psychosocial, suicide, and non-suicidal self-injury variables for the full sample and by sexual orientation.

Overall (N=52)

N (%) /M (SD)		 SMA
(n=27)

N (%) /M (SD)		 HA
(n=25)

N (%)/M (SD)		 Statistic

t/χ2/Mann-
Whitey U		 Effect size

d/φ/r
  A. Suicide/NSSI Behavior
Suicidal Ideation Question (SIQ) 103.02 (42.09) 108.07 (38.18) 97.56 (46.10) 0.90 0.25
Suicide Ideation (PSR) 4.33 (1.29) 4.81 (1.04) 3.80 (1.35) 3.05** 0.84
Suicide Ideation, lifetime (C-SSRS) 4.19 (1.27) 4.33 (1.18) 4.04 (1.37) 0.83
Suicide ideation, past week (C-SSRS) 3.83 (1.51) 4.30 (1.14) 3.32 (1.70) 2.41*
Suicide attempt, lifetime (C-SSRS) 27 (51.9%) 16 (59.3%) 11 (44.0%) 1.21 0.15
Suicide attempt, past week (C-SSRS) 13 (25.0%) 7 (25.9%) 6 (24.0%) 0.03 0.02
NSSI, lifetime (C-SSRS) 44 (84.6%) 27 (100%) 17 (68.0%) 10.21** 0.44
NSSI, past week (C-SSRS) 22 (42.3%) 15 (55.6%) 7 (28.0%) 4.04* 0.28
Any NSSI, lifetime (ISAS) 49 (94.2%) 27 (100%) 22 (88.0%) 3.44 0.26
  B. NSSI Rates/Functions N = 49 n = 27 n = 22
Variety of NSSI (ISAS) 3.61 (2.46) 4.52 (2.17) 2.50 (2.37) 3.11** 0.89
Frequency of NSSI (ISAS) 202.50 (322.24) 254.74 (348.40) 141.17 (283.80)
   median values ‡ 45.50 130.0 12.0 174.00** 0.38
Intrapersonal Functions 17.28 (8.47) 20.59 (5.96) 13.39 (9.41) 3.17** 0.91
Interpersonal Functions 6.94 (7.87) 8.00 (9.14) 5.70 (6.02) 1.03 0.30
  C. Life History/Psychosocial Variables
Any Abuse 29 (55.8%) 19 (70.4%) 10 (40.0%) 4.85* 0.31
Peer Victimization 1.57 (1.28) 1.91 (1.26) 1.20 (1.22) 2.05* 0.57
Depression (BDI) 33.92 (11.57) 34.74 (12.02) 33.04 (11.23) 0.53 0.15
Impairment (CIS) 27.23 (8.99) 28.56 (8.05) 25.80 (9.88) 1.11 0.31
Satisfaction with Life (SLS) 12.62 (6.09) 11.11 (5.25) 14.24 (6.61) 1.90 0.52
Open in a new tab
*

= p <.05,

**

= p <.01

‡

A Mann Whitney U non-parametric test was used to compare median values for Frequency of NSSI given non-normal distribution.

Notes: SMA = sexual minority adolescents; HA = heterosexual adolescents; SIQ = Suicide Ideation Questionnaire; PSR = Psychiatric Status Rating; C-SSRS = Columbia-Suicide Severity Rating Scale; ISAS = Inventory of Statements About Self-Injury; BDI = Beck Depression Inventory; CIS = Columbia Impairment Scale; SLS = Satisfaction with Life Scale; Any Abuse = Chart report or parent endorsement of sexual, physical, psychological abuse and/or neglect.

