RESEARCH PAPER

Nonsuicidal Self-Injury and Suicidality Among

Sexual Minority Youth: Risk Factors and Protective

Connectedness Factors Lindsay A. Taliaferro, PhD, MPH; Jennifer J. Muehlenkamp, PhD

From the Department of Internal Medicine, College of Medicine, University of Central Florida (Dr Taliaferro), Orlando, Fla; and Psychology Department, University of Wisconsin-Eau Claire, Eau Claire, Wis (Dr Muehlenkamp) The authors have no conflicts of interest to disclose. Address correspondence to Lindsay A. Taliaferro, PhD, MPH, Department of Internal Medicine, College of Medicine, University of Central Florida, 6900 Lake Nona Blvd, BBS 426, Orlando, FL 32827 (e-mail: [email protected]). Received for publication August 4, 2016; accepted November 6, 2016.

ABSTRACT

OBJECTIVE: We investigated differences in prevalence of re- petitive nonsuicidal self-injury (NSSI), suicidal ideation, and a suicide attempt among youth who identified as bisexual, gay/lesbian, and questioning. In addition, we examined which types of social connections were associated with reduced risk of repetitive NSSI and suicidality among youth who identified with a specific sexual minority group. METHODS: Data came from the 2013 Minnesota Student Sur- vey. The analytic sample included 77,758 students in grades 9 and 11. Connectedness factors included parent connectedness, teacher caring, connectedness to other nonparental adults, school safety, and friend caring. Logistic regression analyses, stratified according to sexual minority group, determined social connectedness factors associated with repetitive NSSI, suicidal ideation, and a suicide attempt, as well as moderating effects of significant connectedness factors on different risk factors (depression, anxiety, bullying, and violence victimization). RESULTS: Approximately 3% identified as bisexual or questioning their sexual orientation, and <1% identified as

gay/lesbian. Sexual minority youth, particularly bisexual youth, were significantly more likely than heterosexual youth to report repetitive NSSI and suicidality. Effects of connectedness varied across sexual minority groups and outcomes on the basis of types of connections. Parent connectedness emerged as a robust protective factor for all self-harm behaviors across all sexual minority groups. Feeling connected to nonparental adults and safe at school represented additional factors that reduced risk of repetitive NSSI and suicidality among certain groups. CONCLUSIONS: In addition to facilitating connections between youth and parents, clinicians might consider encouraging sexual minority youth to remain connected to trusted nonparental adults who could offer support and care. Schools might consider implementing sociocultural norms of acceptance, tolerance, and positive identity development to reduce risk of self-harm.

KEYWORDS: connectedness; nonsuicidal self-injury; protective factor; sexual minority; suicide

ACADEMIC PEDIATRICS 2017;17:715–722

WHAT’S NEW

We addressed gaps in the literature regarding connect- edness factors that provide subgroups of sexual minor- ity youth protection against repetitive nonsuicidal self-injury and suicidality. Professionals could use these findings to inform programming, research, and clinical work with youth, parents, and schools.

SUICIDAL BEHAVIOR AND nonsuicidal self-injury (NSSI; deliberate destruction of body tissue without sui- cidal intent and not socially sanctioned such as cutting, burning, or scraping one’s skin) represent significant public health problems among adolescents.1,2 Suicide has become the second leading cause of death among adolescents aged 15 to 24 years. In 2015, 17.7% of high school students reported seriously considering attempting suicide, and 8.6% actually attempted suicide during the previous 12 months.3 Further, approximately 18% of adolescents report

engaging in NSSI.1 This behavior might increase risk of suicide,4 especially among youth who engage in repetitive self-injury, defined as 10 or more acts.5

Certain populations show disparities in suicidality and NSSI. Research supports an association between identifying as a sexual minority and increased risk of suicidal behavior among youth.6 Still, the American Foundation for Suicide Prevention called for research that helps confirm a connection between identifying as a sexual minority and suicidal behavior across a wider range of samples (http://afsp.org/wp-content/uploads/2016/ 03/Focus-Grants-Policy-07-01-15.doc). Researchers also identified sexual minority youth as a subpopulation who demonstrates greater prevalence of NSSI.7,8

According to the Minority Stress Theory, stigma, preju- dice, and discrimination create a hostile and stressful social environment, which contributes to mental health problems among sexual minority populations.9 Research supports the application of the minority stress theory in understanding

ACADEMIC PEDIATRICS Copyright ª 2016 by Academic Pediatric Association 715

Volume 17, Number 7 September–October 2017

suicidality and NSSI among youth who identify as a sexual minority.9,10 Investigators showed that minority stress (eg, experiences of discrimination and expectations of rejection) increased risk of NSSI10 and suicidal ideation11

among sexual minority youth through its effect on perceived burdensomeness (perception that one’s existence burdens family, friends, and/or society).

