Self-inflicted injury in community adolescents: a pilot study of interpersonal needs and disclosure

Self-injurious thoughts and behaviors (SITBs) are pervasive among adolescents though there remains limited information regarding these behaviors in community samples. We know relatively little about contextual risk factors in community adolescents, specifically how well-supported adult theories, such as the interpersonal theory of suicide, translate to adolescent SITBs. Likewise, no studies have examined the disclosure of SITBs in detail, beyond a basic assessment of who was told. There has been no exploration of how nondisclosers may differ from disclosers. This study sought to examine the application to adolescents of two primary elements of the interpersonal theory of suicide (perceived burdensomeness and thwarted belongingness) in predicting both overall risk for suicide as well as the frequency of self-injury when interacting with emotion dysregulation and disclosure status. Detailed disclosure data, including reasons for each disclosure, amount of information disclosed, and expectations for as well as immediate and long-term helpfulness, was examined based descriptively. One hundred fourteen adolescents aged 13-18 filled out a variety of questionnaires online. Results indicated that 33.3% (n = 38) of the sample engaged in some form of self-inflicted injury and of those participants, 60.5% (n = 23) had disclosed their self-injury to another person while 26.3% (n = 10) had not. Linear regression analyses indicated that higher emotion dysregulation, perceived burdensomeness, and thwarted belongingness all independently predicted higher frequency of self-inflicted injury. No significant interaction between variables was found. Thwarted belongingness interacted with disclosure status to predict suicide risk, with high levels of thwarted belongingness predicting higher suicide risk for adolescents who had not disclosed their self-injury, but not for those who had disclosed. Descriptive analyses of disclosure data indicated that adolescents often and provided significant information about their self-injury to peers, often disclosed for help-seeking purposes, and felt that responses to disclosures were moderately to very helpful. Based on these results, assessing for levels of perceived burdensomeness and thwarted belongingness in suicide risk assessments may be indicated. Future studies should attempt to extend these findings with larger samples and work to continue developing valid and reliable measures of disclosure.

SELF-INFLICTED INJURY IN COMMUNITY ADOLESCENTS: A PILOT STUDY OF INTERPERSONAL NEEDS AND DISCLOSURE by Chloe Rebecca Skidmore A thesis submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Master of Science Department of Psychology The University of Utah August 2013 Copyright © Chloe Rebecca Skidmore 2013 All Rights Reserved The University of Utah Graduate School STATEMENT OF THESIS APPROVAL The following faculty members served as the supervisory committee chair and members for the thesis of_______Chloe Rebecca Skidmore_____________________. Dates at right indicate the members' approval of the thesis. _____Sheila Crowell_______________________, Chair __6/10/13_________ Date Approved _____David M. Huebner___________________, Member __6/10/13_________ Date Approved ______Lisa Diamond______________________, Member __6/11/13_________ Date Approved The thesis has also been approved by____Carol Sansone_________ Chair of the Department/School/College of _Psychology________ and by Donna M. White, Interim Dean of The Graduate School. ABSTRACT Self-injurious thoughts and behaviors (SITBs) are pervasive among adolescents though there remains limited information regarding these behaviors in community samples. We know relatively little about contextual risk factors in community adolescents, specifically how well-supported adult theories, such as the interpersonal theory of suicide, translate to adolescent SITBs. Likewise, no studies have examined the disclosure of SITBs in detail, beyond a basic assessment of who was told. There has been no exploration of how nondisclosers may differ from disclosers. This study sought to examine the application to adolescents of two primary elements of the interpersonal theory of suicide (perceived burdensomeness and thwarted belongingness) in predicting both overall risk for suicide as well as the frequency of self-injury when interacting with emotion dysregulation and disclosure status. Detailed disclosure data, including reasons for each disclosure, amount of information disclosed, and expectations for as well as immediate and long-term helpfulness, was examined based descriptively. One hundred fourteen adolescents aged 13-18 filled out a variety of questionnaires online. Results indicated that 33.3% (n = 38) of the sample engaged in some form of self-inflicted injury and of those participants, 60.5% (n = 23) had disclosed their self-injury to another person while 26.3% (n = 10) had not. Linear regression analyses indicated that higher emotion dysregulation, perceived burdensomeness, and thwarted belongingness all independently predicted higher frequency of self-inflicted injury. No significant interaction between variables was found. Thwarted belongingness interacted with disclosure status to predict suicide risk, with high levels of thwarted belongingness predicting higher suicide risk for adolescents who had not disclosed their self-injury, but not for those who had disclosed. Descriptive analyses of disclosure data indicated that adolescents often and provided significant information about their self-injury to peers, often disclosed for help-seeking purposes, and felt that responses to disclosures were moderately to very helpful. Based on these results, assessing for levels of perceived burdensomeness and thwarted belongingness in suicide risk assessments may be indicated. Future studies should attempt to extend these findings with larger samples and work to continue developing valid and reliable measures of disclosure. iv TABLE OF CONTENTS ABSTRACT …………………………………………..…………………………………iii LIST OF TABLES ………………………………………………………………………vi LIST OF FIGURES ……………………………………………………………………..vii INTRODUCTION ………………………………………………………………………..1 Defining SITB …………………………………………………………………… 2 Interpersonal Theory of Suicide …………………………………………………. 3 Potential Correlates of Self-Injury Frequency ……………………………..……. 4 Disclosure of SITB ……………………………………………………………….5 Purpose of the Current Study …………………………………………………….6 METHOD …………………………………………………………………..…………….9 Participants ………………………………………………………….………….... 9 Procedures …………………………………………………………….……....…10 RESULTS ……………………………………………………………………………….15 Preliminary Analyses …………………………………………………………... 15 Principle Analyses …………………………………………………………….. 22 DISCUSSION …………………………………………………………………………...32 Perceived Burdensomeness and Thwarted Belongingness………….………….. 34 Disclosure of Self-Injurious Behaviors …………………………………………35 Future Directions ……………………………………………………………..…36 REFERENCES ………………………………………………………………………….38 iii vi vii 1 2 3 4 5 6 9 9 10 15 15 34 36 32 22 LIST OF TABLES Table Page 1. Overall Rates of SITB ……………….…………..……............................................ 2. Methods of Self-Injury Used ……………………………………………………..... 3. Reasons for Engaging in Self-Inflicted Injury …………………………………...… 4. Primary Outcome Measures - MANOVA Results ……….….……………………... 26 27 28 29 LIST OF FIGURES Figure Page 1. Classification of self-injurious thoughts and behaviors……………………………….. 2. Main effect of perceived burdensomeness on risk for suicide ……………………… 3. Interaction of disclosure status and thwarted belongingness predicting suicide risk.... 8 30 31 INTRODUCTION Self-injurious thoughts and behaviors (SITBs) are serious and pervasive public health problems, particularly among adolescents and young adults. As of 2009, suicide was the third leading cause of death among 10-24 year olds in the United States (National Centers for Disease Control [CDC], 2009). Furthermore, there are approximately 100-200 attempts per each adolescent or young adult who completes suicide (CDC, 2010). Thus, SITBs are a significant source of morbidity and early mortality. Recent studies indicate that between 7.5-56% of community adolescents have engaged in at least one episode of self-inflicted injury (Barrocas, Hankin, Young, & Abela, 2012; Heath, Baxter, Toste, & McLouth, 2010; Hilt, Cha, & Nolen-Hoeksema, 2008). However, very few studies have thoroughly assessed histories of suicide attempts and nonsuicidal self-injury (NSSI) among community adolescents. Effective intervention and prevention will require earlier identification of these youth, a better understanding of contextual factors associated with SITB, and differences between adolescents who do and do not disclose their SITB to others. Historically, NSSI was considered an odd or uncommon behavior, sometimes