3.3. Nonsuicidal Self-Injurious Behavior

All of the SMA reported lifetime NSSI, in contrast to 68% of the HA, in the C-SSRS interview (a moderate-to-large effect size); however, when NSSI was assessed with the ISAS self-report measure, which specifically asks about numerous forms of possible NSSI behaviors, the rate of HA endorsing NSSI increased to 88%, and the difference between groups became non-significant (p = .064; see Table 2B). This may be driven in part by the extent to which the ISAS also asks about numerous less severe forms of self-injury (e.g., hair-pulling, pinching, and sticking self with pins). When these less severe NSSI forms were excluded, the difference between groups slightly widened to significant levels (χ2 = 4.68; p = .031), with 100% of SMA and 84% of HA endorsing NSSI (φ = .30; moderate effect size). In contrast, group differences on less severe forms of NSSI were not significant (χ2 = .07; p = .797), with 56% of SMA and 52% of HA endorsing these forms of NSSI. Of the youth endorsing lifetime NSSI on the ISAS, SMA reported engaging in significantly greater frequency of NSSI (moderate effect size) and more types of NSSI (large effect size) compared to the HA.2 Self-reported functions of NSSI were examined for the adolescents reporting a history of NSSI on the ISAS. SMA endorsed intrapersonal functions to a significantly greater extent than the HA (large effect size); however, no significant differences between the groups were found on the interpersonal functions (see Table 2B).

3.4. Life History and Psychosocial Variables

Groups were compared on life history and psychosocial variables (see Table 2C). For the full sample, 19 participants had positive abuse history from parent report on the KSADS and 17 on medical chart review, with 29 (56%) with a positive history from at least one source. SMA reported higher rates of lifetime history of any abuse (70%) than the heterosexual group (40%), with a moderate effect size. SMA reported higher levels of peer victimization than HA, also with a moderate effect size. There were no significant differences in depression, clinical impairment, and life satisfaction ratings between SMA and HA.

Exploratory post-hoc hierarchical regression analyses were conducted to determine if the associations between SMA status and LIFE SI scores, NSSI frequency, NSSI variety, and NSSI intrapersonal functions persisted over and above shared variance with lifetime abuse and peer-victimization (see Table 3). SMA status was entered as a predictor in step 1, and abuse and peer victimization were entered in step 2. For all outcomes, step 2 did not contribute significant incremental validity to the model (p > .472), and SMA status remained a significant predictor with similar effect sizes, while neither abuse nor victimization were significant in any model (p > .256). These findings suggest that while SMA have higher rates of abuse and peer victimization than their heterosexual peers, the association between SMA status and SI and NSSI variables were not attributable to shared variance with abuse history and peer victimization.

Table 3.

Post-hoc hierarchical regression analyses predicting suicidal ideation, NSSI frequency, NSSI variety, and NSSI intrapersonal function from sexual orientation (step 1) and past month peer-victimization and lifetime abuse (step 2).

Dependent
Variable		 Predictors Adjusted
R2 		ΔR2 Step 1 β Step 2 β
Suicide Ideation (PSR score) 1. Sexual Minority .14** .40** .36*
2. Any Abuse .13* .02 −.02
 Peer Victimization .15
Past Week Suicide Ideation (C-SSRS) 1. Sexual Minority .09* .33* .34*
2. Any Abuse .07 .02 −.11
 Peer Victimization .09
NSSI variety (ISAS) 1. Sexual Minority .15** .39** .47**
2. Any Abuse .18** .03 −.11
 Peer Victimization −.13
NSSI frequency (ISAS) 1. Sexual Minority .10* .34* .39**
2. Any Abuse .09 .03 .00
 Peer Victimization −.18
NSSI Intrapersonal Function (ISAS) 1. Sexual Minority .18** .43** .37*
2. Any Abuse .20* .02 .07
 Peer Victimization .12
Open in a new tab
*

= p <.05,

**

= p <.01

Notes: NSSI = Nonsuicidal Self-Injury; PSR = Psychiatric Status Rating; C-SSRS = Columbia-Suicide Severity Rating Scale; ISAS = Inventory of Statements About Self-Injury