Most research on adolescents in general, and sexual minority populations in particular, focuses on health risks.12 In contrast, resilience research focuses on positive features of development. Commonly observed factors associated with resilience among adolescents are strong connections to parents, nonparental adults, school, and pro- social friends.13 Research supports connectedness as a pro- tective factor that mitigates against suicidality among youth in general.2,14,15 Thus, the Centers for Disease Control and Prevention identified connectedness as a public health-based approach to address suicidal behavior.16 Still, limited research has examined protective relationships between connectedness and NSSI and suici- dality among youth who identify as a sexual minority. Within the limited existing research, investigators have re- ported that greater family connectedness, adult caring, and/ or school safety/connectedness are associated with reduced risk of suicidality among sexual minority youth.14,17,18

Although researchers examined different types of social connections, they have not determined which type of con- nections provide youth who identify as a sexual minority with the greatest protection against suicidality and NSSI. This line of research might inform program planning and interventions designed to enhance connections and pro- mote resilience among sexual minority youth, as well as professional practice among those who work with this pop- ulation. Further, a dearth of epidemiological research exists involving population-based samples that examined the association between sexual minority status and NSSI among adolescents. Finally, research remains limited by use of small convenience samples and analytic approaches that involve combining sexual minority groups, which limit the generalizability of results. Very limited research exists on subgroups of sexual minority youth, masking important differences between groups.19,20

To address gaps in the literature, we sought to answer 2 research questions: (1) What is the prevalence of repetitive NSSI, suicidal ideation, and a suicide attempt among youth who identify as bisexual, homosexual, and questioning their sexual orientation?, and (2) Which assessed connect- edness factors show relationships with reduced risk of re- petitive NSSI, suicidal ideation, and a suicide attempt for youth who identify with a specific sexual minority group?

METHODS

STUDY DESIGN AND SAMPLE

Data came from the 2013 Minnesota Student Survey, a population-based survey administered every 3 years to stu- dents in grades 5, 8, 9, and 11.21 The Minnesota Depart- ments of Education, Health, Human Services, and Public Safety conduct this anonymous, paper-and-pencil survey,

which is used to examine a range of health behaviors, as well as potential risk and protective factors. In 2013, the participation rate was 84% of all school districts (n ¼ 162,034), representing 67% of Minnesota school stu- dents across these 4 grades. Most school districts used pas- sive parental consent, and students provided assent. Additional details regarding the survey methodology are available elsewhere.21 The University of Central Florida institutional review board approved this secondary data analysis. For the current analysis, only ninth and 11th grade stu-

dents were included because of differences in survey con- tent for younger students (n ¼ 79,339). The sample included slightly more male (50.2%) than female (49.8%) students, and more students in grade 9 (53.4%) than grade 11 (46.6%). Students reported their race/ ethnicity as white non-Hispanic (73.8%), Asian (5.6%), black (5.1%), Native American (1.0%), Pacific Islander (0.2%), Hispanic (6.8%), and mixed race (6.5%). The ana- lytic sample included 77,758 students in grades 9 and 11 who responded to an item regarding sexual orientation. This sample was significantly more likely than the total sample of ninth and 11th graders to identify as female, in grade 11, and white.

MEASURES

Students were classified into sexual minority groups on the basis of 1 question in the survey: “Which of the following best describes you?” Response options were het- erosexual (straight), bisexual, gay or lesbian, or not sure (questioning). The outcome variables reflected repetitive NSSI, suicidal ideation, and suicide attempt within the past year. NSSI was assessed using the item: “During the past 12 months, how many times did you do something to purposely hurt or injure yourself without wanting to die, such as cutting, burning, or bruising yourself on pur- pose?” Response options provided ranges from 0 to 20 or more times. Students who responded “10 to 19 times” or “20 or more times” were categorized as engaging in repet- itive NSSI and compared with those who responded less than 10 times/never. This criterion conforms to current literature on correctly classifying individuals with a possible NSSI disorder,5 because those who engage in repetitive self-injury show increased risk of psychopathol- ogy, compared with those who engage in fewer NSSI epi- sodes.22 Suicidality was assessed using the following items: “Have you ever thought about killing yourself?” and “Have you ever tried to kill yourself?” Response options for both items were “no,” “yes, more than a year ago,” and “yes, during the past year.” To assess current behavior, we classified students into groups on the basis of their responses of “yes, during the past year.” We assessed 5 continuous connectedness factors exam-

ined in other research with adolescents.14,15 Three items were used to create a composite parent connectedness variable. Students indicated how often they could talk with their mother and father about problems they were having (2 items; response option range: “most of the

716 TALIAFERRO AND MUEHLENKAMP ACADEMIC PEDIATRICS

time” to “mother/father is not around”), as well as how much they believed their parents cared about them (response option range on a 5-point Likert-type scale: “very much” to “not at all”). Perceptions of caring from teachers and friends were assessed using items that inquired about how much students believed “teachers/other adults at school” and “friends,” respectively, cared about them. Response options appeared on a 5-point Likert- type scale from “very much” to “not at all.” Connectedness to other nonparental adults was assessed using 2 items on how much students believed “other adult relatives” and “adults in your community” cared about them, with response options on a 5-point Likert-type scale from “very much” to “not at all.” The last connectedness factor examined involved school safety, which was assessed using the item: “How much do you agree or disagree with the statement I feel safe at school?” Response options were presented on a 4-point Likert-type scale from “strongly agree” to “strongly disagree.”