thought to be pathognomonic to borderline personality disorder (American Psychological Association [APA], 2012). At the very least, it was thought to be nearly exclusive to clinical populations (Lester, 1972; Offer & Barglow, 1960). However, recent literature has indicated that NSSI is seen in both clinical and community populations and appears 2 to be particularly prevalent among adolescents (Nock, 2010). Furthermore, it has become increasingly apparent that SITB is not a behavior of one certain "kind" of adolescent. Teens who are secretive, neglected, absent from school, poorly connected to community resources, higher functioning, or who have not yet come to clinical attention may be especially vulnerable to being missed. Furthermore, many people never disclose SITB and, consequently, fail to receive adequate treatment or social support. For many adolescents, self-injury is a secret, which may be associated with additional psychological distress (Friedlander, Nazem, Fiske, Nadorff, & Smith, 2012). However, there are no studies that explore the reasons for non-disclosure. We also do not know whether key theories of suicide risk in adults translate to disclosing and nondisclosing community adolescents. Specifically, the interpersonal theory of suicide (Joiner, 2005; Van Orden et al., 2008; Van Orden et al., 2010) is a widely accepted theory of suicide risk for adults and, to a lesser degree, adolescents (Timmons, Selby, Lewinsohn, & Joiner, 2011; Van Orden et al., 2007). However, no studies have examined whether the interpersonal theory applies differently to teenagers who have or have not disclosed their self-injuring behaviors. As information regarding SITB in adolescence increases and as intervention and prevention efforts continue to develop, it is becoming even more important to understand the application of leading theories to teenage samples. Defining SITB Many terms have been used to characterize self-injurious thoughts and behaviors (Prinstein, 2008). However, the most widely accepted nomenclature was outlined recently by Nock (2010). At the broadest level, SITB captures the array of self-injurious 3 thoughts/behaviors (see Figure 1). SITB can be divided into (1) suicidal behaviors (SB), which includes ideation, planning, and attempts, and (2) NSSI, which includes nonsuicidal threats or gestures, thoughts of self-injury, and mild, moderate, or severe self-injury (Crowell, Derbidge, & Beauchaine, in press; Nock, 2010). The key difference between suicidal and nonsuicidal behaviors is intent. Suicidal thoughts and behaviors are quantified as acts with any nonzero level of intent to die. In contrast, nonsuicidal thoughts and behaviors involve no suicidal intent. This is a conservative definition of suicidal behavior because it errs on the side of over-identifying those at elevated risk for suicide. Furthermore, evidence suggests that suicidal and nonsuicidal behaviors often serve distinct emotional functions (Muehlenkamp & Gutierrez, 2004; Muehlenkamp & Kerr, 2010). Interpersonal Theory of Suicide Joiner and colleagues have proposed a three-pronged theory of suicide positing that individuals are most likely to make a high lethality suicide attempt when they have high levels of 1) perceived burdensomeness (i.e., the sense of being a burden on one's family or friends), 2) thwarted belongingness (i.e., feeling a lack of connectedness to important others), and 3) acquired capability for suicide. Acquired capability is hypothesized to emerge from a history of painful and provocative experiences (PPEs). These PPEs produce a diminished sense of anxiety associated with the pain and fear of hurting oneself, effectively increasing one's capacity for lethal self-injury (i.e., suicide; Joiner, 2005; Van Orden, Witte, Cukrowicz, Braithwaite, Selby, & Joiner, 2010). The experience of belonging in relationships with others is foundational to healthy development and well-being (Baumeister & Leary, 1995). A sense of 4 connectedness to one's parents, family, nonfamily adults, and school is an established protective factor against health risk behaviors, suicide, and NSSI (Resnick et al., 1997; Resnick, 2000; Timmons et al., 2011). In contrast, the perception of being unwanted, expendable, or otherwise burdensome on loved ones is associated with suicidality in adults (Joiner et al., 2002; Van Orden, Lynam, Hollar, & Joiner, 2006) and adolescents (Woznica & Shapiro, 1990). To our knowledge, no studies have examined the role of perceived burdensomeness and thwarted belongingness in the existence of self-injury outside of its context as a predictor for later suicide attempts. It is also unknown whether individual differences in emotion regulation or status as a discloser or nondiscloser interact with a sense of burdensomeness or disconnectedness to predict higher frequency of self-inflicted injury or higher overall risk for suicide. Because there is considerable variability among self-injurers, many high-risk adolescents are probably overlooked. Potential Correlates of Self-Injury Frequency Suicidal and nonsuicidal self-injury emerge due to a combination of individual vulnerability factors and contextual risks. Some common characteristics of adolescents who engage in SITBs include a history of depression and/or anxiety (Jacobson & Gould, 2007), difficulties with emotion regulation (Adrian, Zeman, Erdley, Lisa, & Sim, 2011; Heath, Toste, Nedecheva, & Charlebois, 2008), and thought suppression (Najmi, Wegner, & Nock, 2007). Although most of the research on emotion dysregulation has focused on extreme emotional lability, the clinical literature indicates that self-injurers also suppress their displays of affect (Linehan, 1993). To date, there are few studies examining detailed correlates of frequency of self-injury in community adolescents. The relation between risk factors and frequency of self-injury is likely to vary 5 across individuals, due to unique constellations of risk and protective factors. We are particularly interested in how Joiner's interpersonal theory may apply to NSSI and disclosure. While the interpersonal theory is well-established in predicting suicidal behavior, its relevance in predicting the frequency of self-injurious behavior overall has, to our knowledge, only been discussed in the context of how NSSI increases acquired capability, and thus risk for suicide (Joiner, Ribeiro, & Silva, 2012). We are interested in how the interpersonal constructs of thwarted belongingness and perceived burdensomeness may influence the frequency of self-injury overall, regardless of level of suicidal desire. Additionally, no study has examined how perceived burdensomeness and thwarted belongingness relate to or may differ between disclosers and nondisclosers. Disclosure of SITB While research on SITB in adolescents has increased significantly, there is still a paucity of information on help-seeking and disclosure. This is particularly true for community adolescents. A significant proportion of adolescents do not tell anyone about their self-injury. In a school-based assessment, researchers found that about 20% of self-injuring adolescents had not reported their behavior and about 40% of ideators had not disclosed their thoughts or attempted to seek help (Evans, Hawton, & Rodham, 2005). Other findings indicate that when an adolescent does choose to report self-injury, the majority will choose to tell a peer first, followed by a mother, a sibling, and another relative. Teachers, psychologists, and other professionals fall significantly lower on the list (DeLeo & Heller, 2004; Evans et al., 2005; Heath et al., 2010). A better understanding of those who do and do not choose to disclose could improve educational and/or parenting strategies for identifying high-risk youth and 6 directing them to appropriate professional care. It is important to know whether there are typical patterns of disclosure - who teens are disclosing to and why, what they are hoping to gain from the disclosure, and if certain disclosures tend to lead to professional help more quickly. It is also critical to learn which teens do not disclose, and why. No study has examined whether disclosure status interacts with contextual factors to predict heightened risk for suicide. However, high levels of perceived burdensomeness and thwarted belongingness may affect one's willingness to disclose SITB. If we know who is more or less likely to ask for help, we can work to develop more tailored and effective public health interventions. Purpose of the Current Study This study is designed to learn more about self-injurious behaviors in a community sample of adolescents. We are particularly interested in Joiner's interpersonal theory of suicide and how its elements, particularly perceived burdensomeness and thwarted belongingness, relate to self-injury in community adolescents. In addition, to our knowledge, no study has explored details of disclosure beyond a basic assessment of the percentage of adolescents who choose to share their self-injury and with whom such behaviors are discussed. Particularly, no studies have examined possible links between perceived burdensomeness, thwarted belongingness, and status as a discloser or nondiscloser, and how these factors may interact to predict risk for suicide. Without more specific knowledge of disclosure and the associated contextual factors, it is difficult to know where to target public health efforts. We hypothesize that (1) the number of disclosures will be associated with the frequency, lethality, and suicidal intent of self-injury. We also hypothesize that (2) the 7 adolescents who score highest on measures of perceived burdensomeness, thwarted belongingness, and emotion dysregulation will report a higher frequency of self-injury, and that each of these interpersonal factors will interact with emotion dysregulation to predict the highest frequency of self-injury in our sample. Finally, we hypothesize that (3) adolescents who score highest on measures of perceived burdensomeness and thwarted belongingness will be at increased risk for suicide if they have not disclosed their self-injury, but their risk will not be affected if they have disclosed. 8 Figure 1 Classification of Self-Injurious Thoughts and Behaviors (Adapted from Nock, 2010) Self-Injurious Thoughts and Behaviors Suicidal (intent to die) Suicide Ideation Suicide Attempt Suicide Plan Nonsuicidal (no intent to die) Self-Injury Threat/Gesture Self-Injury Thoughts Self-Injury Mild Moderate Severe METHOD Participants Participants included 131 adolescents. Due to incomplete survey data, 17 participants were excluded, leaving a final sample of 114. The average age of participants was 15.3 years old (range = 13-18, SD = 1.176). The sample was 81.6% Caucasian (n = 93), 6.1% Asian (n = 7), 6.1% Hispanic (n = 7), and 6.2% Other (Pacific Islander, African American, or multiethnic; n = 7). The majority of the sample was female (63.2%; n = 72). Most of the sample identified as straight/heterosexual (82.5%, n = 94), 5.3% identified as gay or lesbian (n = 6), 4.4% as bisexual (n = 5), and 7% as unsure or questioning (n = 8). Participants were recruited via direct mailers which included an invitation to participate and a consent document. Sealed letters were sent to approximately 10,000 homes across the greater Salt Lake City area. The letters explained the purpose and procedure of the study and were addressed to parents/guardians. Using an IRB-approved passive consent process, parents/guardians who allowed their adolescent(s) to participate were asked to pass along the survey web-link to their teen. As compensation, adolescents were entered in a drawing for each page of the survey they completed, for a total of seven possible entries. 10 Procedures Adolescents who were interested in participating were directed to a fully online informed assent document, where they provided contact information (if given parental permission for recontact) and were assigned an individual identification code. After completing the assent process, participants were linked to a separate secure, de-identified survey. Participants completed a demographic and personal history form, which included information about age, ethnicity, grade level, grade point average, sexual orientation, relationship status, and religious affiliation. This 20-question survey also included questions about substance use history that were written for the current study. A 22-question modified version of the Childhood Maltreatment Interview Schedule-Short Form (CMIS-SF; Briere, 1992) assessed family functioning/conflict, perceived love and support from primary caregivers, and presence of a history of traumatic experiences. This measure has predictive and construct validity and the Psychological Abuse subscale has shown good reliability. Because of the nature of this measure (i.e., items are not summed to form scores), overall reliability estimates are not available. Participants completed the Youth Self-Report (YSR; Achenbach & Rescorla, 2001) in order to assess for the presence of psychopathology. This is a 112-question survey which measures the presence of social, thought, and attention problems, anxiety, depression, the presence of rule-breaking and aggressive behaviors, as well as internalizing and externalizing behaviors. This measure includes competence, problem, and DSM-oriented scales. Test-retest reliability estimates range from .79 to .88, 11 depending on the scale. Content, criterion-related, and construct validity are all well-established for this measure (Achenbach & Rescorla, 2001). Levels of perceived burdensomeness (the extent to which one perceives him or herself as a burden to others) and thwarted belongingness (consistently unmet needs for positive social interactions) were assessed using the Interpersonal Needs Questionnaire (INQ-12; Van Orden, Witte, Gordon, Bender, & Joiner, 2008). These two constructs are thought to be related to capability for suicide. The INQ-12 is an abbreviated version of the original 25 question measure. Internal consistency for the 12-item measure ranges from α = .85 for items related to thwarted belongingness to α = .89 for items related to perceived burdensomeness (Van Orden et al., 2008). To examine emotion regulation, participants completed the Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004), which measures awareness, understanding, and acceptance of emotions, as well as the presence of goal-directed behavior and effective emotion regulation strategies. The 36-item DERS has been used extensively and has high internal consistency (α = .93) and adequate construct and predictive validity (Gratz & Roemer, 2004). Suicidal and nonsuicidal self-injurious behaviors were assessed using a survey that was modified from the Lifetime - Suicide Attempt Self-Injury Count (L-SASI), which is a clinical interview (Linehan & Comtois, 1996). Questions were restructured to allow for multiple choice answering, as well as follow-up questions (i.e., questions that were originally embedded in one were separated in our protocol). Additionally, questions were added regarding suicidal ideation, past and current suicide plans, number of instances of suicide attempts, nonsuicidal self-injurious behaviors, and the number of 12 suicide plans made. Clear distinctions were made between questions referring to suicidal versus nonsuicidal injuries. Questions related to methods of self-injury were presented in matrix form in our survey, with answer choices that allowed participants to select whether or not they engaged in a specific behavior (e.g., cutting) and if, for them, this behavior was done with suicidal intent, no suicidal intent, both, or neither. We assessed the frequency and extent of suicidal intent for the most frequent method of injury as well as the most serious injury. In addition, a list of possible reasons why the participant might engage in self-injurious behavior was included and this was based on a similar list used in the Self-Harm Information Form (SHIF; Croyle & Waltz, 2007). Questions related to current engagement in self-injury and, if none exists, explanation of the end of the behavior was also included, as was a single question regarding general treatment willingness. In addition, we designed a set of questions for the current study regarding the disclosure process. This measure assessed whether or not participants have disclosed their behaviors and, if so, the experiences surrounding these disclosures. Participants are asked who was told, why and when the disclosure happened, the expectation for help following the disclosure (ranked on a scale of 1-4, with 1 being directly asking for help and 4 being uncertainty of whether they wanted or would receive help), the amount of information disclosed (ranked on a scale of 1-5, with 1 being "most/all" details of self-injury and 5 being acknowledgement of the self-injury without providing any details), the perceived immediate and long-term helpfulness of the individual the participant disclosed to (ranked on a scale of 1-5, with 1 being "not at all helpful" and 5 being "extremely helpful"), and the perceived strength of the relationship with the individual (ranked on a 13 scale of 1-5 with 1 being "very close" and 5 being "not close at all"). These data informed the descriptive analyses for our study. Participants were given the option of completing an additional survey and were compensated by having their name added to the drawing one additional time. If participants opted to continue, they completed the Burden of State Emotion Regulation Questionnaire (B-SERQ; Franchow, Suchy, & Thorgusen, unpublished measure). The B-SERQ is a 26-item survey that measures the extent to which participants have recently (or are presently) regulating their affective responses. This survey explores participants' levels of both affect regulation (e.g., smiling, facial expression, tearfulness) and thought regulation (e.g., distraction, suppression, reappraisal) over the past 2 weeks, and separately, over the past 24 hours. This measure is currently being evaluated for validity and reliability (Franchow, Suchy, & Thorgusen, in progress). Upon completion of the set of questionnaires, adolescents were directed to an online debriefing document and given a list of crisis and mental health referrals. For analytic purposes, data on disclosure, severity of self-injury, and overall risk were collapsed. Disclosure data were collapsed into three groups (those with no history of SII, and those with a history of SII who had disclosed and those who had not), as was a variable of severity by number of methods used (1-3 SII methods was coded as mild, 5-7 methods was moderate, and 9-12 methods was severe), a variable of severity by number of self-injuries (1-10 injuries of NSSI only was coded as mild, 11-50 injuries of NSSI only was coded as moderate, and 51-100 injuries of NSSI only, 1 or more suicide attempts, or any participants with both a suicide attempt and any history of NSSI was coded as severe). In addition, we computed an overall risk score for all participants with a 14 history of SII. This variable was created using a sum score based on the number of methods of SII used, the number of times of injury, level of intent based on two questions (their level of suicidal intent usually during episodes of self-injury and their level of intent during their most serious episode of self-injury), and the lethality of means used. The latter score was summed based on the following breakdown of lethality of means: 1 = scratching, digging into wounds, and other/unidentified method; 2= hitting/banging, stabbing/puncturing; 3 = burning, cutting, smothering; 4 = overdosing on medications, drugs, and/or alcohol; 5 = strangling, stepping into traffic; 6 = deliberately crashing a vehicle; 7 = drowning, hanging; 8 = using a gun. Each of these scores was summed to create the overall composite risk score. RESULTS Preliminary Analyses Descriptive statistics were run and all variables with unacceptable skew or kurtosis were transformed. Rates of suicidal ideation, suicide plans, suicide attempts, and nonsuicidal self-injury are summarized in Table 1. Over half (51.8%; n = 59) of participants endorsed suicidal ideation; 24.6% (n = 28) acknowledged lifetime history of a suicide plan; 6.1% (n = 7) endorsed a current suicide plan; 33.2% (n = 38) reported a lifetime history of SII, including NSSI only (21%; n = 24), suicide attempts only (2.6%; n = 3), and both suicidal and nonsuicidal self-injury (9.6%; n = 11). These 38 participants comprised the SII group for analyses. Participants in the SII group were an average of 15.7 years old, 71.1% female (n = 27), and 81.6% Caucasian (n = 31). Forty-seven point four percent identified as nonreligious (n = 18). Fifty-seven point nine percent (n = 22) reported that they were not currently in a relationship and 78.9% (n = 30) identified their sexual orientation as heterosexual. Participants in the non-SII group were an average of 15.2 years old, 57.7% female (n = 41), and 81.7% Caucasian (n = 58). In addition, 38% (n = 17) identified as nonreligious, 81.7% (n = 58) reported that they were not currently in a relationship, and 85.9% (n = 61) identified their sexual orientation as heterosexual. Within the SII group, 60.5% (n = 23) had disclosed their SII to another person ("disclosers,") while 26.3% (n = 10) had not ("nondisclosers"). Five self-injuring participants declined to answer disclosure questions. These participants are included in 16 the overall SII group, but not in either of the disclosure/nondisclosure subgroups. Self-Inflicted Injury Method of self-injury is summarized in Table 2. Participants were asked to select as many relevant self-injury methods from a list of 16 choices. Self-injuring adolescents had engaged in an average of 4.84 methods of self-injury (range = 1-13). Cutting was the most commonly used method of injury (63.1%; n = 24), followed by scratching (52.6%; n = 20), and hitting/banging (47.3%; n = 18). All 16 methods were endorsed by at least 1 participant, including the use of a gun (2.6%; n = 1). Participants were also asked to identify the method they use most frequently in their self-injury (see Table 2). Fifteen participants stated that cutting was their most frequently used method of self-injury. Participants in the SII group were also asked about most common reasons for engaging in self-injurious behaviors. These findings are summarized in Table 3. "To stop bad feelings" was the most commonly endorsed reason (78.7%; n = 26), followed by "to calm down" (48.5%, n = 16), and "to make your thoughts slow down" (45.5%; n = 15). Psychopathology, Self-Regulation, and Interpersonal Needs Table 4 summarizes descriptive statistics and results for the primary outcome questionnaires measuring emotion regulation, psychopathology, interpersonal needs, and affect suppression. MANOVAs were conducted to compare the non-SII group with the SII group. DERS total scores differed significantly between self-injurers and non-self-injurers, with SII participants scoring higher on emotion dysregulation F(1, 107) = 24.30 (p = .000). As expected, participants with a history of SII scored higher on DERS subscales of Nonacceptance of Emotions F(1, 107) = 15.19 (p = .000), Difficulties with 17 Goal-Directed Behavior F(1, 107) = 7.89 (p = .006), Poor Impulse Control F(1, 107) = 19.66 (p = .000), Problematic Emotion Regulation Strategies F(1, 107) = 49.22 (p = .000), and Poor Emotional Clarity F(1, 107) = 15.17 (p = .000). Participants with no history of SII scored higher on the DERS subscale of Difficulty with Emotional Awareness F(1, 107) = 3.96 (p = .049). Comparisons on the INQ Perceived Burdensomeness scale were also different between self-injurers and non-self-injurers F(1, 107) = 51.13 (p = .000), with self-injurers reporting greater feelings of burdensomeness. Likewise, participants with a history of self-injury scored highest on the INQ Thwarted Belongingness scale F(1, 107) = 15.56 (p = .000). The two groups also differed on baseline F(1, 80) = 24.0 (p = .000) and state F(1, 80) =17.23 (p = .000) affect suppression scores. Significant differences between groups were observed on both YSR Externalizing F(1, 107) =12.12 (p = .001) and Internalizing scores F(1, 107) = 68.99 (p = .000), with self-injurers scoring highest. All of the symptom subscales of the YSR differed in the expected direction between groups (all ps < .01 with the exception of social problems, p = .021). Descriptive Analyses of Disclosing versus Nondisclosing Self-Injurers Differences between disclosers and nondisclosers in the SII group were explored using descriptive analyses, such as means, counts, frequencies, and percentages. Nondisclosers appeared similar to disclosers on age and ethnicity, but fewer nondisclosers identified as religious (10% of nondisclosers; 43.4% of disclosers). Nondisclosers acknowledged being single more often (70% vs. 47.8%), and more nondisclosers reported being unsure of or questioning their sexual orientation (40% vs. 4.3%). A demographic measure exploring family relationships and conflict indicated that 18 nondisclosers reported feeling less love from parents and other adults compared with disclosers and the non-SII group. Those in the self-injuring group overall perceived less love from their friends compared to the non-SII group, but disclosers and nondisclosers responded similarly to this question. Likewise, nondisclosers reported more than both disclosers and the non-SII group that they were yelled at, insulted, criticized, made to feel guilty, humiliated, embarrassed, and made to feel like a bad person by their parents over the course of a year. Disclosers also appeared to score more highly on each of these measures than did teens in the non-SII group. With regard to methods and frequency of SII, nondisclosers more often endorsed scratching as their most common method of self-injury (80%, n = 8) rather than cutting (see Table 2). Similarly, nondisclosers reported hitting/banging and scratching as their most frequently used method (both 30%, n = 3) as opposed to cutting for disclosers (see Table 2). Both groups endorsed "to stop bad feelings" as the primary reason for SII. However, nondisclosers more often endorsed "to get other people to act differently or change" (30%, n = 3) and "to get away or escape" (60%, n = 6) as reasons for their self-injury, compared to disclosers. Thirty percent of nondisclosers (n = 3) endorsed "to shock or get a reaction out of people" as a reason for SII, while no disclosers endorsed this reason. Self-injuring adolescents who had disclosed their self-injury answered a series of questions regarding each of their disclosures. Twenty-two disclosing participants provided this information and one declined. Disclosing participants had told an average of 4.27 people about their self-injury, with 90% (n = 20) telling two or more people about their self-injury. For further descriptive analyses, individuals who were disclosed to were 19 collapsed into five groups: family (parents, siblings, extended relatives), peers (friends and significant others), professionals in the community (mental health workers, medical professionals, staff at treatment facilities, probation officers), professionals at school (teachers, coaches), and other adults (those at work, church, friends' parents, etc.). Descriptive analyses indicated that disclosers most often told peers (36.8%, n = 35) followed by family (33.7%, n = 32), community professionals (15.8%, n = 15), other adults (8.4%, n = 8), and school professionals (5.2%, n = 5). Four adolescents endorsed having only disclosed to a peer about their self-injury. The most common reasons for disclosure were "needing or wanting help" (n = 33), "wanting this person to know" (n = 33), "just needing to tell someone" (n = 30), and "being forced to tell" someone (n = 17). When they needed or wanted help, adolescents most often told family (n = 12), followed by community professionals (n = 9), peers (n = 6), other adults (n = 4), and school professionals (n = 2). When they felt as though they "just needed to tell someone," adolescents most often told peers (n = 19), followed by family (n = 7), community professionals (n = 2), school professionals (n = 1) and other adults (n = 1). When they "wanted [someone] to know," this was most often peers (n = 17) followed by family (n = 9), community professionals (n = 4), other adults (n = 2), and school professionals (n = 1). Teens were most often forced to tell family (n = 9), followed by community professionals (n = 4), school professionals (n = 3), and other adults (n = 1). None of our participants endorsed being forced to tell a peer about self-injury. Immediately upon disclosure, the majority of teens reported that the individuals they told reacted by letting the teen know he or she was cared about (n = 61). This reaction was most common of peers (n = 27) followed by family (n = 21), other adults (n 20 = 7), and least often from school and community professionals (n = 3 for each group). The second most frequent immediate reaction was to help the teen find an option for treatment (n = 30), and this was most common of family (n = 12), community professionals (n = 11), school professionals (n = 3), and least common from peers and other adults (n = 2 for each group). "Not doing anything at all" was the third most common immediate reaction to disclosure (n = 16), most often attributed to family members (n = 7), peers (n = 6), and community professionals (n = 3). Fourteen teens endorsed that they had been taken care of physically immediately following disclosure, and this reaction was most common of family members (n = 7), peers (n = 3), other adults (n = 2), and school and community professionals (n = 1 in each group). Eleven teens endorsed that the individual they disclosed to immediately "got angry, upset, or made [them] feel bad," and this was most common of family members (n = 7), peers (n = 3), and community professionals (n = 1). Participants were asked to answer these same questions regarding reactions days, weeks, or months after the disclosure. Results indicated that, after the disclosure, those disclosed to were still most likely to let the teens know they were cared about (n = 56), followed by not doing anything at all (n = 25), helping the teen to get some sort of treatment (n = 24), taking care of the teen physically (n =12), and becoming angry, upset, or making the teen feel badly (n = 9). Reactions following later disclosure by each category of person were distributed similarly to the reactions following immediate disclosure, described above. Overall, adolescents disclosed the most detail about their self-injury to peers (M = 2.22), with less detail told, in rank order, to community professionals (M = 2.43), family 21 members (M = 3.1), school professionals (M = 3.2), and other adults (M = 3.63). They expected the most help from community professionals (M = 1.61), with gradually less help expected from other adults (M = 2), school professionals (M = 2.5), family members (M = 2.7), and the least amount of help expected, but still hoped for, from peers (M = 2.81). Teens found other adults to be the most immediately helpful (M = 4.03), followed by peers (M = 3.64), community professionals (M = 3.43), and school professionals (M = 3.25), while family members were deemed least immediately helpful (M = 3.12). Over the long-term, other adults were rated as the most helpful (M = 3.8), followed by community professionals (M = 3.62), school professionals and peers (M = 3.5 for each group), and family was again deemed the least helpful (M = 3.22). Adolescents reported that they had the strongest relationships with peers (M = 1.83), family members (M = 1.84), other adults (M = 1.96), and the weakest relationships with community (M = 2.92) and school professionals (M = 3.2). With regard to help-seeking, 34.7% (n = 33) of adolescents endorsed not expecting help from the individual they disclosed to, while 23.2% (n = 22) reported that they directly asked the individual for help, 20% (n = 19) were unsure of whether or not they thought they would receive help following the disclosure, and 17.9% (n = 17) reported that they hoped the individual would help them but they did not directly ask for help. 4.2% of participants (n = 4) declined to answer this question. In sum, descriptive analyses indicated that adolescents who disclosed their self-injury typically divulged "some" details of their self-injury, hoped for help following their disclosure but did not directly ask for it, found both immediate and longer-term disclosures to be moderately to very helpful, and disclosed most often to individuals to 22 whom they felt "close" or "somewhat close." Principle Analyses Hypothesis 1: Number of disclosures will be associated with the frequency, lethality, and suicidal intent of the self-injury. To examine whether the frequency, lethality, and suicidal intent of participants' self-injury was associated with the number of disclosures, we conducted bivariate correlation analyses between a count of disclosures, severity scores (by number of times and methods) and two ratings of suicidal intent (assessing intent during typical episodes of self-injury and during the most serious episode of self-injury). These analyses were computed within the SII group only. Number of disclosures was significantly correlated with the number of times a participant had self-injured, with higher frequency of self-injury associated with more disclosures (r = .433, p = .012). Contrary to expectations, number of disclosures was not associated with number of methods used (r = -.049, p = .793), level of suicidal intent during typical episodes of self-injury (r = -.015, p = .933), or level of suicidal intent during the most serious episode of self-injury (r = .259, p = .145). Hypothesis 2: Emotion dysregulation will interact with (1) perceived burdensomeness and (2) thwarted belongingness to predict the frequency of self-injury. In addition, disclosure will interact with (1) perceived burdensomeness and (2) thwarted belongingness to predict overall risk for suicide, with nondisclosers at higher risk than disclosers. 23 Predicting Frequency of Self-Injury To examine whether emotion dysregulation, perceived burdensomeness, and thwarted belongingness accounted for variance in the frequency of self-injury, we conducted a series of linear regressions with frequency of self-injury as the outcome variable. Interaction tests were conducted by entering both predictor variables (i.e., emotion dysregulation and perceived burdensomeness) and the predictor × predictor product vector (i.e., emotion dysregulation × perceived burdensomeness). We took the same approach to testing the interaction between emotion dysregulation and thwarted belongingness. An interaction model is supported when the product vector is significant, regardless of the significance of the main effects. All variables were mean-centered. Due to unacceptable levels of skewness and kurtosis, perceived burdensomeness was transformed using a square root transformation. Does emotion dysregulation independently predict the number of self-inflicted injuries? Regressed independently, emotion dysregulation significantly predicted frequency of self-injury (F(1, 112)=24.8, p = .000). Does perceived burdensomeness interact with emotion dysregulation to predict the number of self-inflicted injuries? As is common with measures of self-injury frequency, the variable was transformed using a natural log transformation to reduce skew and kurtosis. Contrary to our hypothesis, the interaction between emotion dysregulation and perceived burdensomeness did not predict the frequency of self-injury F(3, 109) = 18.06 (p = .205). Likewise, main effects of emotion dysregulation and perceived burdensomeness did not achieve significance (p = .937; p = .610). Regressed independently, perceived burdensomeness did predict frequency of self-injury F(1, 112) 24 = 44.39 (p = .000). Does thwarted belongingness interact with emotion dysregulation to predict the number of self-inflicted injuries? The interaction between emotion dysregulation and thwarted belongingness also did not predict the frequency of self-injury F (3, 109) = 13.01 (p = .073). In this regression equation, the main effect of thwarted belongingness was not significant (p = .249). However, regressed independently, thwarted belongingness did significantly predict frequency of self-injury (F (1, 112) = 12.95, p = .000). Predicting Risk for Suicide To examine whether status as a discloser or nondiscloser moderated the relation between perceived burdensomeness/thwarted belongingness and suicide risk, we conducted a series of linear regressions with the composite risk for suicide score as the outcome variable. Disclosure status was effect coded, with disclosers coded as .5 and nondisclosers coded as -.5. Interaction tests were conducted by entering both predictor variables (i.e., disclosure status and perceived burdensomeness) and the predictor × predictor product vector (i.e., disclosure status × perceived burdensomeness). We took the same approach to testing the interaction between disclosure status and thwarted belongingness. All continuous variables were mean-centered. Does perceived burdensomeness interact with disclosure status to predict risk for suicide? The interaction between disclosure status and perceived burdensomeness did not significantly predict level of suicide risk (F(3, 27) = 1.731, p = .593). However, a main effect for perceived burdensomeness was found to be significant (F(3, 27) = 1.731, p = .032), indicating that teens who endorsed higher levels of perceived burdensomeness 25 tended to be at a higher risk for suicide, regardless of status as a discloser or nondiscloser (see Figure 2). Does thwarted belongingness interact with disclosure status to predict risk for suicide? There was a significant interaction between disclosure status and thwarted belongingness in predicting overall risk for suicide F(3, 27) = 1.656 (p = .041). A positive association was found between overall suicide risk and thwarted belongingness for nondisclosers, but not for disclosers. That is, for adolescents who had disclosed their self-injury, the extent of their experience of thwarted belongingness did not affect their overall risk for suicide. However, for adolescents who had not disclosed, higher scores on thwarted belongingness were associated with higher overall risk for suicide (see Figure 3). 26 Table 1 Overall Rates of SITB n % Suicidal Ideation 59 51.8% Suicide Plan - Lifetime 28 24.6% Suicide Plan - Current 7 6.1% Suicide Attempt 3 2.6% Nonsuicidal Self-Injury 24 21.0% Suicide Attempt and NSSI 11 9.6% No Self-Inflicted Injury 71 62.3% Decline to Answer 5 4.4% 27 Table 2 Methods of Self-Injury Used Method Disclosers n = 23 (%) Nondisclosers n = 10 (%) Cutting 18 (28.2%)± 6 (60%) Scratching 12 (52.2%) 8 (80%)± Hitting or Banging 11 (47.8%)± 7 (70%)± Digging into Wounds 10 (43.4%) 6 (60%) Overdosing on Medication 9 (39.1%) 2 (20%) Burning 7 (30.4%) 2 (20%) Stabbing or Puncturing 6 (26.0%) 4 (40%) Overdosing on Drugs or Alcohol 5 (21.7%) 1 (10%) Strangling 5 (21.7%) 3 (30%) Jumping from Heights 4 (17.3%) 1 (10%) Drowning 3 (13%) 1 (10%) Smothering 3 (13%) 2 (20%) Stepping into Traffic 2 (8.6%) 3 (30%) Hanging 1 (4.3%) 1 (10%) Gun 1 (4.3%) --- Deliberately Crashing Car --- 2 (20%) Other 9 (39.1%) 5 (50%) ± denotes most frequently used methods for each group, while the overall table represents all methods used within the SII sample 28 Table 3 Reasons for Engaging in Self-Inflicted Injury Disclosers n = 23 (%) Nondisclosers n = 10 (%) To stop bad feelings 17 (73.9%) 9 (90%) To communicate or let others know how desperate you are 11 (47.8%) 3 (30%) To calm down 11 (47.8%) 5 (50%) To make your thoughts slow down 10 (43.5%) 5 (50%) To feel something, even if it was pain 10 (43.5%) 5 (50%) Because it felt good 9 (39.1%) 2 (20%) I don't know 8 (34.7%) 4 (40%) To punish yourself 7 (30.4%) 2 (20%) To get away or escape 7 (30.4%) 6 (60%) Because it is a habit 7 (30.4%) 2 (20%) To get help 3 (13.0%) 2 (20%) To prove to yourself that things really were bad and it was okay to feel as bad as you did 6 (26.1%) 3 (30%) To die 6 (26.1%) 2 (20%) To give yourself something to do 5 (21.7%) --- To give you a feeling of accomplishment (that you were doing something well) 4 (17.4%) 2 (20%) To get rid of something about your body you didn't like 4 (17.4%) --- To wake up or get yourself going 3 (13.0%) 1 (10%) To prove you're tough 3 (13.0%) --- To get other people to act differently or change 2 (8.7% 3 (30%) To get back at or hurt someone 2 (8.7%) 2 (20%) To try and improve your appearance 2 (8.7%) --- For another reason not listed here 2 (8.7%) 2 (20%) Decline to answer 2 (8.7%) --- To shock or get a reaction out of people --- 3 (30%) To get out of doing something --- 1 (10%) To look cool --- --- To make others feel needed --- --- 29 Table 4 Primary Outcome Measures - MANOVA Results No History of SII M (SD) Positive for SII M (SD) F Ƞ2 p-value DERS Nonacceptance 11.23 (5.52) 16.11 (7.37) 15.19 .124 .000*** Goals 12.75 (3.86) 14.92 (3.82) 7.896 .069 .006** Impulse 9.08 (2.53) 11.66 (3.45) 19.66 .155 .000*** Awareness 19.69 (5.82) 17.39 (5.58) 3.95 .036 .049* Strategies 13.69 (4.03) 20.37 (5.83) 49.22 .315 .000*** Clarity 11.86 (2.38) 13.81 (2.71) 15.17 .124 .000*** Total 80.25 (17.10) 97.58 (18.18) 24.30 .185 .000*** YSR Withdrawn 10.49 (2.59) 13.42 (2.93) 30.95 .212 .000*** Somatic Complaints 10.77 (2.19) 14.21 (3.89) 33.13 .246 .000*** Anxious/Depressed 23.15 (5.15) 31.82 (5.99) 67.11 .368 .000*** Social Problems 11.09 (2.33) 12.45 (2.75) 5.5 .064 .021** Thought Problems 10.27 (2.60) 13.0 (2.95) 22.99 .188 .000*** Attention Problems 14.55 (3.06) 16.92 (2.55) 18.82 .134 .000*** Delinquent Behavior 13.91 (2.87) 16.34 (3.78) 13.1 .116 .000*** Aggressive Behavior 25.87 (4.62) 28.52 (5.36) 7.05 .064 .009** Externalizing 39.79 (6.49) 44.87 (8.14) 12.11 .106 .001** Internalizing 43.36 (7.55) 57.57 (9.87) 68.99 .397 .000*** INQ Thwarted Belongingness 11.46 (6.95) 17.45 (8.53) 15.59 .127 .000*** Perceived Burdensomeness 12.94 (6.45) 26.18 (12.91) 51.13 .323 .000*** BSERQ Baseline Suppression 26.59 (9.07) 36.67 (8.77) 24.0 .231 .000*** State Suppression 21.67 (7.31) 28.67 (7.42) 17.23 .177 .000*** 30 Figure 2 Main Effect of Perceived Burdensomeness on Risk for Suicide 31 Figure 3 Interaction of Disclosure Status and Thwarted Belongingness Predicting Suicide Risk 32 DISCUSSION In this study, we sought to examine the relation between self-injurious behaviors in community adolescents and levels of emotion dysregulation, perceived burdensomeness, and thwarted belongingness. We were also interested in differences between disclosers and nondisclosers with regard to method and function of the self-injury, as well as the details of disclosure among those who had told someone else about their self-injury. In our sample, 33.3% (n = 38) of participants endorsed a lifetime history of self-injury (nonsuicidal self-injury, a suicide attempt, or both). Of the self-injuring group, 60.5% (n = 23) acknowledged that they had disclosed their self-injury to another person, while 26.3% (n = 10) had not disclosed their self-injury to anyone. We hypothesized that the number of disclosures a teen made would be associated with the frequency, lethality, and suicidal intent of the self-injury and this hypothesis was partially confirmed. While a higher frequency of self-injury was associated with more disclosures, the number of disclosures was not associated with the number of methods a teen used or their level of suicidal intent during typical or most serious episodes of self-injury. This suggests that nondisclosers may have self-injured less often but otherwise have similar suicide-risk behaviors as disclosers. We also hypothesized that the frequency of teens' self-injury would be predicted by an interaction between emotion dysregulation and perceived burdensomeness, as well as an interaction between emotion dysregulation and thwarted belongingness with higher 33 levels predicting the most self-injury. Contrary to expectations, neither perceived burdensomeness nor thwarted belongingness interacted with emotion dysregulation to predict frequency of self-injury. However, when regressed independently, higher levels of thwarted belongingness statistically predicted higher frequency of self-injury. Similarly, emotion dysregulation also significantly predicted frequency of self-injury when regressed independently, indicating that higher levels of emotion dysregulation predicted a higher frequency of self-injury. In addition, we hypothesized that both perceived burdensomeness and thwarted belongingness, individually, would interact with status as a discloser or nondiscloser to predict the highest risk for suicide based on a composite score. While the interaction between perceived burdensomeness and disclosure status was not significant, a main effect of perceived burdensomeness did mark higher suicide risk. Regardless of disclosure status, those with high levels of burdensomeness scored highest on the composite score. This suggests that both disclosers and nondisclosers are equally vulnerable to engaging in suicide risk behaviors if they perceive themselves as a burden on others. Interestingly, the interaction between thwarted belongingness and disclosure status was a significant predictor of suicide risk. For nondisclosers, higher thwarted belongingness was associated with a greater suicide risk score, while this association was not true for disclosers. Future research should examine whether strengthening an adolescent's support network and sense of belongingness could increase the likelihood of disclosure and facilitate the transition to professional care. Our small sample size precluded a more extensive examination of the statistical differences between disclosers and nondisclosers. Furthermore, null findings could be 34 due, in part, to a lack of statistical power, rather than an absence of association between the variables. However, results indicate that nondisclosers engage in equally high rates of suicide risk behaviors and may be uniquely sensitive to the experience of thwarted belongingness. Future studies should continue to examine these variables in the context of adolescent self-injury and disclosure in larger samples. Perceived Burdensomeness and Thwarted Belongingness A primary goal of this study was to examine the applicability to adolescents of two elements of Joiner's interpersonal theory of suicide. We explored levels of perceived burdensomeness and thwarted belongingness, but did not directly assess Joiner's third criterion of acquired capability. Previous literature has established that the combination of perceived burdensomeness and thwarted belongingness predicts higher lethality of suicide attempts, with somewhat equivocal support for acquired capability. This finding has been consistent in adult samples (Van Orden et al., 2007; Van Orden et al., 2008) and similar early findings have been found in adolescents (Stellrecht et al., 2006; Timmons et al., 2011; Van Orden, et al., 2007). This is surprising given that perceived burdensomeness and thwarted belongingness may be uniquely potent during adolescence. Our analyses revealed that both high thwarted belongingness and high perceived burdensomeness independently predict higher frequency of self-injury and risk for suicide. This suggests that clinicians and researchers should consider adding measures of thwarted belongingness and perceived burdensomeness to risk assessment batteries. 35 Disclosure of Self-Injurious Behaviors In addition to our primary aims, we also included a questionnaire assessing disclosure in order to pilot its utility for community adolescents. Although we did not have a large enough sample to detect statistical differences between groups, we were able to explore interesting descriptive findings for disclosers and nondisclosers. Our descriptive data suggest that disclosers and nondisclosers acknowledge somewhat different methods of and reasons for self-injury. Interestingly, nondisclosers in this sample were more likely to acknowledge a need or desire to communicate with others as a reason for self-injury. This suggests that teaching effective communication strategies, improving parental attentiveness, or increasing a sense of belongingness could be useful interventions for nondisclosing adolescents. However, this is speculative and will require further study. The adolescents in our sample most often disclosed their self-injury to peers, and disclosed least often to school professionals. However, they often reported that peers "did not do anything at all" following a disclosure. Disclosing adolescents most often reported seeking out family members when they needed or wanted help, but ranked family lowest on helpfulness. Overall, the adolescents in our sample most often disclosed their self-injury because they needed or wanted help, or wanted a specific person to know about their experience. The teens in our sample were assisted, in some manner, approximately 30 times in getting professional treatment following their disclosure. However, the vast majority of teens reported that the individuals they disclosed to were most likely to let the teen know he/she was cared for. While this is certainly important to communicate, it is likely insufficient on its own in most cases. Likewise, there was a high rate of teens who 36 perceived that those they disclosed to did nothing following the disclosure. That said, very few of the teens in our sample directly asked for help, which may have influenced the outcome of disclosure. A teen's approach to the disclosure is potentially important (e.g., direct communication vs. accidental discovery). Future studies on disclosure should assess this information. Taken together, our descriptive data suggest that teens are often not receiving the help and support they need or expect from family members and school professionals in particular. The latter is consistent with prior research (Heath et al., 2010). This is noteworthy considering the emphasis of the school community (Gould & Kramer, 2001; Lieberman, Toste, & Heath, 2008; Muehlenkamp, Walsh, & McDade, 2010; Toste, & Heath, 2010) and family support (Sharaf, Thompson, & Walsh, 2009; Walsh, 2006) in many NSSI and suicide interventions. Future Directions This is a pilot study examining disclosure of self-injury in an unselected community sample. Future studies would benefit from using similar questions about disclosure and self-injurious behaviors with a larger, school- or community-based sample in an attempt to get a broader, more representative group of adolescents. A strength of this study was that we gathered data on the disclosure processes of self-injuring adolescents. To our knowledge, this type of data has not been gathered in this manner before, with this level of detail. This descriptive information will help to inform future studies and assist in continued efforts to emphasize the importance of information about disclosure and help-seeking in samples of self-injuring community adolescents. In addition, very few studies have attempted to apply Joiner's interpersonal theory of suicide to adolescents. Findings from this study provide a new direction for future studies 37 to continue examining the importance of perceived burdensomeness in predicting frequency of self-injury, disclosure, and overall risk trajectory in adolescence. In addition, future studies should continue to perfect valid and reliable measures that examine details of disclosure and nondisclosure. Establishing measures that are able to meaningfully track temporal history of disclosure (e.g., how long from first disclosure to access to a mental health professional), more accurately assess the content of the disclosure, and that can clearly distinguish each instance of disclosure would assist in clarifying differences that may emerge between disclosers and nondisclosers. Researchers should also examine the disclosure process based upon different patterns of self-injury (e.g., NSSI only, single attempters, multiple attempters). This is consistent with research establishing the existence of distinct subgroups of self-injuring adolescents, who vary across the dimensions of SII frequency, number of methods used, suicidal intent, and overall risk (Hamza & Willoughby, 2013; Klonsky & Olino, 2008; Whitlock, Muehlenkamp, & Eckenrode, 2008). It is possible that disclosure or lack thereof may differ within these subgroups. In sum, improving our understanding of relevant contextual factors while increasing knowledge about disclosure may have important practical implications for future research and treatment as we continue to clarify differences between groups of self-injuring adolescents. 38 REFERENCES Achenbach, T. M., & Rescorla, L. A. (2001). Manual for the ASEBA School-Age Forms & Profiles. University of Vermont, Research Center for Children, Youth, & Families, Burlington, VT. Adrian, M., Zeman, J., Erdley, C., Lisa, L., & Sim, L. (2011). Emotional dysregulation and interpersonal difficulties as risk factors for nonsuicidal self-injury in adolescent girls. Journal of Abnormal Child Psychology, 39, 389-400. American Psychological Association (2012). DSM-5 Rationale for Proposed Revision to Nonsuicidal Self-Injury. Access at: http://www.dsm5.org/ProposedRevisions/Pages/proposedrevision.aspx?rid=443# Barrocas, A. L., Hankin, B. L., Young, J. F., & Abela, J. R. Z. (2012). Rates of nonsuicidal self-injury in youth: Age, sex, and behavioral methods in a community sample. Pediatrics, 130, 39-45. Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497-529. Briere, J. (1992). Child abuse trauma: Theory and treatment of the lasting effects. Newbury Park, CA: Sage Publications. Crowell, S. E., Derbidge, C. M., & Beauchaine, T. P. (in press). Developmental approaches to understanding suicidal and self-injurious behaviors. The Oxford Handbook of Suicide and Self-Injury. Croyle, K. L., & Waltz, J. (2007). Subclinical self-harm: Range of behaviors, extent, and associated characteristics. American Journal of Orthopsychiatry, 77, 332-342. DeLeo, D., & Heller, T. S. (2004). Who are the kids who self-harm? An Australian self-report school survey. Medical Journal of Australia, 181, 140-144. Evans, E., Hawton, K., & Rodham, K. (2005). In what ways are adolescents who engage in self-harm or experience thoughts of self-harm different in terms of help-seeking, communication and coping strategies? Journal of Adolescence, 28, 573-587. 39 Franchow, E. I., Suchy, Y., & Thorgusen, S. R. (unpublished measure). The Burden of State Emotion Regulation Questionnaire. Friedlander, A., Nazem, S., Fiske, A., Nadorff, M. R., & Smith, M. D. (2012). Self-concealment and suicidal behaviors. Suicide and Life-Threatening Behavior, 42, 332-340. Gould, M. S., & Kramer, R. A. (2001). Youth suicide prevention. Suicide and Life-Threatening Behavior, 31, 6-31. Gratz, K. L., & Roemer, L. (2004). Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the difficulties in emotion regulation scale. Journal of Psychopathology and Behavioral Assessment, 26, 41-54. Hamza, C. A., & Willoughby, T. (2013). Nonsuicidal self-injury and suicidal behavior: A latent class analysis among young adults. PLoS ONE, 8, 1-7. Heath, N. L., Baxter, A. L., Toste, J. R., & McLouth, R. (2010). Adolescents' willingness to access school-based support for nonsuicidal self-injury. Canadian Journal of School Psychology, 25, 260-276. Heath, N. L., Toste, J. R., Nedecheva, T., & Charlebois, A. (2008). An examination of nonsuicidal self-injury among college students. Journal of Mental Health Counseling, 30, 137-156. Hilt, L. M., Cha, C. B., & Nolen-Hoeksema, S. (2008). Nonsuicidal self-injury in young adolescent girls: Moderators of the distress-function relationship. Journal of Consulting and Clinical Psychology, 76, 63-71. Jacobson, C. M., & Gould, M. (2007). The epidemiology and phenomenology of nonsuicidal self-injurious behavior among adolescents: A critical review of the literature. Archives of Suicide Research, 11, 129-147. Joiner, T. E., Pettit, J. W., Walker, R. L., Voelz, Z. R., Cruz, J., Rudd, M. D., & Lester, D. (2002). Perceived burdensomeness and suicidality: Two studies on the suicide notes of those attempting and those completing suicide. Journal of Social and Clinical Psychology, 21, 531-545. Joiner, T. E., Conwell, Y., Fitzpatrick, K. K., Witte, T. K., Schmidt, N. B., Berlim, M. T.,… Rudd, M. D. (2005). Four studies on how past and current suicidality relate even when "everything but the kitchen sink" is covaried. Journal of Abnormal Psychology, 114, 291-303. Joiner, T. (2005). Why people die by suicide. Cambridge, MA: Harvard University Press. 40 Joiner, T. E., Ribeiro, J. D., & 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. Klonsky, E. D., & Olino, T. M. (2008). Identifying clinically distinct subgroups of self-injurers among young adults: A latent class analysis. Journal of Consulting and Clinical Psychology, 76, 22-27. Lester, D. (1972). Self-mutilating behavior. Psychological Bulletin, 78, 119-128. Lieberman, R. A., Toste, J. R., & Heath, N. L. (2008). Nonsuicidal self-injury in the schools: Prevention and intervention. In M. K. Nixon & N. L. Heath (Eds.), Self-injury in youth: The essential guide to assessment and intervention (pp. 195-216). New York, NY: Routledge. Linehan, M. M. (1993). Cognitive behavioral treatment of borderline personality disorder. New York: Guilford Press. Linehan, M. M., & Comtois, K. (1996). Lifetime parasuicide history. University of Washington, Seattle, WA. Unpublished work. Muehlenkamp, J. J., & Gutierrez, P. M. (2004). An investigation of differences between self-injurious behavior and suicide attempts in a sample of adolescents. Suicide and Life-Threatening Behavior, 34, 12-23. Muehlenkamp, J. J., & Kerr, P. L. (2010). Untangling a complex web: How nonsuicidal self-injury and suicide attempts differ. The Prevention Research, 17, 8-10. Muehlenkamp, J. J., Walsh, B. W., & McDade, M. (2010). Preventing nonsuicidal self-injury in adolescents: The signs of self-injury program. Journal of Youth and Adolescence, 39, 306-314. Najmi, S., Wegner, D. M., & Nock, M. K. (2007). Thought suppression and self-injurious thoughts and behaviors. Behavior Research and Therapy, 45, 1957-1965. National Center for Injury Control and Prevention (2010). Suicide: Facts at a glance. US Department of Health and Human Services, Centers for Disease Control and Prevention. Available at: http://www.cdc.gov/ViolencePrevention/pdf/Suicide_DataSheet-a.pdf. National Center for Injury Control and Prevention. (2009). Ten leading causes of death by age group, United States - 2009. US Department of Health and Human Services, Centers for Disease Control and Prevention. Available at: http://www.cdc.gov/Injury/wisqars/pdf/10LCD-Age-Grp-US-2009-a.pdf 41 Nock, M. K. (2010). Self-injury. Annual Review of Clinical Psychology, 6, 15.1-15.25. Offer, D., & Barglow, P. (1960). Adolescent and young adult self-mutilation incidents in a general psychiatric hospital. Archives of General Psychiatry, 3, 194-204. Prinstein, M. J. (2008). Introduction to the special section on suicide and nonsuicidal self-injury: A review of unique challenges and important directions for self-injury science. Journal of Consulting and Clinical Psychology, 76, 1-8. Resnick, M. D., Bearman, P. S., Blum, R. W., Bauman, K. E., Harris, K. M., Jones, J.,…Udry, R. (1997). Protecting adolescents from harm: Findings from the national longitudinal study on adolescent health. Journal of the American Medical Association, 278, 823-832. Resnick, M. D. (2000). Protective factors, resiliency, and healthy youth development. Adolescent Medicine, 11, 157-164. Sharaf, A. Y., Thompson, E. A., & Walsh, E. (2009). Protective effects of self-esteem and family support on suicide risk behaviors among at-risk adolescents. Journal of Child and Adolescent Psychiatric Nursing, 22, 160-168. Stellrecht, N. E., Gordon, K. H., Van Orden, K., Witte, T. K., Wingate, L. R., Cukrowicz, K. C., Butler, M., Schmidt, N. B., Fitzpatrick, K. K., & Joiner, T. E. (2005). Clinical applications of the interpersonal-psychological theory of attempted and completed suicide. Journal of Clinical Psychology: In Session, 62, 211-222. Timmons, K. A., Selby, E. A., Lewinsohn, P. M., & Joiner, T. E. (2011). Parental displacement and adolescent suicidality: Exploring the role of failed belonging. Journal of Clinical Child and Adolescent Psychology, 40, 807-817. Toste, J. R., & Heath, N. L. (2010). School response to nonsuicidal self-injury. The Prevention Researcher, 17, 14-17. Van Orden, K. A., Lynam, M. E., Hollar, D., & Joiner, T. E. (2006). Perceived burdensomeness as an indicator of suicidal symptoms. Cognitive Therapy and Research, 30, 457-467. Van Orden, K. A., Witte, T. K., Gordon, K. H., Bender, T. W., & Joiner, T. E. (2008). Suicidal desire and the capability for suicide: Tests of the Interpersonal-Psychological Theory of Suicidal Behavior among adults. Journal of Consulting and Clinical Psychology, 76, 72-83. Van Orden, K. A., Witte, T. K., Selby, E. A., Bender, T. W., & Joiner, T. E. (2007). Suicidal behavior in youth. In J. R. Z. Abela & B. L. Hankin (Eds.), Handbook of depression in children and adolescents (pp. 441-465). New York, NY: Guilford Press. 42 Van Orden, K. A., Witte, T. K., Cukrowicz, K. C., Braithwaite, S. R., Selby, E. A., & Joiner, T. E. (2010). The interpersonal theory of suicide. Psychological Review, 117, 575-600. Walsh, B. (2006). Treating self-injury: A practical guide. New York, NY: Guilford Press. Whitlock, J., Muehlenkamp, J., & Eckenrode, J. (2008). Variation in nonsuicidal self-injury: Identification and features of latent classes in a college population of emerging adults. Journal of Clinical Child & Adolescent Psychology, 37, 725-735. Woznica, J. G., & Shapiro, J. R. (1990). An analysis of adolescent suicide attempts: The expendable child. Journal of Pediatric Psychology, 15, 789-796.