4. Discussion

The current study examined differences by sexual orientation in suicidal and self-harming behaviors, as well as a range of potentially related psychosocial variables, in a sample of adolescents recruited on an inpatient psychiatric hospital. In this high-risk sample, SMA did not differ from heterosexual adolescents on rates of prior suicide attempts, but findings for suicide ideation differed for the two measures used. SMA reported higher levels of SI than the heterosexual group on the A-LIFE (a clinician rated measure based on interview report of suicidal ideation severity over the past week) but not the SIQ (a self-report measure based on frequency of thoughts over preceding month). These findings converged with SI severity ratings on the C-SSRS. In that interview, SMA reported experiencing higher levels of SI than the heterosexual group in the week prior to hospitalization, but the group difference on lifetime SI severity was nonsignificant. While all of the inpatient youth report similar lifetime histories of SI, SMA are reporting more intense acute SI at hospitalization, perhaps reflecting a more severe state of crisis. This could be driven by greater experiences of adverse and stigma-specific events precipitating hospitalization, a tendency not to ask for help or disclose SI until more severe, and/or other factors not assessed in the present study. Further investigation is needed to better understand implications of these differences in acute SI prior to hospitalization. SMA also endorsed higher rates of NSSI than heterosexual adolescents, particularly more severe forms of NSSI, as well as higher levels of engaging in NSSI for the intrapersonal functions.

Engaging in NSSI to for intrapersonal functions, such as to achieve management of suicidal urges, as self-punishment, or in order to generate feelings, has been particularly strongly linked to suicide attempts in prior studies of college students (Paul et al., 2015), suggesting that these functions may reflect elevated risk for suicidal thoughts and behaviors. While recently hospitalized adolescents are generally at elevated risk for suicidal behavior regardless of sexual orientation, the differences in motivation for NSSI among SMA (versus heterosexual adolescents) suggests that SMA may be at higher risk of engaging in self-harming behavior to manage their distress. The elevated prevalence of NSSI among SMA, as well as the different reasons for SMA in this group, may heighten risk for and severity of future suicide attempts, through increasing acquired capability for and desensitization to self-destructive behavior (Whitlock et al., 2013). The greater number of methods of NSSI utilized by the SMA may also reflect elevated risk for suicide attempts (J. G. Stewart et al., 2017). Future work should examine these associations longitudinally in high-risk youth to understand whether NSSI may contribute to the elevated suicide rates found in SMA over time.

Several possibilities may explain the lack of observed differences in suicide attempts between SMA and HA in the present sample, despite the elevated SI and NSSI in the SMA group. Given that this is a high severity sample recruited based on suicidal behavior, it is possible that the sample approaches a ceiling effect for attempts, typically a relatively rare behavior, whereas there is more variability to capture for SI and NSSI. There also may be only a small effect on attempts in this more severe sample that the present sample is underpowered to detect. Additionally, one of the specific functions of NSSI endorsed to a greater extent by the SMA is using NSSI to manage suicide-related thoughts, suggesting that for these SMA experiencing SI, these high rates of NSSI may reflect emotion regulation strategies that function to prevent suicide attempts, at least in the short-term. If so, engaging in NSSI may then paradoxically increase risk over time as described above (Whitlock et al., 2013) or have other maladaptive effects; longitudinal research would be needed to examine that possibility, as well as any other potential developmental or age-related effects on differences suicide attempts between these groups.