Risk factors associated with NSSI and suicidality were assessed using similar measures as used in other epidemi- ological research with adolescents.15 Depressive and anxi- ety symptoms were assessed with items asking students whether or not (yes/no) they experienced significant prob- lems with feeling very “trapped, lonely, sad, blue, depressed, or hopeless about the future” (depression) and “anxious, nervous, tense, scared, panicked, or like some- thing bad was going to happen” (anxiety) during the past 12 months. Bully victimization was assessed using the item: “During the past 30 days, how often have other stu- dents harassed or bullied you because you are gay or lesbian or because someone thought you were?” Response options were presented on a 5-point Likert-type scale from “never” to “every day.” Students who responded once or more were categorized as experiencing bullying. Finally, the experience of violence at school was assessed using the item: “During the past 30 days, how often have other students at school pushed, shoved, slapped, hit, or kicked you when they weren’t kidding around?” Response options appeared on a 5-point Likert-type scale from “never” to “every day.” We categorized students as victims of violence at school if they responded once or more. Demographic variables included grade (ninth vs 11th grade), sex (female vs male), and race/ethnicity (white vs nonwhite).

DATA ANALYSIS

Analyses were performed using SPSS version 23 (IBM Corp, Armonk, NY). Chi-square and analysis of variance tests were used to examine differences between the sexual minority groups on the study outcomes and primary inde- pendent variables. For the primary analyses, we performed logistic regression analyses stratified according to sexual minority group, with listwise deletion. Analyses were per- formed in 3 stages. First, we examined relationships be- tween each connectedness factor and each outcome 1 at a time, controlling for the demographic variables and risk factors. During the second stage, connectedness factors

that were significantly associated with an outcome were all entered together into a model, with the demographic and risk factor variables. Finally, during the third stage, we examined interactions, separately, between any risk and connectedness factors that remained significant in the final model to determine the moderating effects of connectedness factors on identified risk factors. Signifi- cance was determined by a P value < .05. We only present results for sexual minority groups because they were most relevant to our research questions. Results for heterosexual youth are available upon request.

RESULTS Almost 3.0% (n ¼ 2223) of students identified as

bisexual, 2.7% (n ¼ 2082) indicated they were questioning their sexual orientation, and 0.8% (n ¼ 655) identified as gay/lesbian. Sexual minority youth reported significantly lower levels on all the connectedness factors and higher scores on all the risk factors, compared with their hetero- sexual counterparts (Tables 1 and 2). Further, students who identified as a sexual minority were significantly more likely than heterosexual youth to report NSSI and suicidality during the previous 12 months (Table 3). In particular, bisexual youth showed a significantly greater likelihood of reporting repetitive self-injury (23.7%), sui- cidal ideation (43.8%), and a suicide attempt (19.5%), compared with their heterosexual, gay/lesbian, and ques- tioning counterparts. In Tables 4 through 6 the findings from the final logistic

regression models are shown. Among bisexual youth (Table 4), repetitive NSSI was significantly associated with all of the examined risk factors—depressive symp- toms, anxiety symptoms, bullying, and violence victimiza- tion. Parent connectedness and perceived caring from friends emerged as significant protective factors. Risk fac- tors significantly associated with suicidal ideation included depressive and anxiety symptoms, and being the victim of bullying. Connectedness factors that protected against sui- cidal ideation included greater connectedness to parents and nonparental adults. Further, we found a significant interaction between connectedness to nonparental adults and depressive symptoms (P ¼.012). Among those who re- ported depressive symptoms, youth who reported higher levels of connectedness to nonparental adults were less likely to report suicidal ideation. Risk factors significantly associated with attempting suicide also included depres- sive and anxiety symptoms, being a victim of bullying, and being the victim of violence at school. Parent connect- edness emerged as the only significant protective factor, when all connectedness factors were examined together. Further, parent connectedness moderated the effect of depressive symptoms on suicide attempts (P ¼ .016). Among those who reported depressive symptoms, higher levels of parent connectedness was significantly associated with reduced risk of a suicide attempt. Among youth who identified as gay/lesbian (Table 5),

repetitive NSSI was significantly associated with depres- sive symptoms, and school safety emerged as a significant

ACADEMIC PEDIATRICS PREVENTING NSSI AND SUICIDALITY AMONG SEXUAL MINORITY YOUTH 717

protective factor. A significant interaction emerged be- tween school safety and depressive symptoms (P ¼ .008), indicating that for gay/lesbian youth who reported depres- sive symptoms, those with higher levels of perceived school safety were less likely to report repetitive NSSI. Suicidal ideation was significantly associated with depres- sive symptoms, and perceived caring from friends and school safety emerged as significant connectedness factors. Perceived safety at school also protected against a suicide attempt, and depressive and anxiety symptoms increased risk of a suicide attempt. No significant interactions emerged in analyses used to examine suicidality among gay/lesbian students.