Although SMA reported higher incidence of physical abuse and peer-victimization than heterosexual participants — confirming others’ findings that this group is at particular risk for abuse from family and peers (see Friedman et al., 2011 for a meta-analysis) — the presence of abuse and victimization was not, surprisingly, associated with NSSI or SI after adjustment for SMA status, and the magnitude and significance of the relations between SMA status and both NSSI and SI were unchanged by controlling for abuse and victimization. Other studies have implicated these elevated rates of abuse and the associated experiences of minority stress as risk factors for suicide and self-harming behavior in SMA among general populations (Baams, Grossman, & Russell, 2015; Bouris, Everett, Heath, Elsaesser, & Neilands, 2016; Mustanski & Liu, 2012; Puckett et al., 2017); however, in our study on inpatient participants, the link between sexual orientation and NSSI and SI appeared unrelated to abuse and peer-victimization. However, the present study is underpowered for mediation or moderation analyses, so further research with larger samples powered for mediational and moderation analyses is needed to determine whether trauma history and peer victimization may directly contribute to the elevated rates of NSSI used to manage distress in SMA or affect which SMA engage in NSSI or SI. Future research should also incorporate a measure of lifetime peer victimization, as well as recent victimization, given that important experiences shaping current behavior may have occurred prior to the previous month’s assessment period of the measure utilized in the present study. More nuanced assessment of forms of victimization, including sexual minority-specific victimization and types and intensity of behaviors involved, would also be informative. Clinically, careful assessment of interpersonal victimization and stressors may be particularly important for SMA to determine whether these factors may be present and contributing to distress.

While SMA demonstrated a trend toward lower levels of satisfaction with life, SMA and heterosexual adolescents did not differ in their levels of symptoms of depression and clinical impairment. These findings likely reflect the high severity of symptoms present across most adolescents warranting inpatient levels of care. These results are also consistent with prior findings demonstrating that suicide risk differences between SMA and heterosexual adolescents may persist even when differences in depression are accounted for (Safren & Heimberg, 1999). Furthermore, while understanding how sexual minority status may confer risk within the high-risk population of recently hospitalized adolescents is important for clinicians in these settings, restriction of range in some variables due to the heightened severity may obscure some associations that would be observed in broader samples across ranges of clinical severity.

Future, larger studies should examine additional constructs as potential factors in elevated self-harm and SI, such as those that have been found to promote resilience in community SMA samples, including family support and connectedness (Reisner et al., 2014; Taliaferro & Muehlenkamp, 2017). Further work could also include a priori assessment of gender identity, as well as measures beyond self-report, such as task-based assessments of suicide and self-harm biases. Additionally, the SMA in this sample were primarily bisexual-identified girls, so these findings may not generalize to youth of other sexual minority identities or genders; further research is needed with broader samples. Similarly, our sample was predominately White; as such, additional research is warranted to understand the experiences of SMA who are racial and ethnic minorities. We were unable to record information from patient and parent pairs who were approached for the study but declined to participate, so we cannot determine whether self-selection bias may have occurred and further affected demographic representation or other aspects of the study.

These findings suggest that sexual minority orientation may exacerbate self-harm risk, perhaps through increased need to regulate distress, in addition to the underlying psychopathology common across all youth in these settings. Interventions such as Dialectical Behavior Therapy (Linehan, 2014; Rathus & Miller, 2014) that specifically target difficulties with emotion regulation and distress tolerance, challenge self-punitive cognitions, and provide alternative coping skills to NSSI, might be particularly useful for increasing skills for managing stressors. However, evidence of differential efficacy of these types of interventions across sexual minorities suggest they may also need to be adapted specifically to meet SMA needs and address relevant concerns (Beard et al., 2017).

Highlights.

  • Examined self-harming behavior in sexual minority and heterosexual high-risk youth

  • No group differences found in depression, clinical impairment, or life satisfaction

  • Sexual minorities endorsed greater degree of non-suicidal self-injury

  • Sexual minorities reported higher levels of suicide ideation but not behavior

  • Abuse and victimization did not account for sexual minorities’ greater self-harm

Acknowledgements

Funding: This work was supported by grants from the National Institute of Mental Health (R34 MH101272; T32MH019927; K23MH112889) and the National Institute on Alcohol Abuse and Alcoholism (K08AA025011).

Footnotes

Conflicts of interest: The authors have no conflicts of interest to declare.

1

Youth who are questioning their orientation or who have conflict around it have been shown to demonstrate even higher levels of risk than those self-identified as sexual minorities, including depression (Birkett, Espelage, & Koenig, 2009), difficulties with adjustment in response to bullying and victimization (Poteat, Aragon, Espelage, & Koenig, 2009), and suicidal ideation (Batejan et al, 2015). Less extant work examines risk of individuals who select not to report; however, prior research has demonstrated non-disclosure to be a form of stigma management among sexual minority individuals (Schrimshaw, Downing, & Cohn, 2018). This suggests that the youth choosing not disclose may also be at elevated risk for minority stress and at-risk behaviors, and thus of particular relevance to the research question.

2

For youth reporting NSSI, frequency scores ranged from 1 to 1323 and reflected a highly positively skewed count variable. Accordingly, a Mann Whitney U non-parametric test was utilized and median values, in addition to means, are provided in the table. Significant results were also achieved with an alternate approach of t-tests using log-transformed variables.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Baams L, Grossman AH, Russell ST 2015. Minority stress and mechanisms of risk for depression and suicidal ideation among lesbian, gay, and bisexual youth. Developmental Psychology 51, 688–696. 10.1037/a0038994 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Batejan KL, Jarvi SM, Swenson LP 2015. Sexual Orientation and Non-Suicidal Self-Injury: A Meta-Analytic Review. Archives of Suicide Research, 19,131–150. 10.1080/13811118.2014.957450 [DOI] [PubMed] [Google Scholar]
  3. Beard C, Kirakosian N, Silverman AL, Winer JP, Wadsworth LP, Björgvinsson T 2017. Comparing treatment response between LGBQ and heterosexual individuals attending a CBT- and DBT-skills-based partial hospital. Journal of Consulting and Clinical Psychology, 85,1171–1181. 10.1037/ccp0000251 [DOI] [PubMed] [Google Scholar]
  4. Beck AT, Steer RA, Brown GK 1996. Beck Depression Inventory-II. San Antonio: Psychological Corporation. [Google Scholar]
  5. Bird HR, Shaffer D, Fisher P, Gould MS 1993. The Columbia Impairment Scale (CIS): Pilot findings on a measure of global impairment for children and adolescents. International Journal of Methods in Psychiatry Research, 3,167–176. [Google Scholar]
  6. Birkett M, Espelage DL, Koenig B 2009. LGB and questioning students in schools: The moderating effects of homophobic bullying and school climate on negative outcomes. Journal of Youth and Adolescence, 38(7), 989–1000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bouris A, Everett BG, Heath RD, Elsaesser CE, Neilands TB 2016. Effects of victimization and violence on suicidal ideation and behaviors among sexual minority and heterosexual adolescents. LGBT Health, 3, 153–161. 10.1089/lgbt.2015.0037 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Burke TA, Hamilton JL, Cohen JN, Stange JP, Alloy LB 2016. Identifying a physical indicator of suicide risk: Non-suicidal self-injury scars predict suicidal ideation and suicide attempts. Comprehensive Psychiatry, 65, 79–87. 10.1016/j.comppsych.2015.10.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cohen CJ, Fowler M, Medenica VE, Rogowski JC 2018. Millennial Attitudes on LGBT Issues: Race, Identity, and Experience. GenForward. [Google Scholar]
  10. Diener E, Emmons RA, Larsen RJ, Griffin S, 1985. The Satisfaction With Life Scale. Journal of Personality Assessment, 49, 71–75. 10.1207/s15327752jpa4901_13 [DOI] [PubMed] [Google Scholar]
  11. Field A, 2013. Discovering Statistics Using IBM SPSS Statistics (4 ed.). London: Sage. [Google Scholar]