Among youth who indicated they were questioning their sexual orientation (Table 6), depressive symptoms and be- ing the victim of bullying represented significant risk fac- tors associated with engaging in repetitive NSSI. Connectedness to parents and nonparental adults repre- sented significant protective factors against NSSI. Further, a significant interaction emerged between connectedness to nonparental adults and being the victim of bullying (P ¼ .021). Among youth who reported questioning their sexual orientation and being the victim of bullying because of their perceived sexual orientation, those with higher levels of connectedness to nonparental adults were less likely to report repetitive self-injury. Suicidal ideation was significantly associated with depressive symptoms, anxiety, and bully victimization. Parent connectedness and perceived teacher caring represented the most impor- tant protective factors, and a significant interaction emerged between parent connectedness and being bullied because of one’s sexual orientation (P ¼ .032). For youth who reported feeling uncertain about their sexual orienta- tion and being bullied for their perceived sexual orienta- tion, those with higher levels of parent connectedness were less likely to report suicidal ideation. A suicide attempt was significantly associated with depressive

symptoms and being the victim of violence at school, whereas connectedness to parents and nonparental adults represented significant protective factors.

DISCUSSION In this study we sought to address gaps in the literature

regarding risk of self-harm among youth who identify as a sexual minority, and social connectedness factors that might protect against these behaviors. Findings support previous research showing that youth who identify as a sexual minority show significantly greater risk of repetitive NSSI, suicidal ideation, and a suicide attempt, compared with their heterosexual counterparts.20,23 In the current study, youth who identified with certain sexual minority groups reported 2 to 8 times greater prevalence of these self-harm behaviors as heterosexual youth. Consistent with previous research, bisexual youth were

most likely to report NSSI and suicidality.20,23 These findings suggest the need to focus public health prevention programming on this high-risk subpopulation, as well as conduct research with these vulnerable adoles- cents to further identify factors that increase and decrease risk of self-destructive behaviors. Bisexual youth might face unique challenges during adolescence associated with their sexual orientation, such as lack of acceptance by heterosexual as well as homosexual peers, thus, experi- encing less connectedness within the lesbian, gay, bisexual, transgender, and queer (LGBTQ) community.20,24 The current data further underscore this idea by showing that nonacceptance through bully victimization because of one’s sexual orientation represented a significant risk factor for all 3 self-harm behaviors among bisexual youth. Prevention programming and clinical interventions could target the distinct interpersonal and intrapersonal chal- lenges experienced by youth who identify as bisexual. Further, professionals who provide care to bisexual youth

Table 1. Differences in Perceived Connectedness Among Sexual Minority Youth

Variable

Heterosexual Bisexual Gay or Lesbian Not Sure (Questioning)

Mean (SD)

Parent connectedness 12.8 (2.3) 10.6 (2.8)a 10.8 (3.0)a 11.6 (2.8) Teacher caring 9.2 (2.1) 8.2 (2.4)a 8.4 (2.5)a 8.9 (2.4) Connectedness to nonparental adults 7.3 (1.9) 5.8 (2.1) 6.1 (2.3) 6.5 (2.2) School safety 3.4 (0.7) 3.1 (0.8)a 3.0 (0.9)a 3.2 (0.8) Friend caring 4.1 (1.0) 3.8 (1.2)a 3.7 (1.3)a,b 3.7 (1.2)b

Variables sharing the same superscript letter were not significantly different at P < .05.

Table 2. Prevalence of Risk Factors Among Sexual Minority Youth

Depressive Symptoms Anxiety Symptoms Bully Victim Violence Victim at School

n (%) P n (%) P n (%) P n (%) P

Sexual Orientation <.001 <.001 <.001 <.001 Heterosexual 19,575 (28.8) 20,793 (30.6) 3172 (4.4) 7136 (10.0) Bisexual 1438 (69.8) 1328 (64.5) 841 (38.6) 423 (19.3) Gay or lesbian 345 (57.7) 324 (54.4) 370 (58.3) 142 (22.2) Not sure (questioning) 910 (48.8) 853 (45.9) 412 (20.6) 380 (18.9)

Post hoc chi-square tests showed a statistically significant difference (P < .001) between each group for all of the variables.

718 TALIAFERRO AND MUEHLENKAMP ACADEMIC PEDIATRICS

might benefit from training to help ensure they possess the understanding and capacity to provide effective support and care to these young people.20

Findings from this study suggest connectedness to important individuals and/or institutions within adoles- cents’ social ecologies might buffer the effect of stressors experienced from identification with a sexual minority group. Although connectedness showed a protective effect for each sexual minority subgroup, the effect varied across groups and outcomes on the basis of types of connections. For example, perceived connection to peers was only protective for NSSI among bisexual youth; perceptions of teachers’ caring was only protective against suicidal ideation among questioning youth; and school safety was protective against all 3 outcomes for gay/lesbian youth. These unique differences in the types of connections across sexual minority groups are impor- tant for prevention, because they suggest a variety of pos- itive relationships are necessary for reducing risk for self-harm. Prevention strategies need to focus on fostering positive connections across a range of social roles, relationships, and environments. Sexual minority youth want health care providers to promote their wellness and resilience,25 which includes prosocial con-

nections across adolescent social ecologies. Thus, clini- cian training should address strength-based approaches to the care of youth, especially vulnerable groups such as the LGBTQ population.26

Aligning with previous research,14 parent connected- ness emerged as an important protective factor among most youth. For bisexual youth, perceiving strong con- nections with parents mitigated risk of a suicide attempt conferred by depressive symptoms, and for questioning youth, parent connectedness moderated the relationship between being a victim of bullying and suicidal ideation. In this study, the effects of parent connectedness were not as strong among gay/lesbian youth. Still, supportive parent-child relationships likely reduce risk of NSSI and suicidality. Thus, primary prevention strategies should focus on fostering positive connections with par- ents to reduce suicide attempts and ideation, as well as NSSI, among sexual minority youth. The findings also might support the use of family-based therapy strategies when working with sexual minority youth,27 particularly those reporting symptoms of depression, bully victimiza- tion, NSSI, or suicidal behaviors. In addition to the protective effect of parents, feeling

connected to nonparental adults as well as connected and

Table 3. Prevalence of Repetitive NSSI, Suicidal Ideation, and Suicide Attempts Among Sexual Minority Youth

Repetitive NSSI Suicidal Ideation Suicide Attempt

% (n) P % (n) P % (n) P

Sexual Orientation <.001 <.001 <.001 Heterosexual 2.6 (1798) 9.5 (6478) 2.7 (1859) Bisexual 23.7 (491) 43.8 (900) 19.5 (402) Gay or lesbian 16.3 (98) 29.0 (175) 12.5 (76) Not sure (questioning) 9.2 (173) 22.0 (409) 6.9 (130)

NSSI indicates nonsuicidal self-injury.

Post hoc chi-square tests showed a statistically significant difference (P < .001) between each group for all of the variables.

Table 4. Risk and Protective Factors Associated With Repetitive Self-Injury, Suicidal Ideation, and Suicide Attempt Among Bisexual Youth

Repetitive NSSI

(n ¼ 491) Suicidal Ideation

(n ¼ 900) Suicide Attempt

(n ¼ 402) Odds Ratio (95% Confidence Interval)

Risk factor Depressive symptoms 5.61 (3.53–8.91)*** 5.67 (4.16–7.74)*** 3.00 (1.96–4.61)*** Anxiety symptoms 2.10 (1.50–2.92)*** 1.95 (1.50–2.52)*** 1.91 (1.34–2.72)*** Bully victim 1.34 (1.04–1.73)* 1.40 (1.11–1.76)** 1.51 (1.16–1.97)** Violence victim at school 1.39 (1.02–1.88)* 1.15 (0.86–1.54) 1.68 (1.24–2.28)***

Protective factor Parent connectedness 0.94 (0.89–0.99)* 0.89 (0.85–0.93)*** 0.93 (0.88–0.98)** Teacher caring 0.95 (0.89–1.01) 1.02 (0.96–1.08) 1.03 (0.96–1.10) Other nonparental adult connectedness 0.93 (0.86–1.01) 0.92 (0.85–0.99)* 0.92 (0.85–1.01) School safety 0.92 (0.78–1.09) 0.89 (0.76–1.04) 0.91 (0.76–1.08) Friend caring 0.89 (0.80–0.99)* 0.94 (0.85–1.04) 0.94 (0.84–1.06)

Significant interactions Depressive symptoms � non-parental adult

connectedness 1.21 (1.04–1.39)*

Depressive symptoms � parent connectedness 1.17 (1.03–1.32)* NSSI indicates nonsuicidal self-injury.

All models were controlled for sex, grade, and race/ethnicity. Results shown in bold were statistically significant at P < .05.

*P < .05.

**P # .01.

***P # .001.

ACADEMIC PEDIATRICS PREVENTING NSSI AND SUICIDALITY AMONG SEXUAL MINORITY YOUTH 719

safe at school, might represent additional factors that reduce risk of repetitive NSSI and suicidality among sexual minority youth. Feeling connected to and supported by prominent individuals and institutions within one’s envi- ronment likely offers adolescents who identify as a sexual minority healthful coping strategies to manage the stressors and negative emotions they experience, thus, mitigating risk for self-harm. Perceptions of a safe school environment was particularly protective against NSSI and suicidality among youth who identified as gay/lesbian, which is consistent with other research showing that a sup- portive social culture can protect against suicide risk.28 For example, gay/lesbian students in schools with gay-straight

alliances show lower rates of suicidality.29 Although in the current study we did not explicitly ask about perceptions of social culture and acceptance of sexual minority youth in the school, such aspects likely contribute to perceiving a school environment as safe for sexual minority youth. Follow-up studies should carefully examine specific fac- tors influencing perceptions of school safety among LGBTQ youth and how each of those might influence risk behaviors. Prevention strategies that promote devel- oping sociocultural norms of acceptance, tolerance, and positive identity development, particularly within the school setting, might have success in decreasing suicide risk for sexual minority youth.