  12. Friedman MS, Marshal MP, Guadamuz TE, Wei C, Wong CF, Saewyc E, Stall R 2011. A meta-analysis of disparities in childhood sexual abuse, parental physical abuse, and peer victimization among sexual minority and sexual nonminority individuals. American Journal of Public Health, 101, 1481–1494. 10.2105/AJPH.2009.190009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gratz K, Lacroce D, Gunderson J 2006. Measuring changes in symptoms relevant to borderline personality disorder following short-term treatment across partial hospital and intensive outpatient levels of care. Journal of Psychiatric Practice, 12, 153–159. [DOI] [PubMed] [Google Scholar]
  14. Grov C, Bimbi DS, Nanín JE, Parsons JT, 2006. Race, ethnicity, gender, and generational factors associated with the coming-out process among gay, lesbian, and bisexual individuals. Journal of Sex Research, 43, 115–121. [DOI] [PubMed] [Google Scholar]
  15. Hamza CA, Stewart SL, Willoughby T, 2012. Examining the link between nonsuicidal self-injury and suicidal behavior: a review of the literature and an integrated model. Clin Psychol Rev, 32, 482–495. 10.1016/j.cpr.2012.05.003 [DOI] [PubMed] [Google Scholar]
  16. Joiner TE, Ribeiro JD, Silva C, 2012. Nonsuicidal self-injury, suicidal behavior, and their co-occurrence as viewed through the lens of the interpersonal theory of suicide. Current Directions in Psychological Science, 21, 342–347. 10.1177/0963721412454873 [DOI] [Google Scholar]
  17. Kaufman J, Birmaher B, Brent D, Rao U, Flynn C, Moreci P, et al. 1997. Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version (K-SADS-PL): Initial reliability and validity data. Journal of the American Academy of Child & Adolescent Psychiatry, 36, 980–988. [DOI] [PubMed] [Google Scholar]
  18. Keller MB, Lavori PW, Friedman B, Nielsen E, Endicott J, McDonald-Scott P, Andreasen NC, 1987. The Longitudinal Interval Follow-up Evaluation: A comprehensive method for assessing outcome in prospective longitudinal studies. Archives of General Psychiatry, 44, 540–548. 10.1001/archpsyc.1987.01800180050009 [DOI] [PubMed] [Google Scholar]
  19. Klonsky ED, Glenn CR, 2008. Assessing the functions of non-suicidal self-injury: Psychometric properties of the Inventory of Statements About Self-injury (ISAS). Journal of Psychopathology and Behavioral Assessment, 31, 215–219. 10.1007/s10862-008-9107-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Klonsky ED, Glenn CR, Styer DM, Olino TM, Washburn JJ, 2015. The functions of nonsuicidal self-injury: Converging evidence for a two-factor structure. Child and adolescent psychiatry and mental health, 9, 44 10.1186/s13034-015-0073-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Linehan MM 2014. DBT® Skills Training Manual, Second Edition. Guilford Publications. [Google Scholar]
  22. Marshal MP, Dietz LJ, Friedman MS, Stall R, Smith HA, McGinley J, et al. 2011. Suicidality and depression disparities between sexual minority and heterosexual youth: A meta-analytic review, J of Adol Health, 49, 115–123. 10.1016/j.jadohealth.2011.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Mustanski B, Liu RT, 2012. A longitudinal study of predictors of suicide attempts among lesbian, gay, bisexual, and transgender youth. Archives of Sexual Behavior, 42, 437–448. 10.1007/s10508-012-0013-9 [DOI] [PubMed] [Google Scholar]
  24. Paul E, Tsypes A, Eidlitz L, Ernhout C, Whitlock J, 2015. Frequency and functions of non-suicidal self-injury: Associations with suicidal thoughts and behaviors. Psychiatry Research, 225, 276–282. 10.1016/j.psychres.2014.12.026 [DOI] [PubMed] [Google Scholar]