Table 6. Risk and Protective Factors Associated With Repetitive Self-Injury, Suicidal Ideation, and Suicide Attempt Among Youth Question-

ing Their Sexual Orientation

Repetitive NSSI

(n ¼ 173) Suicidal Ideation

(n ¼ 409) Suicide Attempt

(n ¼ 130) Odds Ratio (95% Confidence Interval)

Risk factor Depressive symptoms 3.40 (1.92–6.00)*** 6.60 (4.45–9.78)*** 2.77 (1.46–5.24)** Anxiety symptoms 1.39 (0.88–2.20) 1.85 (1.35–2.55)*** 1.53 (0.90–2.62) Bully victim 1.62 (1.05–2.49)* 1.79 (1.27–2.50)*** 1.58 (0.97–2.56) Violence victim at school 1.19 (0.75–1.91) 1.12 (0.78–1.62) 2.74 (1.68–4.48)***

Protective factor Parent connectedness 0.86 (0.79–0.92)*** 0.90 (0.85–0.95)*** 0.86 (0.79–0.94)*** Teacher caring 0.95 (0.86–1.05) 0.90 (0.84–0.97)** 1.07 (0.95–1.19) Other non-parental adult connectedness 0.85 (0.75–0.96)** 0.93 (0.84–1.02) 0.77 (0.67–0.89)*** School safety 1.02 (0.80–1.32) 1.06 (0.87–1.28) 0.97 (0.73–1.28) Friend caring 0.96 (0.82–1.13) 1.00 (0.88–1.13) 1.11 (0.92–1.33)

Significant interactions Bully victim � nonparental adult connectedness 1.23 (1.03–1.46)* Bully victim � parent connectedness 1.11 (1.01–1.21)* NSSI indicates nonsuicidal self-injury.

All models were controlled for sex, grade, and race/ethnicity. Results shown in bold were statistically significant at P < .05.

*P < .05.

**P # .01.

***P # .001.

Table 5. Risk and Protective Factors Associated With Repetitive Self-Injury, Suicidal Ideation, and Suicide Attempt Among Gay or Lesbian

Youth

Repetitive NSSI

(n ¼ 98) Suicidal Ideation

(n ¼ 175) Suicide Attempt

(n ¼ 76) Odds Ratio (95% Confidence Interval)

Risk factor Depressive symptoms 3.87 (1.75–8.54)*** 6.79 (3.64–12.67)*** 4.17 (1.72–10.07)** Anxiety symptoms 1.87 (0.92–3.81) 1.60 (0.93–2.74) 2.28 (1.06–4.91)* Bully victim 1.12 (0.59–2.12) 0.93 (0.96–1.54) 1.60 (0.79–3.23) Violence victim at school 0.94 (0.48–1.84) 1.01 (0.56–1.54) 1.37 (0.72–2.63)

Protective factor Parent connectedness 0.93 (0.83–1.03) 0.96 (0.88–1.04) Teacher caring 0.90 (0.79–1.03) 0.97 (0.88–1.04) Other nonparental adult connectedness 1.02 (0.88–1.19) School safety 0.56 (0.40–0.80)*** 0.63 (0.47–0.85)** 0.65 (0.47–0.91)* Friend caring 0.83 (0.69–0.99)* 0.83 (0.68–1.02)

Significant interaction Depressive symptoms � school safety 2.93 (1.32–6.50)** NSSI indicates nonsuicidal self-injury.

All models were controlled for sex, grade, and race/ethnicity. Results shown in bold were statistically significant at P < .05. Missing esti-

mates indicate the associated variable was not included in the final model because the factor was not significant in preliminary analyses.

*P < .05.

**P # .01.

***P # .001.

720 TALIAFERRO AND MUEHLENKAMP ACADEMIC PEDIATRICS

Further, clinical practices that address barriers to optimal care of sexual minority youth, including facili- tating discussions about sexual orientation and attractions, as well as supportive connections in adolescents’ lives, should help reduce health disparities and promote resil- ience among this population.26,30 Kitts31 reported that only 57% of physicians knew an association exists between identifying as a sexual minority and risk of suicide, and most did not think they possessed the skills needed to address issues related to sexual orientation with adoles- cents. The limited time spent in medical school on LGBTQ content likely contributes to discomfort and potentially cultural incompetence by well-meaning physicians to address unique needs and health disparities among sexual minority youth.

31,32 These factors might contribute to

adolescents’ reluctance to seek needed care and share their sexual orientation with clinicians.33 Fortunately, providing clinicians with LGBTQ-specific training en- hances their comfort and knowledge to work with sexual minority youth.34 Thus, medical educators are encouraged to enhance their curricula in this area,32 and clinicians in practice are encouraged to engage in continuing medical education (eg, Fenway Guide to Lesbian, Gay, Bisexual, and Transgender Health35).

This study included several strengths and weaknesses. Strengths included the large population-based sample, which provided sufficient power to stratify analyses ac- cording to sexual minority group and examine NSSI and suicide outcomes during the previous 12 months, filling important gaps in the literature. This survey also included different types of connectedness factors we could examine together to determine those that showed the strongest rela- tionships with our outcomes for each sexual minority group. Still, even with much power, some groups remained relatively small because of the low base rate of the out- comes. Also, the findings might not generalize to youth outside of Minnesota. Additional research supporting these findings among nationally representative samples of ado- lescents would enhance the generalizability of the results. In addition, the cross-sectional nature of the data precluded us from establishing cause and effect relationships. Future research should include longitudinal designs that track youth over time. Because of the limited research in this area, analyses also remained exploratory rather than hypothesis-testing. Finally, all variables were measured us- ing 1 or 2 items. The reliability of these measures might not be as high as multi-item scales that measure the constructs.

Despite these limitations, professionals who work to prevent NSSI and suicide could use these findings to inform programming, research, and clinical work with youth, parents, and schools. For example, public health ef- forts to address NSSI and suicidality among sexual minor- ity youth could involve schools implementing and enforcing bullying and violence prevention programs that address stereotypes and homophobic perceptions, as well as include components that facilitate students’ feelings of safety in and connectedness to their schools and adults in the school. In addition, primary care and mental health care providers who work with youth must ensure parents

recognize and appreciate the importance of their continued presence and support in the lives of their adolescent chil- dren. However, clinicians also should consider that sexual minority youth might create “families of choice” that pro- vide greater support and connectedness than their families of origin.28 Public health professionals and clinicians might consider encouraging youth who identify as a sexual minority to remain connected to trusted nonparental adults who could offer support and care. In particular, bisexual youth might benefit from increased connectedness to pro- social adults who might help buffer against negative emo- tions induced by perceived burdensomeness.11,20

Collectively, promoting positive social connections, particularly with adults, within supportive environments appear to constitute important protective elements against self-harm for sexual minority youth.

ACKNOWLEDGMENTS The authors thank Ashley Cadman for her assistance reviewing related

research literature for this article.

Financial disclosure: Funding was provided, in part, by a Dana Grant

from the University of Tampa to the first author.

SUPPLEMENTARY DATA Supplementary data related to this article can be found

online at http://dx.doi.org/10.1016/j.acap.2016.11.002.

REFERENCES 1. Muehlenkamp JJ, Claes L, Havertape L, et al. International preva-

lence of adolescent non-suicidal self-injury and deliberate self-

harm. Child Adolesc Psychiatry Ment Health. 2012;6:10.

2. Whitlock J, Wyman PA, Moore SR. Connectedness and suicide pre-

vention in adolescents: pathways and implications. Suicide Life

Threat Behav. 2014;44:246–272.

3. Centers for Disease Control and Prevention. Youth Risk Behavior

Surveillance - United States, 2015. MMWR Surveill Summ. 2016;

65:1–174.

4. Joiner T, Ribeiro JS, Silva C. Nonsuicidal self-injury, suicide

behavior, and their co-occurrence as viewed through the lens of the

interpersonal theory of suicide. Curr Dir Psychol Sci. 2012;21:

342–347.

5. Muehlenkamp J, Brausch A. Reconsidering criterion A for the diag-

nosis of non-suicidal self- injury disorder. J Psychopathol Behav

Assess. doi:10.1007/s10862-016-9543-0.

6. Pl€oderl M, Wagenmakers EJ, Tremblay P, et al. Suicide risk and sex- ual orientation: a critical review. Arch Sex Behav. 2013;42:715–727.

7. Reisner S, Biello K, Perry N, et al. A compensatory model of risk and

resilience applied to adolescent sexual orientation disparities in non-

suicidal self-injury and suicide attempts. Am J Orthopsychiatry. 2014;

84:546–556.

8. Tsypes A, Lane R, Paul E, et al. Non-suicidal self-injury and suicidal

thoughts and behaviors in heterosexual and sexual minority young

adults. Compr Psychiatry. 2016;65:32–43.

9. Meyer IH. Prejudice, social stress, and mental health in lesbian, gay,

and bisexual populations: conceptual issues and research evidence.

Psychol Bull. 2003;129:674–697.

10. Muehlenkamp J, Hilt L, Ehlinger P, et al. Nonsuicidal self-injury in

sexual minority college students: a test of theoretical integration.

Child Adolesc Psychiatry Ment Health. 2015;9:16.

11. Baams L, Grossman AH, Russell ST. Minority stress and mechanisms

of risk for depression and suicidal ideation among lesbian, gay, and

bisexual youth. Develop Psychol. 2015;51:688–696.

ACADEMIC PEDIATRICS PREVENTING NSSI AND SUICIDALITY AMONG SEXUAL MINORITY YOUTH 721

12. Colpitts E, Gahagan J. The utility of resilience as a conceptual frame-

work for understanding and measuring LGBTQ health. Int J Equity

Health. 2016;15:60.

13. Masten A. Ordinary magic: lessons from research on resilience in hu-

man development. Educ Can. 2009;49:28–32.

14. Eisenberg ME, Resnick MD. Suicidality among gay, lesbian and

bisexual youth: the role of protective factors. J Adolesc Health.

2006;39:662–668.

15. Taliaferro LA, Muehlenkamp JJ, Borowsky IW, et al. Factors distin-

guishing youth who report self-injurious behavior: a population-

based sample. Acad Pediatr. 2012;12:205–213.

16. Centers for Disease Control and Prevention. Strategic direction for

the prevention of suicidal behavior: promoting individual, family,

and community connectedness to prevent suicidal behavior. Available

at: https://www.cdc.gov/violenceprevention/pdf/suicide_strategic_

direction_full_version-a.pdf. Accessed July 1, 2016.

17. Duong J, Bradshaw C. Associations between bullying and engaging in

aggressive and suicidal behaviors among sexual minority youth: the

moderating role of connectedness. J Sch Health. 2014;84:636–645.

18. Seil KS, Desai MM, Smith MV. Sexual orientation, adult connected-

ness, substance use, and mental health outcomes among adolescents:

findings from the 2009 New York City Youth Risk Behavior Survey.

Am J Public Health. 2014;104:1950–1956.

19. Kaestle C, Ivory A. A forgotten sexuality: content analysis of bisex-

uality in the medical literature over two decades. J Bisexuality.

2012;12:35–48.

20. Shearer A, Herres J, Kodish T, et al. Differences in mental health

symptoms across lesbian, gay, bisexual, and questioning youth in pri-

mary care settings. J Adolesc Health. 2016;59:38–43.

21. Minnesota Student Survey Interagency Team. Minnesota Student Sur-

vey 2013. Available at: http://www.health.state.mn.us/divs/chs/mss/

statewidetables/statetablesbygrade13.pdf. Accessed July 1, 2016.

22. Victor SE, Klonsky ED. Correlates of suicide attempts among self-

injurers: a meta-analysis. Clin Psychol Rev. 2014;34:282–297.

23. Stone DM, Luo F, Ouyang L, et al. Sexual orientation and suicide

ideation, plans, attempts, and medically serious attempts: evidence

from local youth risk behavior surveys, 2001-2009. Am J Public

Health. 2014;104:262–271.

24. Brewster M, Moradi B. Perceived experiences of anti-bisexual preju-

dice: instrument development and evaluation. J Couns Psychol. 2010;

57:451–468.

25. Hoffman ND, Freeman K, Swann S. Healthcare preferences of

lesbian, gay, bisexual, transgender, and questioning youth. J Adolesc

Health. 2009;45:222–229.

26. Levine D, Committee on Adolescence. Office-based care for lesbian,

gay, bisexual, transgender, and questioning youth. Pediatrics. 2013;

132:e297–e313.

27. Woodward E, Willoughby B. Family therapy with sexual minority

youths: a systematic review. J GLBT Fam Stud. 2013;10:380–403.

28. American Psychological Association. Guidelines for psychological

practice with lesbian, gay, and bisexual clients. Am Psychol. 2012;

67:10–42.

29. Goodenow C, Szalacha L, Westheimer K. School support groups,

other school factors, and the safety of sexual minority adolescents.

Psychol Sch. 2006;43:537–589.

30. Coker TR, Austin B, Schuster MA. The health and health care of

lesbian, gay, and bisexual adolescents. Ann Rev Public Health.

2010;31:457–477.

31. Kitts RL. Barriers to optimal care between physicians and lesbian,

gay, bisexual, transgender, and questioning adolescent patients. J Ho-

mosex. 2010;57:730–747.

32. Fallin-Bennett K. Implicit bias against sexual minorities in medicine:

cycles of professional influence and the role of the hidden curriculum.

Acad Med. 2015;90:549–552.

33. Meckler GD, Elliott MN, Kanouse DE, et al. Nondisclosure of sexual

orientation to a physician among a sample of gay, lesbian, and

bisexual youth. Arch Pediatr Adolesc Med. 2006;160:1248–1254.

34. Lurie B. A curriculum for educating residents to care for LGBTQ

youth. Acad Pediatr. 2016;16:e6.

35. Makadon H, Mayer K, Potter J, et al. Fenway Guide to Lesbian, Gay,

Bisexual, and Transgender Health. 2nd ed. American College of Phy-

sicians; 2015.

722 TALIAFERRO AND MUEHLENKAMP ACADEMIC PEDIATRICS

©2017 Elsevier