  25. Posner K, Brown GK, Stanley B, Brent DA, Yershova KV, Oquendo MA, et al. , 2011. The Columbia-Suicide Severity Rating Scale: Initial validity and internal consistency findings from three multisite studies with adolescents and adults. The American Journal of Psychiatry, 168,1266–1277. 10.1176/appi.ajp.2011.10111704 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Poteat VP, Aragon SR, Espelage DL, Koenig BW 2009. Psychosocial concerns of sexual minority youth: Complexity and caution in group differences. Journal of Consulting and Clinical Psychology, 77(1), 196. [DOI] [PubMed] [Google Scholar]
  27. Poteat VP, Scheer JR, DiGiovanni CD, Mereish EH, 2014. Short-term prospective effects of homophobic victimization on the mental health of heterosexual adolescents. Journal of Youth and Adolescence, 43, 1240–1251. 10.1007/s10964-013-0078-3 [DOI] [PubMed] [Google Scholar]
  28. Puckett JA, Horne SG, Surace F, Carter A, Noffsinger-Frazier N, Shulman J, et al. , 2017. Predictors of sexual minority youth’s reported suicide attempts and mental health. Journal of Homosexuality, 64, 697–715. 10.1080/00918369.2016.1196999 [DOI] [PubMed] [Google Scholar]
  29. Rathus JH, Miller AL, 2014. DBT® Skills Manual for Adolescents. Guilford Publications. [Google Scholar]
  30. Reisner SL, Biello K, Perry NS, Gamarel KE, Mimiaga MJ, 2014. A compensatory model of risk and resilience applied to adolescent sexual orientation disparities in nonsuicidal self-injury and suicide attempts. American Journal of Orthopsychiatry, 84, 545–556. 10.1037/ort0000008 [DOI] [PubMed] [Google Scholar]
  31. Reynolds W, 1988. Suicide Ideation Questionnaire (SIQ). Odessa, FL: Psychological Assessment Resources. [Google Scholar]
  32. Russell ST, Fish JN, 2016. Mental health in lesbian, gay, bisexual, and transgender (LGBT) youth. Annual Review of Clinical Psychology, 12, 465–487. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Safren SA, Heimberg RG 1999. Depression, hopelessness, suicidality, and related factors in sexual minority and heterosexual adolescents. Journal of Consulting and Clinical Psychology, 67, 859–866. [DOI] [PubMed] [Google Scholar]
  34. Schrimshaw EW, Downing MJ, Cohn DJ 2018. Reasons for non-disclosure of sexual orientation among behaviorally bisexual men: Non-disclosure as stigma management. Archives of Sexual Behavior, 47(1), 219–233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Spirito A, Williams CA, Stark LJ, Hart KJ, 1988. The Hopelessness Scale for Children: psychometric properties with normal and emotionally disturbed adolescents. Journal of Abnormal Child Psychology, 16, 445–458. [DOI] [PubMed] [Google Scholar]
  36. Stewart JG, Esposito EC, Glenn CR, Gilman SE, Pridgen B, Gold J, Auerbach RP, 2017. Adolescent self-injurers: Comparing non-ideators, suicide ideators, and suicide attempters. Journal of Psychiatric Research, 84,105–112. 10.1016/j.jpsychires.2016.09.031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Tabachnick BG, Fidell LS, 2000. Using Multivariate Statistics (4 ed.). Allyn & Bacon. [Google Scholar]
  38. Taliaferro LA, Muehlenkamp JJ, 2017. Nonsuicidal self-injury and suicidality among sexual minority youth: Risk factors and protective connectedness factors. Acad Pediat 17, 715–722. 10.1016/j.acap.2016.11.002 [DOI] [PubMed] [Google Scholar]
  39. Whitlock J, Muehlenkamp J, Eckenrode J, Purington A, Baral Abrams G, Barreira P, Kress V, 2013. Nonsuicidal self-injury as a gateway to suicide in young adults. The Journal of Adolescent Health, 52, 486–492. 10.1016/j.jadohealth.2012.09.010 [DOI] [PubMed] [Google Scholar]
  40. Yen S, Ranney ML, Krek M, Peters JR, Mereish EH, Tezanos KM, Kahler CW, Solomon JB, Beard C, Spirito A, in press. Skills to enhance positivity in suicidal adolescents: Results from a pilot randomized clinical trial. Journal of Positive Psychology, doi: 10.1080/17439760.2019.1615105 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES