This item has the following downloads:
1 THE RELATIONSHIP BETWEEN NONSUICIDAL SELF INJURY (NSSI) AND PSYCHOSOCIAL FUNCTIONING IN COLLEGE STUDENTS By STACEY M. RICE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Stacey M. Rice
3 To my grandparent s, who I know would be so proud
4 ACKNOWLED GMENTS Completing this dissertation was not an easy task, and I would not have been able to do it without the many wonderful people I am lucky to have in my life. Thank you to my committee chair, Dr. Nancy Waldron who helped me through this process and kep t me going during these past six years. I would also like to thank Dr. Diana Joyce. Without your support, none of this would have been possible. Your mentorship, advice, and warmth will not be forgotten as I continue my academic pursuits. I would also like to acknowledge my committee members, Drs. Wayne Griffin, Gary Geffken, and James Algina. I am lucky to have worked with all of you, and appreciate all the feedback, encouragement, and advice I have received over the past few years. Dr. Jeanna Mastrodicasa also played an instrumental role in the completion of this dissertation. Thank you for continually helping me navigate the world of academic research and cut through lots of university red tape! And a very special thank you to Yujeong Park, whose statist ical knowledge and amazing friendship helped me through the often grueling process of data analysis. On a personal note, I would like to express my gratitude to Dave Dobrinsky, who has never doubted me, even when I doubted myself. Your unwavering support a nd love means so much to me. Thanks for teaching me how important it is to laugh through the hard times. I would also like to thank my parents, Phil and Candee Rice, who have listened patiently to every crisis, concern, and triumph I have ever had, and hav e always supported me without question. I love you both so much. And thank you to my sister, Jamie. I am so lucky to have you in my life and be able to call you both my sister and my friend.
5 Finally, I would like to thank Cathy, Susan, Suzie, Jenny, and S ally. Throughout these six years, we have shared countless memories and laughs. I could not have asked truly helped me get through these last few years in one piece. I love you all so much and am so thankful to have you in my life.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ 4 LIST OF TABLES ................................ ................................ ................................ ........... 8 LIST OF FIGURES ................................ ................................ ................................ ......... 9 ABSTRACT ................................ ................................ ................................ .................. 10 CHAPTER 1 REVIEW OF THE LITERA TURE ................................ ................................ ............ 12 College Students and Mental Health ................................ ................................ ...... 15 Definition of Non Suicidal Self injury ................................ ................................ ...... 18 Suicide and NSSI ................................ ................................ ................................ ... 22 Prevalence ................................ ................................ ................................ ............. 23 Prevalence among adolescents ................................ ................................ ....... 24 Prevalence among young adults ................................ ................................ ...... 25 Gender Differences ................................ ................................ ................................ 26 Racial, Ethnic, and Socio economic Differences ................................ ...................... 28 Common Methods ................................ ................................ ................................ .. 28 Form of Self Injury ................................ ................................ ................................ .. 30 Etiology ................................ ................................ ................................ .................. 32 Biological Factors ................................ ................................ ............................ 32 Personality Traits ................................ ................................ ............................. 33 Childho od Trauma and Maltreatment ................................ ............................... 35 Co morbid Disorders ................................ ................................ ........................ 39 Epidemiology and Functional Models ................................ ................................ ..... 43 Four function Model (FFM) ................................ ................................ .............. 44 Emotional Regulation Theory ................................ ................................ ........... 47 Anxiety Reduction Model ................................ ................................ ................. 48 Hostility Model ................................ ................................ ................................ 48 Experiential Avoidance Theory ................................ ................................ ........ 49 Social Learning, Modeling, and Reinforcement Perspectives ................................ 49 NSSI Contagion ................................ ................................ ............................... 49 Social Learning Theories ................................ ................................ ................. 51 Treatment Approaches ................................ ................................ ........................... 52 Limitations of Current Research ................................ ................................ ............. 58 DSM V Proposed Revisions ................................ ................................ ................... 63 Problem Statement ................................ ................................ ................................ 70 2 METHODS AND PROCEDUR ES ................................ ................................ ........... 75 Participants ................................ ................................ ................................ ............ 75
7 Measures ................................ ................................ ................................ ............... 76 The Counseling Center Assessment of Psychological Symptoms 62 (CCAPS 62) ................................ ................................ ................................ ...................... 77 NSSI Questionnaire ................................ ................................ ................................ 82 Demographic Information ................................ ................................ ....................... 84 Procedures ................................ ................................ ................................ ............. 84 3 RESULTS ................................ ................................ ................................ .............. 91 4 DISCUSSION ................................ ................................ ................................ ....... 106 Evolution of Research Question and Analysis ................................ ...................... 108 Study Implications ................................ ................................ ................................ 111 Depression ................................ ................................ ................................ ..... 114 Eating Concerns ................................ ................................ ............................ 115 Substance Abuse ................................ ................................ ........................... 115 Family Distress ................................ ................................ .............................. 116 Academic Distress ................................ ................................ ......................... 117 Summary of Implications ................................ ................................ ...................... 117 Limitations ................................ ................................ ................................ ............ 118 Internal Validity ................................ ................................ .............................. 118 External Validity ................................ ................................ ............................. 120 Future Directions ................................ ................................ ................................ .. 120 APPENDIX A IRB AND INFORMED CONSENT ................................ ................................ ........ 122 B CCAPS 62 ................................ ................................ ................................ ........... 126 C CCAPS 62 AUTHOR APPROVAL ................................ ................................ ........ 128 D NSSI SCREENING QUEST ION & QUESTIONNAIRE ................................ .......... 129 E DEMOGRAPHIC INFORMAT ION ................................ ................................ ........ 131 LIST OF REFERENCES ................................ ................................ ............................. 132 BIOGRAPHICAL SKETCH ................................ ................................ ......................... 147
8 LIST OF TABLES Table p age 1 1 Four Function Model of NSSI ................................ ................................ ................ 73 2 1 Ethnicity Breakdown of Sample ................................ ................................ ............. 88 2 2 Gender Breakdown of Sample ................................ ................................ ............... 88 2 3 CCAPS 62 Correlat ion Coefficients ................................ ................................ ....... 89 2 4 Pearson Product Moment Correlation Coefficients of the CCAPS 62 .................... 89 2 5 Data collection timeline ................................ ................................ .......................... 90 3 1 Gender Breakdown by Group ................................ ................................ .............. 100 3 2 Ethnic Breakdown by Group ................................ ................................ ................ 100 3 3 Means and Standard Deviations by Gender and CCAPS 62 Scale for the Past NSSI Group ................................ ................................ ................................ ......... 101 3 4 Means and Standard Deviations by Gender and CCAPS 62 Scale for the Present NSSI Group ................................ ................................ ............................ 102 3 5 Means and Standard Deviations by Gender and CCAPS 62 Scale for the non NSSI Group ................................ ................................ ................................ ......... 103 3 6 Present NSSI compared to no NS SI ................................ ................................ .... 104 3 7 Past NSSI compared to no NSSI ................................ ................................ ......... 104 3 8 Means and Standard Deviations for Each Variable ................................ .............. 105
9 LIST OF FIGURES Figure page 1 1 The experiential avoidance model (EAM) of deliberate self harm .......................... 74
10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of t he Requirements for the Degree of Doctor of Philosophy THE RELATIONSHIP BETWEEN NONSUICIDAL SELF INJURY (NSSI) AND PSYCHOSOCIAL FUNCTIONING IN COLLEGE STUDENTS By Stacey Rice August 2013 Chair: Nancy Waldron Cochair: Diana Joyce Major: School Psychology Over the past few decade s, the psychological needs of college students have become increasingly complex. College counseling centers have seen an increase in the severity of mental health needs in students being seen for counseling (Gallagher, 2009), and the number of students usi ng nonsuicidal self injury (NSSI) as a coping mechanism appears to be increasing as well (Gallagher, 2009). While NSSI scholarship has grown significantly over the past decade, significant gaps in the literature still exist. Conflicting definitions and pre valence rates have made it difficult for researchers to accurately identify and understand individuals engaging in NSSI. The Diagnostic and Statistical Manual for Mental Disorders, Fifth Edition (DSM V) proposed the addition of NSSI as a separate diagnosti c category in order to help accurately diagnose and treat individuals engaging in this behavior. Although NSSI was ultimately not included in the upcoming edition of the DSM, this study used the proposed diagnostic criteria to identify self injurers. Speci fically, this study examined the psychosocial
11 functioning of individuals who currently engage in NSSI, those who engaged in NSSI in the past, and those who have never engaged in NSSI. Seven hundred thirty three college students completed an online survey e xamining eight variables of psychosocial functioning and NSSI behaviors. Multinomial logistic regression found that high levels of anxiety and hostility were the best predictors of current NSSI behaviors. Hostility was also predictive of past NSSI behavior s. Additionally, one way Analyses of Variances (ANOVAs) were used to determine if there were significant differences between current, past, and non NSSI groups and psychosocial functioning in the areas of depression, eating concerns, social anxiety, gener alized anxiety, hostility, academic distress, family distress, and substance abuse. Significant differences were found between groups on all variables except substance abuse. These findings align with prior studies indicating that individuals who engage in NSSI have worse psychological outcomes than those who do not engage in NSSI (Cheng, Mallinckrodt, Soet, & Sevig 2010).
12 CHAPTER 1 REVIEW OF THE LITERA TURE Throughout the last decade, the number of demands on college students has seen a dramatic increase Incidents such as the April 16, 2007 tragedy at Virginia Polytechnic Institute and State University, the January 8, 2011 shooting of U.S Representativ e Gabrielle Giffords, and the July 20, 2012 shooting in a Colorado movie theater have increased public a wareness of the mental health aged population. Increasingly, college much greater than the traditional presenting problems of adjustment and individuation that were seen for college students in counseling center research The increasing mental health demands of the college student population have coincided wit h a significant population increase in community college and university students. According to the U.S Department of Education, between 1997 and 2007, the number of individuals enrolled in college has increased from 14.5 million to 18.2 million (a 26% incr ease) (National Center for Education Statistics, U.S. Department of Education, Institute of Education Sciences). Increased mental health demands combined with an increase in college student population have created significant strains on university counseli ng and mental health centers. Specifically, 93.4% of college counseling center directors report that they have seen a recent trend towards increasing numbers of students presenting for services with severe psychological problems (Gallagher, 2009).
13 Amidst these increasingly complex psychological needs, there is growing concern that more college aged individuals are turning to self injurious behaviors as a coping mechanism. Over the past five years, 55.7% of counseling center directors have identified incre ases in students presenting for self injury concerns. On large college campuses, this number increases to 75% of counseling center directors (Gallagher, 2009). Self injury has become a well known phenomenon, both among clinical and school personnel, as we ll as the general population and popular culture. However, as A The Bright Red Scream, eloquently describes, mutilation has been trivialized (wrist cutting), misidentified (suicide attempt), regarded merely as a sympto m (borderline 1998, p. xii) These discrepancies in identifying and understanding nonsuicidal self injury (NSSI) have made it difficult for researchers and clinicians to properly r ecognize, assess, and treat individuals engaging in NSSI. Significant incongruities have also been found within NSSI scholarship. The definition of NSSI varies considerably among researchers, causing largely different results in studies. Some research inc ludes relatively minor behaviors of NSSI, while others only take into account more severe types of self injury, such as burning or carving skin (Klonsky, 2007). Incidence and prevalence rates of NSSI, along with the functions of NSSI have also been debated among researchers, creating a large amount of inconsistent data within the field.
14 As an introduction to this study, a comprehensive review of the literature on NSSI is prov ided. This review will begin with a synopsis of the current status prevalence rates of NSSI will be discussed. Specifically, research will focus on the adolescent and young adult populations, as they are found to be the most likely individuals to self injure. Following this section, the etiology of self injury will be presented, particularly examining the role biology, personality, childhood maltreatment, and co occurring dis orders play in the development and maintenance of NSSI. The next section will examine the current functional models of self injury, as well as empirically supported treatment options for individuals engaging in self injury. Finally, a critical review of th e literature and the current gaps and limitations in scholarship will be presented. This section will include an argument for considering NSSI as a diagnosis in the May 2013 release of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Editio n (DSM V). Following a review of the literature, a brief description of the current study and methodology will be described. Within the methodology, study participants will be discussed, and the four measures that were used in this study will be described in detail. Additionally, an overview of the procedure for data collection will be provided, as well as the data analysis methods that were u tilized The goal of this study was to determine whether a significant difference in psychosocial functioning exist s between individuals who do not engage in NSSI, those who engaged in NSSI in the past, and those who presently engage in NSSI.
15 Specifically, areas of depression, anxiety, academic distress, eating concerns, family distress, hostility and substance abuse w ill be examined in conjunction with self injurious behaviors. College Students and Mental Health The college age population has recently become a critical component to understanding mental health issues and promoting successful mental health policies. Bec ause most mental illnesses emerge by ages of 15 24, effective interventions with college students may help reduce the chronic prevalence of severe mental health issues (Eisenberg, Golberstein, & Gollust, 2007). College has also proven to be an especially r isky environment for psychological and behavioral problems, e specially when combined with access to alcohol and other substances (Weitzman, 2004). In a nationally representative survey of 27,000 college students (ages 18 24), 4.8% of students were found t o have poor mental health and/or depression within the past 30 days (Weitzman, 2004). The study also found that women, racial or ethnic minorities, and students whose parents did not attend college were more likely to be affected by mental health issues (W eitzman, 2004). A longitudinal study of psychological distress in college students revealed that stress levels generally peaked during freshman year and began a steady decline throughout the remaining years of college, with the exception of a small group o f students who continued to experience significant psychological difficulties (Kitzrow, 2003). Interestingly, college student data has also indicated that undergraduate students are more likely to seek services the longer they remained in school.
16 This may be related to increased access or knowledge regarding services at college campuses. For example, in an average year, the Kansas State University counseling center had the following breakdown of clients: 16.1% of clients were freshmen, 18.3% were sophomore s, 22.7% were juniors, 26.8% were seniors, and 15.4% were graduate students (Benton et al., 2003). Of students surveyed in this study, 75.4% of students were under the age of 25, which falls in the typical age range for college undergraduates (Benton et al ., 2003). Additionally, a 2001 national poll of college freshman found that 28% felt The importance of college student mental health is emphasized by findings indicating that men tal health issues interfere with college attendance, performance, retention and graduation rates (Blanco et al., 2008; Kitzrow, 2003). Between 2001 2006, universities reported between a 40 55% increase in students seeking services through university counse ling centers (Soet & Sevig, 2006). Counseling directors also reported that in the 2002 2003, 40.7% of college age clients presented with severe psychological problems (Kisch, Leino, & Silverman, 2005). Further, recent findings in the literature suggest th at the mental health needs of college aged students have been increasing over the past two decades (Yorgason, Linville, & Zitzman, 2008). Specifically, it has been found that college students are presenting with increasingly severe mental health needs comp ared with the relatively benign concerns of students presenting to counseling centers
17 the National Survey of Counseling Center Directors, 56% of college counseling centers no ted an increase in self injury cases within a one year period (Gallagher, 2009). It is important to note that while many surveys of college counseling centers have noted a perceived increase the mental health needs of college students, these studies genera lly relied on retrospective, reflective surveys by individuals working in these centers ( Benton et al., 2003). R esearch attempting to examine this issue using data such as client perceived scores of distress at intake, have found no significant differences across 6 8 years ( Cornish, Kominars, Riva, McIntosh, & Henderson, 2000). Benton et al., (2003 ), however, conducted a study using more objective data in college counseling centers, and found that in recent years, students being se rved in counseling centers have more complex problems than in the past. Additionally, they found that the number of students seen each year with depression doubled over the examined time period of 13 years, the number of suicidal students tripled, and the number of students seen af ter a s exual assault quadrupled (Benton et al., 2003). The effectiveness of new psychiatric medications is one possible reason for the increase in students with complex psychological needs. These medications have allowed students with significant psycholo gical disorders to attend college, where as in the past it may have been difficult for them to do so (Kitzrow, 2003). It is also important to note that although there is increased concern about mental health on college campuses, the improved health promotio n and mental health services on campuses have helped improve some
18 aspects of the health for students. For example, between 1990 and 2004, the suicide rate for students at 309 colleges and universities was approximately 6.5 per 100,000. In a matched sample of the general U.S. population not attending college (by gender and age), the suicide rate was approximately 12.6 per 100,000; or roughly double that on college campuses (Kraft, 2009). The increased demand in counseling services does not necessarily corre late with an increase in knowledgeable professionals at the university level. For example, one study found that 63% of campus counseling centers reported that an increased demand in services without a corresponding increase in resources is a major challeng e and concern (Kitzrow, 2003). Of the campus counseling centers, only 21% reported an increase in professional staff during the p rior year (Kitzrow, 2003). One longitudinal study found that mental health issues found in the college age population were larg ely persistent problems that remained throughout adulthood (Zivin, Eis enberg, Gollust, & Golberstein 2 009). Definition of Non Suicidal Self injury Non suicidal self injury (NSSI) is defined as purposeful and direct injury to ct intent to die (Nock & Favazza, 2009). Other terms for this behavior have included self harm, parasuicide, deliberate self harm (DSH), self injurious behaviors (SIB), and self mutilation (SM) (Borrill, Fox, Flynn, & Roger, 2009). In 1979, it was reported that there were as many as 33 terms for this behavior (MacAniff Zilla & Kiselica, 2001). The range of terms and defin itions for the behavior has often caused confusion in the research literature, primarily around the ideas of suicidal intent and how direc t the injury to the self is. For example, the category of DSH often includes risk taking behaviors such as
19 jumping from heights or drug overdoses (Heath, Baxter, Toste, & McLouth, 2010) that are not usually included in other definitions of self injury. Cur rently, NSSI is the most widely recognized term for these behaviors in the literature and will be used throughout this paper. In response to the lack of consensus regarding the definition of NSSI, a group of leading researchers and clinicians established t he International Network for the Study of Self Injury (ISSS) in 2006. One year later, the ISSS developed a definition of NSSI to help clarify the term and its associated behaviors (Heath, Toste, Nedecheva, & Charlebois, 2008). As such, NSSI is defined as: The deliberate, self inflicted destruction of body tissue resulting in immediate damage, without suicidal intent and for purposes not socially sanctioned. As such, this behavior is distinguished from: suicidal behaviors involving an intent to die, drug o verdoses, and other forms of self injurious behaviors, including culturally sanctioned b ehaviors performed for display or aesthetic purposes; repetitive, stereotypical forms found among individuals with developmental disorders and cogn itive disabilities, a nd severe forms (e.g. self immolation and auto castration) found among individuals with psychosis. (ISSS, 2007). Historically, self injury has been classified into four broad categories. This taxonomy was developed by Armando Favazza, one of the earliest self injury researchers, and separates self injury by method, frequency and intensity (Yates, 2004). Sterotypic SIB includes individuals with severe and pervasive developmental disorders (DD) such as mental retardation (MR; currently kn own
20 as Intellectual Disability [InD]), Lesch This type of self injury often has a repetitive, rhythmic quality, is generally performed without regard for social context, and without significant affect (Yates, 2004). The most common type o f stereotypic SIB is head banging (Favazza, 1996). Major SIB usually occurs suddenly and often involves a great deal of tissue damage and bleeding (Favazza, 1996). Generally, Major SIB occurs within the presence of a psychotic episode and often includes se lf enucleation and autocastration (Yates, 2004). Compulsive SIB is often associated with disorders such as trichotillomania and includes repetitive behaviors such as hair pulling and nail biting that occur many times each day. It is usually classified as a n impulse control disorder (Yates, 2004). Finally, Impulsive SIB can be categorized as episodic or repetitive. Individuals engaging in Impulsive SIB may injure themselves intermittently, and the behavior often functions to regulate emotion or decrease tens ion (Yates, 2004). Impulsive SIB may begin to take on repetitive qualities over time, and may take on addictive qualities as more time is devoted to self injury (Yates, 2004). T he definition of NSSI rules out two of the four categories that Favazza posite d; Sterotypic SIB related to D evelopmental D isabilities or In tellectual D isabilities and Major SIB related to psychosis. Therefore, these will not be the Impulsive SIB are distinguishe d by either impulse control or emotional regulation functions respectively. Compulsive SIB is noted to be associated with DSM IV (American Psychiatric Association [APA], 2000) mental health disorders such as
21 Trichotillomania. The repetitive nature of thes e specified behaviors as well as the daily frequency are consistent with Obsessive Compulsive D isorder (OCD). Although both Trichotillomania and OCD can be exacerbated by stress, precipitating emotional regulation stressors are not diagnostic criteria. For example, Trichotillomania often manifests during periods of relaxing or distracting activities and is often preceded by an itchy scalp feeling. In fact, gratification, pleasure, or relief from hair pulling activities is a required criterion. Although prec eding tension is also a required symptom, it can be related to resisting the hair pulling urge and is not necessary general tension related to stress (APA, 2000, p. 674 675). For OCD, the precipitating features are recurrent obsessions or compulsions. The obsessions criteria requires persistent thoughts, impulses or images that are not worries about real life problems and for adults, the person 462) and unreasonable or excessive. Most obsessions are related to contamination, doubt, aggressive impulse, or sexual imagery. Compulsions are behaviors that directly reduce the tension related to the unrealistic thought, impulse or image (e.g., hand washing). In contrast to Compulsive S IB components, an intermittent and episodic frequency as well as a general emotional regulation function are the core aspects consistent with literature on NSSI. Therefore, for the pur poses of this study, NSSI will only describe impulsive self injury. Further, the te rms NSSI or self injury will only be used to describe directly injurious behaviors lacking clear suicidal intent.
22 That is, inclusion of individuals engaging in self injuriou s behavior while denying the desire to end their life, and often citing other reasons for self injuring, such as tension relief, coping skills, or to distract oneself (Brown, Comtois, & Linehan, 2002). Suicide and NSSI Although self injury has been report ed as a significant risk factor for suicidal behavior and gestures (Brausch & Gutierrez, 2010), research has determined that NSSI is a behavior distinct from the suicidal continuum (Stanley et al., 2009). Generally, NSSI is understood as separate from suic idality in terms of motivation and medical severity. Specifically, those engaging in NSSI often utilize the behavior as a way to prolong their life, rather than end it. That is, a large number of individuals engaging in NSSI use it as a functional coping m echanism to improve their mood and decrease feelings of helplessness (Klonsky, 2007). Additionally, NSSI usually involves less tissue damage and attention from medical professionals (Glenn & Klonsky, 2009). NSSI and suicide are also distinct in several o ther ways. A suicide attempt encouraging therapy or hospitalization, subsequently decreasing the number of attempts. However, when NSSI is discovered, family and other individ uals may react with disgust, anger or denial, which does little to decrease the number of subsequent self injurious episodes (MacAniff Zila & Kiselica, 2001). Further, suicidal individuals who are removed from stressful environments often improve, while NS SI individuals tend to continue self injurious patterns of behavior, even when removed from stressful situations (MacAniff Zila & Kiselica, 2001).
23 While, it is important to note that although the behaviors themselves are separate, NSSI is a known risk fac tor for suicidal thoughts and attempts (Glenn & Klonsky, 2009). Approximately 50 70% of adolescents engaging in NSSI also have a history of suicide attempts (Dougherty et al., 2009), and between 34 45% of high school and college aged self injurers report s uicidal ideation (Whitlock & Knox, 2007). Although suicide attempts and NSSI are distinct behaviors, one theory posits that self injurious behavior may serve as a precursor to suicide attempts. Specifically, individuals engaging in NSSI may become more con fident and less concerned by the pain and associated feelings of NSSI. Consequently, it may allow an individual to habituate to the idea of suicide and give them the confidence to make a suicide attempt (Nock, Joiner, Gordon, Lloyd Richardson, & Prinstein, 2006). It has also been found that suicide attempts are more likely among adolescen ts who repetitively self injured over long periods of time, use d different methods of self injury, and repo rted feeling no pain while engaging in NSSI (Plener, Libal, Kelle r, Fegert, & Muehlenkamp, 2009). Prevalence Current prevalence rates of NSSI behavior appear to be increasing, specifically among the nonclinical adolescent and young adult populations (Brausch & Gutierrez, 2010). Lifetime prevalence rates of NSSI in the general population are estimated to be between 10 15%, with 5 10% of individuals engaging in repeated episodes (Yates, 2004). Self injurious behaviors also have long been associated with intellectual disabilities or organic neurological disorders (Favazza 1996).
24 Prevalence among adolescents Historically, self injury has been viewed as a behavior that occurs primarily within the inpatient psychiatric population, with research suggesting that upwards of 80% of adolescent inpatient samples engage in NSSI be haviors (Glenn & Klonsky, 2009). However, recent research indicates that NSSI behaviors are becoming an alarming trend among nonclinical, community samples of adolescents. These samples have estimated that as many as 13% of adolescents have engaged in NSSI behaviors (Ross & Heath, 2002). According to Ross & Heath, (2002), m ost adolescents engage in NSSI in the absence of serious pathology or mental illness The average age of onset for self injury occurs between 11 15 years old (Heath, Toste, & Beettam, 20 07), with some studies suggesting that individuals as young as 10 engage in NSSI. In the middle school population, self injury rates appear to be slightly lower, with approximately 7.5% of young adolescents engaging in self injury (Hilt, Nock, Lloyd Richar dson, & Prinstein, 2008). Most adolescents in the community population engage in what is referred injury is generally characterized as a compulsive behavior that is ritualistic and often occurs with out premeditation. It is also usually episodic and repetitive, in that, it occurs periodically, and the individual often does not identify as a self injurer (Whitlock, 2010). This population may create injuries that are mild, moderate, or severe (Whitlock, 2010). While NSSI often co occurs with psychiatric disorders, common NSSI is more likely to occur in the absence of a comorbid psychiatric diagnosis (Whitlock, 2010).
25 Prevalence among young adults Currently, young adults (ages 18 25) appear to have th e highest risk for engaging in NSSI (Rodham & Hawton, 2009; White, Trepal Wollenzier, & Nolan, 2002). Literature suggests that NSSI prevalence rates range from approximately 14 38% of college age students ), with a 12 month prevalence rate of 7.3% (Gollust, Eisenberg, & Golberstein, 2008 ; Laye Gindhu & Schonert Reichl, 2005 ). The large discrepancies in college age prevalence rates present a methodological concern for researchers and practitioners seeking more precise rates of NSSI within this population. Cl ose examination of the literature shows significant difference s in methodology, sampling, and survey techniques, which may lead to skewed estimations of individuals engaging in NSSI. For example, in a 2007 study conducted by Lloyd Richardson, Perrine, Dier ker, and Kelley, initial findings indicated 46% of a community sample of adolescents had engaged in NSSI. However, after closer review of the data, they determined that only 60% of injury, which was characte rized by cutting/carving, burning self (rubbing an eraser over the skin until burning or bleeding results) (Lloyd injury, which included less clinically significant behaviors such as, hitting self, pulling hair, biting self, picking at a wound, or picking at the skin to draw blood (Lloyd injurio us behaviors, many studies do not distinguish fall into the same classification as someone engaging in severe NSSI. In
26 addition, studies often do not distinguish the frequency of b ehaviors. Therefore, a person engaging in an incident once may be categorized with individuals who chronically engage in the behavior. Discrepancies in the literature also may be due to differences in measurement sensitivity (Heath et al., 2008). While som e studies ask participants to identify specific NSSI behaviors, others ask participants broader 2008). Still other studies provide students with an extensive list of self injurious behaviors and ask them to endorse any in which they have engaged. This method also tends to produce higher prevalence rates, as checklists may include behaviors not always associated with self injury (e.g. scraping skin or interfering with wound healing) (Heath et al., 2008). Studies using measures with broad definitions of NSSI are likely to yield higher prevalence rates among college students. For example, Gratz (2006) utilized a broad measure and reported a 37% prevalence rate for college students, whil e Heath et al., (2008) found a rate of only 11.68% of college students with a more specific measure. Gender Differences Currently, there is little consensus regarding possible gender differen ces in NSSI behaviors, although preliminary studies suggest tha t the frequency and severity of self injury may be mediating factors by gender. While it is commonly believed that females engage in self injurious behaviors more frequently than males, research has remained inconclusive (Laye Gindhu & Schonert Reichl, 200 5). Several studies investigating prevalence rates of NSSI in community samples have found no significant difference in gender (Hilt, Nock, Lloyd
27 Richardson, & Prinstein, 2008; Lloyd Richardson, Perrine, Dierker, & Kelley, 2007) while others have found tha t females are more likely to engage in NSSI behaviors (Hawton, Rodham, Evans, & Weatherall, 2002; Rodham & Hawton, 20 09). Whitlock, Eckenrode, & Silverman (2006) conducted a study in a sample of college students (excluding students participating in counsel ing) and found that among isolated self injurious episodes; there w as no significant gender difference. However, the reported rate for repeated NSSI episodes was approximately 1.5 times higher for college women than for men. When examining possible gender differences in NSSI behaviors, it is important to note that significant differences in gender are found more often when examining clinical rather than community samples (Heath et al., 2008). This may be because females are more likely to seek help or beca use overdose or inappropriate ingestion of medication is often included in the clinical definition of NSSI (Heath et al., 2008). These behaviors are more typically found in the female population and therefore, may contribute to some of the gender differenc es found in the research. Females also tend to be overrepresented in the general clinical population, making gender differences more likely (Whitlock, Eckenrode, & Silverman, 2006). Some research suggests that homosexual and bisexual individuals, or those questioning their sexuality are more likely to engage in NSSI than their heterosexual counterparts (Lofthouse & Yager Schweller, 2009; Whitlock, Eckenrode, & Silverman. 2006). Rates of self injury continue to be disproportionate among the Lesbia n, Gay, B isexual, Transgender, & Questioning
28 (LGBT Q ) individuals, specifically among younger members of the population (Alexander & Clare, 2004). Alexander and Clare (2004) argue that the reason for increased rates of NSSI among this population is associated with t he continued social pressure and marginalization of the LGBT population. One British newspaper conducted interviews of homosexuals engaging in self injury and found many common themes among them, including feelings of having no one to turn to for advice, t he tendency to self their sexual identity, family hostility, discrimination and homelessness (Alexander & Clare, 2004). Racial, Ethnic, and Socioeconomic Differences There is currently a dearth of information reg arding possible racial or socioeconomic status (SES) differences among individuals engaging in NSSI. Several studies suggest that the incidence of self injury is higher among Caucasian adolescents, (Ross & Heath, 2002) but few systematic research studies h ave been conducted to determine if there is any significant difference in NSSI behaviors among other ethnic groups or SES levels. Some researchers have hypothesized that NSSI may be overrepresented in the middle to upper class Caucasian culture (Yates, 200 4), however, a recent study by Hilt, Cha, and Nolen Hoeksema (2008) found no significant differences within ethnic groups in a sample of 94 adolescent girls. Common Methods Although various methods of self injuring have been identified, cutting is the mo st common method, occurring in 70 97% of those who self injure (Klonsky, 2007), followed by banging or hitting (21% 44%) and burning (15 35%) (Rodham
29 & Hawton, 2009). Other behaviors classified as NSSI include carving of the skin, pulling skin or hair, and bruising or breaking bones (Whitlock, Eckenrode, & Silverman, 2006). More mild forms of self injury may include scratching the skin or preventing wounds from healing (Kanan, Finger, & Plog, 2008). The location of the body where self injury occurs is also an important component of understanding NSSI The majority of self injury occurs on the abdomen or extremities (Walsh, 2007). Individuals engaging in self injury on the face, eyes, neck, jugular region, breast, or genitals, may exhibit more severe psychol ogical problems than those who self injure in other areas and are more often found in the clinical population rather than among community samples (Whitlock, 2010). Often, these individuals are experiencing psychotic decompensation or trauma related behavio r and should be referred immediately for emergency services (Walsh, 2007). The direct harmful consequences associated with NSSI also distinguishes it from other risk taking behaviors such as smoking and using drugs, which generally have unintended negativ e consequences. Culturally sanctioned body modification such as tattoos and body piercing are also considered outside the realm of NSSI behaviors (Nock, 2009). While definitions of self harm often include forms of indirect self injury such as ingesting an illicit or recreational drug, jumping from a height, or ingesting a non digestible substance, these acts generally do not fall within the American and Canadian definitions of NSSI (Heath, Schaub, Holly, & Nixon, 2009).
30 Form of Self Injury Currently, there is scant literature investigating how often individuals self injure and under what circumstances. This represents a significant gap in linking scholarship to intervention, as practitioners have little information about the circumstances surrounding specifi c incidents of NSSI. A study conducted by Nock, Prinstein, and Sterba, (2009) used a community sample of 30 adolescents engaging in NSSI to investigate the environmental context and circumstances surrounding instances of NSSI thoughts and behaviors. This s tudy was conducted using an ecological momentary assessment (EMA) method, wherein adolescents are able to use a computer to record their self injurious thoughts and behaviors in real time, outside the laboratory or clinic (Nock, Prinstein, & Sterba, 2009). The EMA method is an important data collection innovation as it allows participants to record their thoughts and behaviors without relying on retrospective recording, reducing errors caused by memory. Individuals participating in the study experienced an average of five NSSI thoughts per week. These thoughts were reported to be of moderate intensity and lasted between 1 30 minutes. Approximately 86% of these individuals acted on at least one of these thoughts, with an average of 1.6 NSSI episodes per week Additionally, 33.3% of participants experienced at least one suicidal thought during the study period, and averaged 1.1 suicidal thoughts per week. A qualitative study conducted with 154 self injuring individuals found that 51.6% of participants self inj ured at least once a week, and 24.5% self injured at least once a day (Polk & Liss, 2009). Other research suggests that an individual engaging in repetitive NSSI will engage in self injurious behavior an average of
31 50 times, although some individuals repor t 400 or more independent episodes of self injury (Muehlenkamp, 2005). Nock, Prinstein, and Sterba (2009) also found that adolescents were most often alone when they began experiencing self injurious thoughts, although they still experienced a significant number of thoughts when they were with peers and friends. Self injurious thoughts were experienced least often in the presence of family or strangers. Adolescents who self injure d only while alone were also found to have a higher association with suicidal thoughts and behaviors (Glenn & Klonsky, 2009). Additionally, although earlier research has found that self injurious thoughts often occur in conjunction with alcohol or substance use, Nock, Prinstein and Sterba (2009) found that the majority of NSSI rel ated thoughts occurred while adolescents were sober. These studies have significant implications for improving interventions by providing an environmental context to NSSI thoughts and behaviors. Interventions that take environmental context of NSSI into ac count can include coping skills specific to the time and place associated with NSSI. For example, knowing that most incidents occur when a student is alone would imply the importance of ready access to support networks as an intervention component (e.g., h otline or family member to call). However, because these studies still rely on self report measures, it is difficult to assess the accuracy of the results. The EMA method improves on self report measures but it may still be difficult to ascertain the validity of these studies.
32 Etiology The etiology of NSSI is complex, and can best be understood through analysis of both biological vulnerabilities and environmental risk factors (Crowell, Beauchaine, McCaul ey, Smith, Vasilev, & Stevens, 2008). The subsequent section explains some of the many factors associated with NSSI, including biology, personality traits, childhood trauma, and common co morbid disorders. Biological Factors The role of biology in NSSI is complex, and cannot fully explain the etiology of this complicated behavior. However, the serotonergic and opioid systems have been consistently implicated in NSSI, both within the developmentally disordered and psychiatric populations (Yates, 2004). Decr eased levels of serotonin production have been associated with increased aggression, impulsivity, suicidality and NSSI (Yates, 2004). These findings have been replicated among both humans and rhesus monkeys, and contribute strong support to the contributio n of a dysregulated serotonergic system in NSSI behaviors (Sher & Stanley, 2009). Selective Serotonin Reuptake Inhibitors (SSRIs), a class of medications used to maintain the availability of serotonin in the brain, has been associated with reducing the inc idence of NSSI behaviors (Schroeder et al., 2001). Endogeneous opioids are also implicated in the development and maintenance of NSSI behaviors (Yates, 2003). Research has determined that individuals experience increased rates of opioids when they self in jure ( Chapman, Gratz, & Brown, 2006). Notably, most individuals engaging in NSSI report that they feel little or no pain prior to or during self injurious episodes. This
33 stress induced analgesia may be related to the dissociation that individuals often rep ort before they self injure (Bohus, Limberger, Ebner, Glocker, Schwarz, Wernz, & Lieb, 2000). As a result of the role endogenous opioids play in NSSI, pharmacological treatments such as naltrexone have been found to reduce self injurious behavior in many i ndividuals (Bohus et al., 2000). Personality Traits Due to the complex nature of self injury, it is not possible to conclusively determine why all individuals engage in NSSI. However, several theoretical frameworks have been developed to propose reasons w hy some individuals engage in this behavior (Klonsky, 2007). In addition, other researchers have attempted to isolate specific personality constructs that are more highly correlated with NSSI (Andover et al., 2005; Dougherty et al., 2009; Janis & Nock, 200 9). Impulsivity and perfectionism, two personality traits often associated with NSSI, are discussed below. Research has implicated impulsive behaviors or difficulty with impulse control as a feature of NSSI (Janis & Nock, 2009). Impulsivity can be define d as the tendency to act quickly and without regard for consequences. Impulsive individuals often have difficulty inhibiting their responses (Herpertz, Sass, & Favazza, 1997), and impulsivity has also been linked to emotional dysregulation (Herpertz, Sass, & Favazza, 1997). A 2009 study by Stanford and Jones attempting to determine whether self injurers could be considered a psychologically homogeneous group found that impulsivity played a role in approximately one third of adolescents engaging in NSSI. B ased on their answers to a self report questionnaire, the individuals in the study were
34 organized into normal pathological, and impulsive subtypes. More specifically, this study found that the correlation between impulsivity and NSSI was higher for males (34 out of 76) than females (33 out of 142) (Stanford & Jones, 2009). Individuals engaging in NSSI with higher levels of impulsivity have also been found to have increased risk for subsequent suicidal ideation or attempts (Dougherty et al., 2009). While s ome individuals report impulsively engaging in self injury, other individuals report a more compulsive component to the behavior. Studies have examined how much time is usually spent thinking about engaging in N SSI before engaging in the act with inconsist ent results (Janis & Nock, 2009). It has been noted that some individuals ritualize the behavior and spend long periods of time planning or thinking about NSSI before actually completing it (Connors, 1996). Another construct that has been studied in conce rt with NSSI is that of perfectionism. Perfectionism is defined in the literature as attempting to attain unrealistically high expectations in several areas of functioning (Hoff & Muehlenkamp, 2009). Perfectionism has been linked to cognitive distortions t hat may lead to even more serious forms of maladaptive coping or problem solving abilities. Maladaptive perfectionism also has been shown to correlate strongly with several other forms of pathology, including suicidal ideation (Hoff & Muehlenkamp, 2009) an d Obsessive Compulsive Personality Disorder (APA, 2000, p. 729). Several studies have suggested perfectionism as a possible clinical correlate to NSSI (Cross, 1993; Strong, 1998; White, Trepal Wollenzier, &
35 Nolan, 2002); however, only limited empirical evi dence has been found to support this notion. NSSI indicated that individuals did not score significantly higher on all subscales of perfectionism, although they did differ significantly on three perfectionism subscales; concern over mistakes, parental criticism and organization. This study suggests that while overall perfectionism may not strongly influence NSSI, it is possible that specific aspects may indicate vulnerability to self inju ry (Hoff & Muehlenkamp, 2009). Ch ildhood Trauma and Maltreatment Historically, self injury has been associated with psychoanalytic models that were often based on childhood abuse and trauma, and more specifically, sexual abuse. For example, in her book, T he Bright Red Scream, Marilee The only recourse [to sexual abuse] is psychic defenses denial, self blame, dissociation, repression to blunt the overwhelming horror of the experience and feel some sense of control. This can lead to a quite successful front presented to the public and a secretive, shame filled inner self compulsively re enacting the trauma in a futile attempt to master it (1998, p. 67). Early studies demonstrated significant, positive associations between childhood sexual abuse and subsequent self injury (Muehlenkamp, Kerr, Bradley, & Larsen, 2010), but this research was often based on case studies and ethnographic accounts of adolescent and young adult women engaging in self injurious behavior. Many early self injury researchers suggested that NSSI may
36 them (Klonsky & Glenn, 2009). NSSI has also been described as a the functions of NSSI following abuse as fourfold. Self injury is used to 1) re enact the original trauma, 2) organize the self and regain and maintain homeostasis, 3) manage dissociative symptoms and 4) express feelings and needs. While some researchers suggest a link between sexual abuse and NSSI, it is difficult to fully validate this relationship, as the majority of research has been cond ucted using very small sample sizes or qualitative samples (Cavanaugh, 2002; Klonsky & Moyer, 2008). Systematic, quantitative research has yielded less consistent results (Klonsky & Glenn, 2009). For example, a meta analysis including 43 studies examining the association between childhood sexual abuse and NSSI found a relatively small relationship between the two (Klonsky & Moyer, 2008). Specifically, these aggregated studies found that childhood sexual abuse accounts for no more than 5 % of the variance in the development of NSSI (Klonsky & Moyer, 2008). Several studies suggest that family difficulties or childhood maltreatment may play a role in subsequent NSSI behaviors (Prinstein, Guerry, Browne, & Rancourt, 2009). A well established theory put forth by Marsha Linehan suggests that early invalidating environments are directly associated with poor emotional regulation and interpersonal skills. She proposes that these early invalidating environments put individuals at higher risk for developing NSSI as a m aladaptive
37 theory and demonstrated that environments with problematic attachments or abusive environments put children at higher risk for NSSI (Klonsky & Glenn, 2009). It h as also been shown that those individuals engaging in NSSI reported a lower quality of relationship with their parents than individuals not engaging in NSSI (Hilt et al., 2008). Although it has been noted that many individuals engaging in NSSI have a hist ory of childhood maltreatment, it is unclear whether the association is a specific risk factor for NSSI or a predictor of psychopathology in general (Prinstein et al., 2009). Some studies suggest that individuals who engage in self injury experienced more emotional or physical neglect as children (Klonsky & Glenn, 2009), while others note that childhood sexual abuse and physical neglect appear to be the most strongly correlated with NSSI, with physical abuse and emotional neglect less predictive of NSSI beh aviors (Glassman, Weierich, Hooley, Deliber to & Nock, 2007). Yates, Carlson, & Egeland, (2008) found that physical abuse better predicted intermittent NSSI, while sexual abuse was a better predictor of recurrent NSSI, suggesting complex relationships betwe en early abuse and subsequent NSSI. Still other researchers have argued that there are much stronger associations between NSSI and physical abuse and neglect than sexual abuse (Muehlenkamp, Kerr, Bradley, & Larson, 2010). Specifically, physical abuse is f ound to correlate more strongly with repetitive NSSI than those engaging in episodic NSSI (Muehlenkamp et al., 2010). Although correlational data cannot be
38 that suffering from ph ysical abuse may desensitize a person to physical pain, making it more likely that the individual will engage in NSSI as a coping mechanism rather than a less painful method such as substance abuse (Muelenkamp et al., 2010). Current research suggests that the relationship between abuse and NSSI may be more thoroughly und erstood through the use of medi ational models (Muehlenkamp et al., 2010). That is, other clinically relevant variables may help at least partially explain the relationship between NSSI and childhood abuse. For example, one study found Post Traumatic Stress Disorder (PTSD) was a mediating variable for the abuse/NSSI relationship, while research conducted by Glassman et al., (2007) indicated self criticism to be a primary mediating variable in the relationship (Muehlenkamp et al., 2010). Additionally, a study conducted by Cheng, Mallinckrodt, Soet and Sevig (2010) found that experiencing a traumatic event was a significant predictor for NSSI only among women, whereas witnessing trauma predicted self injury for both men and women. Overall, data from current research suggests a complex and possibly mediated association between NSSI and abuse that have a strong component of emotional regulation difficulties (Muehlenkamp et al., 2010). For example, Muehlenkamp et al., (2010) noted that individuals engaging in NSSI, as well as those experiencing significant childhood abuse reported significant difficulties regulating their emotions. Additionally, Paivio and McCulloch (2004) suggested that alexithymia (the inability to accurately identify and express emotions appropriately) mediated the relationship between abuse and SIB. Specifically, abuse is related to self injury,
39 especially when the individual experiences difficulty expressing negative affect or em otion (Polk & Liss, 2007). Although there is disagreement regarding the correlation between NSSI and early childhood abuse, most studies agree that self injurers are more likely to report early negative family environments (Klonsky & Glenn, 2009). Co morb id Disorders While NSSI is not listed as a separate diagnostic category, the behavior is often associated with a variety of DSM IV Axis I and Axis II disorders, (APA, 2000) including eating disorders, posttraumatic stress disorder (PTSD), anxiety, depressi on, borderline personality disorder (BPD) and obsessive compulsive disorder (OCD) (Ballard, Bosk, & Pao, 2010; Kress & Hoffman, 2008). In fact, many adolescents engaging in NSSI also experience symptoms consistent with internalizing, externalizing, and sub stance use disorders on Axis I and borderline, avoidant, and paranoid personality disorders on Axis II (Nock, Teper, & Hollander, 2007). Borderline Personality Disorder (BPD) is most commonly associated with self injury, and is usually associated with the clinical population. In fact, the only time self injury is mentioned in the DSM IV TR is as a symptom of BPD. This is problematic, because BPD is generally seen as an inappropriate diagnosis for children under the age of 18 because of their developing per sonalities (Wilkinson & Goodyer, 2011). It is estimated that between 70 80% of individuals diagnosed with BPD engage in some form of self injurious behavior (Kress & Hoffman, 2008). BPD is characterized as a disorder in which individuals experience affecti ve instability, dysregulated behaviors, and an intense fear of abandonment
40 (Selby, Anestis, Bender, & Joiner, 2009). The diagnostic criterion for BPD also mutilating Emotional dysregulation is one of the hallmark symptoms 2006). It has also been suggested that emotional dysregulation for individuals with BPD is experienced as 1) extrem ely intense experience of emotion, 2) increased sensitivity to emotional stimuli, and 3) difficulty returning to emotional baseline (Linehan, 1993). Although diagnosis prior to age 18 is generally discouraged, many adolescents (especially in the inpatient setting) may show signs that are indicative of BPD, including using self injury as an impulsive reaction to real or perceived abandonment (Kress & Hoffman, 2008). Adolescents with features of BPD may also engage in a wide variety of high risk behaviors th at often include substance abuse, promiscuity, self mutilation and eating disorders. These high risk behaviors are often the determining factor in hospitalization, especially when the adolescent has accidentally overdosed, cut themselves too deeply or bec ome pregnant (Bleiberg, 2001). Anxiety is also strongly associated with self injurious behaviors in both the clinical and community populations, and NSSI is believed to coincide with a significant reduction in tension (Andover, Pepper, Ryabchenko, Orrico, & Gibb, 2005). Individuals engaging in NSSI generally have maladaptive coping strategies, and it has been hypothesized that as anxiety increases, these individuals use self injury as a way to cope with feelings of tension (Ross &
41 Heath, 2003). That is, as anxiety builds to intolerable levels, some individuals use NSSI as a strategy to decrease anxiousness (Ross & Heath, 2003). For individuals engaging in NSSI for anxiety reduction, there is generally an immediate feeling of relief following self injury (Fa vazza, 1998). Specifically, individuals with high levels of anxiety showed significant decreases in respiration, skin conductancy level and heart rate in response to NSSI scripts (Andover et al., 2005). These findings suggest that individuals with high lev els of anxiety may be more likely to engage in NSSI to decrease their levels of anxiousness (Andover et al., 2005). While there is significant co morbidity between anxiety and NSSI, the connection between NSSI and depression is less consistent (Andover et al., 2005). Some research has suggested that there is a link between individuals experiencing major depressive episodes and NSSI, while other studies have found no link between a diagnosis of major depression and self injury (Andover et al., 2005). Guerry & Prinstein, (2010) found a cognitive vulnerability stress interaction as a significant predictor of NSSI in the time period of 9 18 months following a hospital discharge. They noted that individuals with more negative attributional styles in combination with more stressful interpersonal events reported increasing levels of NSSI behaviors over time. They also found that individuals engaging in NSSI experience significantly higher levels of negative affect, as well as lower levels of distress tolerance (Gue rry & Prinstein, 2010). Additionally, eating disorders also demonstrate co morbidity with NSSI. Research suggests that approximately 22 62% of individuals diagnosed with
42 anorexia (Croyle & Waltz, 2007; Hilt et al., 2008; Serras et al., 2010), and up to 7 2% of individuals diagnosed with bulimia (Croyle & Waltz, 2007) also engage in self injurious behaviors. There are several common characteristics among those who engage in NSSI and individuals identified with eating disorders. In both cases, adolescence is the typical age of onset for these disorders, and they are both commonly associated with body dissatisfaction and self punishment (Favaro & Santonastaso, 2000). Both eating disorders and NSSI are generally conceptualized as a set of behaviors that serve a n emotional regulation function (Muehlenkamp, Engel, Wadeson, Crosby, Wonderlich, Simonich, & Mitchell, 2009). There is also significant evidence to suggest co morbidity between substance abuse and NSSI. Self injury is common among clinical substance usin g samples, with prevalence ratings ranging from 34 50% (Serras, Saules, Cranford, & Eisenberg, 2010). In the Serras et al., (2010) study, drug use was found to be associated with higher rates of all self injurious behavior, however, the relationship betwee n alcohol use and NSSI was less clear. Serras et al., (2010) found that, while binge drinking behavior within the past two weeks did not correlate with NSSI, frequent binge drinking behavior did. This distinction is consistent with the research that sugges among college students (consuming 4 or more drinks on one occasion within the past two weeks) is not significantly correlated with poorer mental health outcomes (Serras et al., 2010). Gollust and colleagues (2008) found similar results in their investigation of binge drinking and self injury. That is, no
43 significant relationship was found between binge drinking and self injury. However, in this study, no distinction was made between binge drinking and frequent binge drinking, which may have accounted for the lack of statistically significant results. Epidemiology and Functional Models Although research in the area of NSSI has increased substantially, it is still difficult for psychologists and researchers to determ ine the multi faceted etiology of this complex behavior. As Lloyd contextually complex, meaning that an individual is inextricably tied to his p. 31). That is, an individual may engage in sel f injury for different reasons at different times and contexts, which changes the functions of the behavior. Additionally, self over determined behavior, meaning that it may simultaneously serve several functions for an i ndividual (Lloyd Richardson, Nock, & Prinstein, 2009). As a result, multiple models of NSSI have been proposed to explain the functionality of the phenomenon. The theories presented in this section have been developed by key self injury researchers and ge nerated significant empirical support. However, it should be noted, there is significant overlap within theories, and researchers often use different terms to describe similar functions and behaviors. Therefore, while the theories are presented as separate there are many significant aspects of each theory that may be closely associated with other models.
44 Four function Model (FFM) The four function model (FFM) has recently been posited to help understand the processes that produce and maintain NSSI behavio r (Nock & Cha, 2009). The model was developed as a comprehensive, integrative model that draws on information from learning theory and behavior therapy (Lloyd Richardson, Nock, & Prinstein, 200 8 ) According to this model, self injury functions along two di chotomous dimensions: reinforcement that is either positive or negative, and contingencies that are either autom atic (i.e., intrapersonal) or so cial (i.e., interpersonal) (Nock & Cha, 2009). These four functions are not mutually exclusive, and individuals often engage in self injury for more than one purpose. (Hilt, Nock, Lloyd Richardson, & Prinstein, 2008). The four function model is illustrated in Table 1 1 (Lloyd Richardson, Nock, & Prinstein, 2008 p. 34). The first function proposed by this model is the automatic negative reinforcement (ANR) function. In this function, self injury serves to regulate an emotional state (Lloyd Richardson, Nock, & Prinstein, 2009). The ANR fu nction is the one most often endorsed by self injurers in research studies, (Nixon, Cloutier, & Aggarwal, 2002; Nock & Cha, 2009; Polk & Liss, 2009) and is also the only function that is closely associated with a history of suicidal attempts and hopelessne ss (Lloyd Richardson, Nock, & Prinstein, 2009; Nock & Prinstein, 2005). Studies with both the community population and hospitalized inpatients lend strong support to the ANR function, with individuals often citing reasons for
45 self Richardson, Nock, & Prinstein, 2009, p. 33). The second function proposed by the FFM is automatic positive reinforcement (APR). These individuals engage in NSSI for the purp ose of generating feeling and minimizing anhedonia (Nock & Cha, 2009). Prior to engaging in self injury, these individuals often report feelings of numbness or disassociation, which is alleviated by engaging in the behavior (Hilt, Cha, & Nolen Hoeksema, 20 08). Dissociation occurs when an individual feels overwhelmed by an external stressor and begins to disconnect and disengage from reality (Low, Jones, MacLeod, Power, & Duggan, 2000). Although the dissociation function is often adaptive during times of sev ere trauma, many individuals continue to dissociate throughout their lives, which may prompt NSSI behavior (Low et al., 2000). Individuals engaging in NSSI for automatic positive reinforcement often have a higher pain threshold and may be more likely to e ngage in NSSI to feel more alive (Hilt, Cha, & Nolen Hoeksema, 2008; Polk & Liss, 2009). Individuals engaging in NSSI for the purpose of APR are also more likely to experience symptoms of major depressive disorder (MDD), Borderline Personality Disorder (BP D), or posttraumatic stress disorder (PTSD) (Klonsky, 2007; Nock & Prinstein, 2005). As a result of its strong association with psychopathology, APR is the second most endorsed function within the clinical population (Lloyd Richardson, Nock, & Prinstein, 2 009).
46 The third function proposed by the FFM model is social positive reinforcement (SPR), in which individuals engage in NSSI to obtain attention or environmental resources (Nock & Cha, 2009). This may also be used as a social environment by eliciting a response from others (Prinstein et al., 2009). SPR is endorsed almost as often as the ANR function by community adolescents, who may use self Richardson, Nock, & Prinstein, 2009, p. 35). Although many adolescents do not initiate NSSI behaviors for social positive reinforcement reasons, they may find that when others discover their injuries they receive access to the care they need (Lloyd Richardson, Nock, & Prinstein, 2009). Finally, social negative reinforcement (SNR) functions to remove some interpersonal demand or task (Nock & Cha, 2009). Individuals who endorse this function of self Richardson, Nock, & Prinstein, 2009, p. 35). Among college students, NSSI may be used as a way to escape from the increasing demands of adult life and acade mic responsibilities (Lloyd Richardson, Nock, & Prinstein, 2009). While the ANR function has received the most support from both community and clinical populations, the other three functions have received significant support as well. Research by Lloyd Ric hardson, Nock, & Prinstein (2009) has suggested that while hospitalized samples are more likely to report automatic functions of NSSI, samples from the community are as likely to report
47 social functions as they are to report automatic ones. This theoretica l functional framework has gained considerable empirical support, through both self report studies and psychological and behavioral studies (Glenn & Klonsky, 2009; Nock, 2009; Nock & Cha, 2009; Nock & Prinstein, 2004; Nock, Teper, & Hollander, 2007). Emoti onal Regulation Theory The emotional regulation theory, which is most similar to the ANR function of the four function model, also has garnered significant support to explain the functions of NSSI (Prinstein, Guerry, Browne, & Rancourt, 2009). Individuals engaging in self injury for the purpose of regulating emotions do so as a way to remove unpleasant affective states (Hilt, Cha, & Nolen Hoeksema, 2008). The emotional regulation theory suggests that early unstable environments decrease ility to learn to cope with difficult situations (Klonsky, 2007). As a result, these individuals are less able to manage their own affect and may use self injury as a maladaptive means of coping or to alleviate acute emotional distress of affective arousal (Klonsky, 2007). The emotional regulation model posits that NSSI is often related to poor emotional regulation skills (Hilt, Cha, & Nolen Hoeksema, 2008). Thus, these individuals use NSSI to mediate or avoid unwanted emotional states. Often, these indivi duals experience heightened levels of distress that they feel unable to control through other coping mechanisms. In support of this model, Nock & Mendes (2008) found that adolescents who engage in NSSI generally exhibit higher levels of physiological react ivity to stress, decreased problem solving skills and a reduced ability to tolerate stress.
48 Anxiety Reduction Model The anxiety reduction model represents an extension of the emotional regulation model and offers another functional explanation for NSSI. T his model proposes that individuals use NSSI when they experience tension or anxiety that reaches an intolerable level (Ross & Heath, 2003). NSSI often allows these individuals to experience immediate relief from anxiety following the act (Favazza, 199 8). Following NSSI, these individuals are able to return to a normal emotional state (Ross & Heath, 2003). Ross & Heath (2003) argue that individuals engaging in NSSI for anxiety regulation are more likely to experience higher levels of generalized anxiety (i .e. trait anxiety) and more feelings of anxiety immediately preceding self injury (i.e. state anxiety). These individuals often lack adaptive coping mechanisms, and use NSSI as a way to regulate their affect. Hostility Model The hostility model, also deve loped by Ross and Heath (2003) is an extension of the anxiety reduction model, wherein individuals use NSSI as a cathartic action to reduce hostility in times of stress. This theory is more often associated with male self injurers who feel it is unacceptab le to outwardly express hostility. This model suggests that individuals are unable to appropriately express feelings of hostility, causing an increase in feelings of tension, and the direction of anger on an acceptable source ( i.e., the self) (Ross & Heath 2003). As in the anxiety reduction model, it is expected that individuals who self injure report greater generalized feelings of hostility (i.e. trait hostility) as well as greater feelings of hostility before engaging in NSSI (i.e. state hostility) (Ros s & Heath,
49 2003). Ross & Heath hypothesize that some individuals experience both anxiety and hostility before NSSI, while others experience only anxiety or hostility prior to the act (Ross & Heath, 2003). Experiential Avoidance Theory The experiential avo idance theory posits that NSSI is a way for individuals to reduce or control high levels of emotional arousal (Chapman, Gratz, & Brown, 2006). Experiential avoidance includes any behavior that serves to avoid or escape from unwanted internal conditions an d the external conditions that surround them (Hayes, Wilson, Gifford, Follette, & Strosahl, 1996). The class of behaviors defined as experiential avoidance are maintained through negative reinforcement, and include other behaviors besides NSSI, such as avo idant coping styles, thought suppression, drug or alcohol use, or avoidance of certain objects and situations (Chapman, Gratz, & Brown, 2006) (See Fi gure 1) Social Learning, Modeling and Reinforcement Perspectives While there are varying degrees of supp ort for each model described above, it is also important for researchers to consider the interpersonal perspectives of NSSI, especially when examining adolescent and young adult populations. Research has suggested that interpersonal relationships may influ ence the development of NSSI (Bureau, Martin, Freynet, Poirier, Lafontaine, & Cloutier, 2010), and that interpersonal stressors often immediately precede NSSI behaviors in adolescents (Prinstein et al., 2009). NSSI Contagion A contagion effect also has be en noted among self injurers, wherein adolescents normalize NSSI behavior within their group of friends and use it as a
50 way to create or solidify friendships (Heilbron & Prinstein, 2008). Research indicates that one of the most reliable predictors of wheth er an adolescent will attitudes or beliefs (Heilbron & Prinstein, 2008). A causal link has also been suggested between affiliations with high risk peers and the likelihood of adolescents engaging in high risk behavior (Prinstein, Boergers, & Spirito, 2001). For example, adolescents engaging in a high frequency of physical aggression or substance abuse tend to have larger proportions of friends engaging in these same behaviors ( Prinstein et al., 2001). The association between peers and high risk behavior may be a combination of selection effects, where adolescents prefer peers who engage in similar types of behavior, as well as socialization effects, wherein adolescents implicit ly or explicitly influence others to engage in similar behaviors (Prinstein et al., 2001). Accumulating evidence suggests that NSSI is a behavior strongly subject to peer influence (Prinstein et al., 2009). In fact, one study found that adolescents had th oughts of NSSI after being encouraged by others approximately 1.7% 3.8% of the time (Nock, Prinstein, & Sterba, 2009). Adolescents and young adults may also engage in self injury in each ieberman, Toste, & Heath 2009). This form of contagion is thought to be more prevalent among males (Lloyd Richardson et al., 2007). The contagion effect makes it even more difficult to effectively treat NSSI, because there is concern that treating the beh avior within certain therapeutic group settings may increase the
51 rates of NSSI, as individuals share their stories of self injury in detail, or in a positive light (Kibler, 2009). Social Learning Theories Social psychological theories have determined that individuals may imitate the behavior of peers to maintain or elevate their self image (Prinstein et al., 2009). As a result, if adolescents perceive NSSI to be a behavior of high status, it is more likely that they will engage in the behavior as a way to seek positive reinforcement from their environment (Prinstein et al., 2009). Social learning theory dictates that learning about and observing the NSSI behaviors of others injuring (Nock, 2009). Inte restingly, a correlation has been found between the increasing prevalence of NSSI and the increase in references to NSSI in movies, songs, print media and the internet in the past ten years (Whitlock, Purington, & Gershkovich, 2009). Social learning effec ts may also be heightened when adolescents perceive low support from peers, increasing the possibility that they will imitate risky behaviors of their friends (Prinstein et al., 2001). Behavioral theorists also suggest that positive reinforcement and socia l modeling may influence self injuring behavior (Prinstein et al., 2009). Granic and Dishion (2003) determined that the manner in which adolescents talk about deviant acts may serve to reinforce externalizing and aggressive behavior over time. Their work a lso noted that deviant adolescents were more likely to engage in verbal positive reinforcements towards peers when discussing antisocial acts than prosocial acts (Granic & Dishion, 2003).
52 The social functions of NSSI may also serve as a form of communicat ion to others about thoughts and feelings. Nock (2009) notes that when higher order processes such as language fail, individuals often feel compelled to use more primitive means to convey messages. Thus, it can be theorized that these individuals do not fe el successful communicating their message, and feel that they must resort to increasingly intense modes of communication until their message is recognized (Nock, 2009). Evidence from the literature suggests that adolescents engaging in NSSI do not differ f rom other adolescents on general intelligence, problem solving or design fluency, but they do have significantly poorer verbal fluency than non self injurers, and report being less aware of their own emotions and having difficulty expressing their emotion s (Nock, 2009). It has also been suggested that emotional dysregulation stems from a lack of social support, both in family and peer relationships (Adrian, Zeman, Erdley, Lisa, & Sim, 2011). In their 2011 study, Adrian and colleagues found that family an d peer interpersonal problems have negative and independent effects on an for developing NSSI behaviors. Further, their model gave support to the notion that poor familial r elationships often predicted poor outcomes in peer relationships (Adrian et al., 2011). Treatment Approaches While many elements of NSSI have been examined in the research, effective treatment options are still relatively limited for this population. No nsuicidal self injury has typically been regarded as a treatment resistant behavior that has historically created great frustration among the medical and
53 psychological communities (Muehlenkamp, 2006). Although it is considered a behavior distinct from suic ide, self injury can often lead to serious injury and accidental death (Muehlenkamp, 2006). It is important to note that even though NSSI is considered a separate behavior from suicide, a correlation still exists between the two (Newman, 2009). Thus, NSSI should be taken seriously, and examined in conjunction with a thorough assessment for suicide. Individuals with NSSI are commonly treated in an inpatient hospital setting, and usually experience minimal success (Muehlenkamp, 2006). Because of the signific ant expense and minimal effectiveness of inpatient hospitalization, it is important to find empirically supported treatments that work on an outpatient basis with NSSI. Currently, treatments utilizing a cognitive behavioral therapy (CBT) approach have demo nstrated the most effectiveness when treating NSSI (Muehlenkamp, 2006). These approaches help individuals manage the emotions that lead to repeated episodes of self injury. CBT treatments aim to help individuals successfully regulate their emotions by unde rstanding the functionality of all emotions, rather than simply using thought suppression strategies to deal with certain emotions. Research has found that suppressing negative or unwanted emotions or thoughts can ultimately create negative consequences fo r individuals engaging in NSSI (Gratz, 2007) by increasing distressing thoughts and emotions. Individuals may engage NSSI as a way to reduce the emotional Wegner, & Nock, 2007)
54 Behavioral interventions also help adolescents find alternative means of coping as well as adaptive ways to understand and manage their stress and emotions (Lieberman, Toste, & Heath 2009). Behavioral strategies to help reduce depression and increase s elf 2009). A cognitive framework of NSSI asserts that individuals engaging in self injurious behaviors harbor maladaptive beliefs a nd flawed coping mechanisms. These behaviors are maintained by a variety of environmental and internal consequences that prove to be reinforcing to the individual (Newman, 2009). It is important for cognitive treatment to focus on these maladaptive beliefs and allow the client to modify these beliefs into something more productive and adaptive (Newman, 2009). Namely, individuals engaging in NSSI usually have a variety of automatic negative thoughts about themselves or others (Lieberman, Heath, & Toste, 2009 ). These negative cognitions often include a) the behavior is necessary and acceptable, b) the individual is disgusting and deserves to be punished, c) the action of self injury is the only way to reduce unpleasant feelings to solve crises, and d) the acti on is needed to help communicate feelings (Walsh & Rosen, 1985). Treatments targeting the emotional dysregulation that is associated with self injury may also be useful. However, treatments aimed at helping individuals better regulate their emotions should focus on learning other more adaptive ways to experience emotions, rather than suppressing or attempting to control them (Gratz, 2007).
55 Problem solving therapy (PST) and dialectical behavior therapy (DBT) are two cognitive behavioral approaches that have been identified as having some effectiveness with self injuring individuals (Muehlenkamp, 2006). While both treatments are short term and focus directly on NSSI and its associated skill deficits, DBT is considered an intense treatment paradigm that is oft en difficult to utilize (Muehlenkamp, 2006; Slee, Arensman, Garnefski, & Spinhoven, 2007). However, both of these models are widely used to treat NSSI, and can be adapted for a variety of settings or populations. PST is based on the research supported no tion that individuals who self injure generally present with specific problem solving deficits (Townsend et al., 2001). As a result, this treatment focuses on teaching problem solving skills and general coping strategies, as well as helping individuals ide ntify and solve problems in their lives (Muehlenkamp, 2006). Problem solving treatment is considered a pragmatic approach for helping people with emotional regulation problems, suicidal behavior and NSSI (Townsend et al., 2001). The primary goal of PST is to help patients develop skills to problem solve current difficulties, with a secondary aim of teaching them overall strategies which will serve to help them better solve problems in the future (Townsend et al., 2001). During therapy, individuals are taug ht the different steps in effective problem solving, including identification and operationalization of the problem, generating possible solutions, choosing a solution and evaluating its results (Muehlenkamp, 2006; Townsend et al., 2001). More flexible thi nking styles are also encouraged in PST, as it has been shown that individuals engaging in NSSI
56 often have rigid, inflexible ways of thinking (Muehlenkamp, 2006). The formation of a strong therapeutic alliance is also stressed in PST (Muehlenkamp, 2006). R esearch on the effectiveness of PST has been limited, and the results are often inconclusive because of the small sample sizes used and limited studies conducted (Townsend et al., 2001). One study conducted by Hawton et al., (2000), found that PST was more effective in decreasing the repetition of NSSI (15.5% repetition for PST and 19.2% for control patients), however, these results were not statistically significant, which may be a result of the small sample size used in this study. Therefore, more systema tic research in this area is needed with larger samples to determine treatment efficacy (Muehlenkamp, 2006). DBT is a treatment originally developed for individuals with borderline personality disorder, but is now often used with individuals engaging in N SSI. This treatment has recently emerged as one of the most effective methods of treating BPD, and has been shown in randomized clinical trials to reduce the overall incidence of self injurious behaviors (Nock, Teper & Hollander, 2007). DBT is generally c onducted by utilizing a combination of individual psychotherapy as well as group therapy and skills training (Gratz, 2007). It incorporates aspects of mindfulness, behaviorism and dialectical philosophy (Lynch & Cozza, 2009) and includes a significant elem ent of self acceptance and chang e (Robins & Chapman, 2004). Thus, DBT seeks to balance the idea of accepting individuals where they are, while gently requesting and implementing the idea of change (Trupin, Stewart, Beach, & Boesky, 2002). Individual therap y focuses on achieving the primary targets of DBT, which are 1) decreasing life
57 threatening behaviors, 2) decreasing therapy interfering behaviors, 3) decreasing quality of life interfering behaviors, and 4) increasing behavioral skills (Nock, Teper & Holl ander, 2007). Group based skills training focuses primarily on 1) mindfulness, 2) emotional regulation, 3) interpersonal effectiveness, 4) distress 2007). The last group skill is specific to adolescent DBT and includes suggestions for navigating family interactions (Nock, Teper, & Hollander, 2007). DBT also can help individuals become more aware and manage the behaviors associated with their emotions, allowing them to more succes sfully utilize adaptive coping strategies (Gratz, 2007). Individuals are taught to identify an emotional response and the associated physiological, subjective and behavioral aspects of the emotion and event (Gratz, 2007). The functions of emotions are als o explained, which allows self injurers to gain increased understanding and subsequent acceptance of negative emotions (Gratz, 2007). Distress tolerance skills are also taught to increase emotional regulation and dividuals are taught to better manage the behaviors associated with emotions by considering the short and long term consequences of behaviors. Self soothing techniques are demonstrated to help self injurers more adaptively cope with negative emotions (Grat z, 2007). The four functions of DBT treatment include helping the individual develop adaptive skills, addressing obstacles to using new skills, allowing the individual to generalize new skills to their daily life, and keeping therapists motivated and skil led, while minimizing burnout (Robins & Chapman, 2004). DBT can be
58 adapted for adolescents and young adults by shortening the duration of treatment, reducing the number and complexity of skills taught, including other family members and engaging caregivers in skill building groups (Robins & Chapman, 2004). Limitations of Current Research Literature on NSSI has expanded considerably within the last decade as researchers continue to gain understanding of the different components of this complex behavior. How ever, significant limitations and inconsistent findings still exist in the research in several important areas of self injury. Defining self injury continues to be one of the most discrepant and difficult issues su rrounding NSSI l self has recently become a standard term for the behavior (ISSS, 2007), there remains a multitude of different names for self injury, which often causes disagreements in the literature. Terms such as parasuicide, self injurious behavior, self mut ilation, deliberate self harm, self carving, and self cutting generally encompass the behaviors found in NSSI (Nixon & Heath, 2009). Current research reflects the idea that NSSI exists within a continuum of deliberate self harming behaviors. That is, while NSSI is considered a self harm behavior, there are many other self harming behaviors (either direct or indirect) that cannot be included within the NSSI spectrum (Nixon & Heath, 2009). The true prevalence of self injurious behavior is also difficult to ascertain, both within community and clinical samples. Community based estimates range from .75% to 46% for adolescents (Carlson et al., 2005; Heilbron & Prinstein, 2008; Latzman, Gratz, Young, Heiden, Damon, & Hight, 2010; Levesque,
59 Lafontaine, Bureau, Cl outier, & Dandurand, 2010; Nock, 2009; Ross & Heath, 2002; Weismoore & Esposito Smythers, 2010) and 21 to 80% in clinical samples (Glenn & Klonsky, 2009; Heilbron & Prinstein, 2008; Hilt et al., 2008). Prevalence estimates for young adults have similar dis crepancies, ranging from approximately 14 to 38% of college age students (Gollust, Eisenberg, Golberstein, 2008; Laye Gindhu & Schonert Reichl, 2005). These wide ranges reflect disagreement over the actual rates of self injury as well as the specificity of the term. Prevalence rates likely differ due to variability in the terms and definitions used by researchers. For example, some surveys ask students if they have ever rates th 2008). Other studies utilize a definition of self injury that includes high risk behaviors such as self poisoning, jumping from heights, and drug overdoses (De Leo & Heller, 2004; Hawton et al., 1998) Nixon & Heath, 2009). Most of the recent literature concurs that while high risk behaviors pose potential risks, they should not be included in the NSSI paradigm (Nixon & Heath, 2009). Including these behaviors in the definition of self inj ury usually yield higher prevalence ratings than studies excluding these behaviors from the definitions. A study conducted by Lloyd Richardson et al., (2007) found a 12 month prevalence rate of 46.5% of self injury of adolescents in a community sample. The measure used in this study included more innocuous behaviors such as picking at an area of skin until it bleeds as a type of self injury. However, when this item was removed,
60 the prevalence rate dropped to 27.7% (Lloyd Richardson et al., 2007). The use of broad definitions in studies is especially problematic if the questionnaires are retrospective and encompass all of childhood as some innocuous behaviors have apture behaviors that are not uncommon for children. Debate over the methods and functions of NSSI has also caused problems when considering various research studies as researchers may limit or broaden questionnaire items for behaviors that reflect their particular model preference. For example, while many studies investigate all forms of self injury, other studies limit the definition of self methods such as burning, self hitting and erasing skin (Heath, Schaub, Holl y, & Nixon, 2009). Furthermore, European studies often use the Child and Adolescent Self Harm in Europe (CASE) group definition, whereas self harm is defined as : An act with a non fatal outcome in which an individual deliberately did one or more of the fo llowing: Initiated behavior (for example, self cutting, jumping from a height), which they intended to cause self harm; Ingested a substance in excess of the prescribed or generally recognized therapeutic dose; Ingested a recreational or illicit drug that was an act that the person regarded as self harm; Ingested a non ingestible substance or object (Hawton, Rodham, Evans, & Weatherall, 2002, p. 1208). This definition further complicates the study of NSSI because several of these risky behaviors (e.g. jumpi ng from a height, ingesting a substance) are not considered self injury in most American scholarship. Another significant difference between American and European research is the idea of suicidal intent. While the large majority of American researchers con ceptualize NSSI as a behavior without suicidal intent, many British researchers include behaviors that
61 indicate suicidal intent in the NSSI spectrum (Fliege, Lee, Grimm, & Klapp, 2009). This limits the utility of comparing data cross nationally, as prevale nce rates will differ based on definitional issues (Fliege et al., 2009). To date, only one study has compared international rates of NSSI using cross nationally validated assessment scales (Plener et al., 2009). This study, conducted by Plener and collea gues (2009) assessed the rates of NSSI in a community sample of students from the United States and Germany. The data derived from this study suggested no significant differences between the prevalence rates found in Germany (25.6%) and those found in the United States (23.2%) (Plener et al., 2009). Research methodology may also affect prevalence rates of NSSI. It has been shown that interviews are likely to yield lower prevalence rates than anonymous surveys because of social desirability bias (Heath, Sch aub, Holly, & Nixon, 2009). This may be because individuals are more comfortable disclosing self injury anonymously, or because interviewers are more discriminating in what they consider self injury when talking to participants (Heath, Schaub, Holly, & Nix on, 2009). The disagreement over prevalence rates also makes it difficult to determine how widespread the behavior is among adolescents and young adults. Some of the disagreements over prevalence rates relate to the methods of data collection, as well as concerns over accurate self disclosure. It is also important for researchers to appropriately distinguish between incidence and prevalence rates of self injury. While prevalence refers to the proportion of the sample that is
62 currently engaging in the behav ior, incidence refers to the proportion that has exhibited the behavior anytime during a specific time period (Cheng, Mallinckrodt, Soet, & Sevig, 2010). In a survey, questions that begin with the essentially assess the incidence of NSSI. As a result, surveys attempting to assess lifetime incidence rates of NSSI consistently obtain higher rates than surveys assessing prevalence rates of NSSI (Cheng, Mallinckrodt, Soet, & Sevig, 20 10). Many studies examining community populations of self injurers rely on self report measures given at a school. While this method offers the best way to survey large numbers of adolescents, it may be difficult to determine if all students who self inju re are identifying themselves. These students may have concerns that their behaviors will be disclosed to school staff or their parents; even if they are assured their information will remain confidential. While many researchers disagree about the exact pr evalence of adolescents and young adults engaging in NSSI, the existing estimates of this behavior are alarming. While prevalence rates of NSSI present a significant challenge for researchers, there are also significant discrepancies regarding the etiolog y of self injury within the NSSI literature. Although many researchers believe that NSSI is an over determined behavior, caused by multiple psychological vulnerabilities and environmental deficits (Prinstein et al., 2009), there are significant disagreemen ts over the correlation between events such as childhood maltreatment and NSSI. Much of the research conducted in these areas has
63 been unsystematic or utilized small case study samples (Klonsky & Glenn, 2009). Early psychoanalytically focused research in t his area strongly linked NSSI to early childhood maltreatment, specifically sexual abuse. For example, individuals to reconnect with their bodies during periods of dissociati on. Conversely, more recent literature has failed to find significant correlations between early sexual abuse and subsequent NSSI (Klonsky & Moyer, 2008). Other recent studies have linked sexual abu se with NSSI in a multiple medi ational model, which uses PTSD, depressive symptoms and psychological dysregulation as mediating variables in the relationship between childhood maltreatment and NSSI (Shenk, Noll, & Cassarly, 2010). Some of the disagreements over the prevalence and etiology in the literature may also contribute to obstacles regarding intervention and prevention of NSSI. It may be difficult to properly assess and plan interventions for NSSI if the behavior itself has not been appropriately operationalized and understood. Currently, the DSM IV TR (A PA, 2000) identifies NSSI only in the context of Borderline Personality Disorder. This presents challenges for mental health professionals assessing and treating individuals engaging in these behaviors. DSM V Proposed R evisions The Diagnostic and Statisti cal Manual for Mental Disorders, Fourth Edition, Text Revision, (DSM IV TR) (APA, 2000) does not conceptualize NSSI as a separate psychiatric disorder. Currently, NSSI is listed only once in the DSM IV TR, as a symptom classification for Borderline Persona lity Disorder (BPD) (Glenn & Klonsky, 2010). There has been significant controversy
64 regarding the addition of NSSI as a clinically distinct syndrome (Muehlenkamp, 2005). The earliest argument for the addition of self injury to the DSM was made in 1967 by G raff and Malin, and then again in 1969 by Pao (Muehlenkamp, harm control disorders based on their belief that NSSI stemmed from the inability to resist an urg e to self injure (Lynam et al., 2011). Since that time, many researchers have attempted to describe a syndrome of self injury, however, these proposed syndromes were often difficult to substantiate due to the lack of research in the area (Muehlenkamp, 2005 ). One of the strongest arguments against including NSSI as a separate disorder in the DSM V comes from the lack of large scale, empirically sound research (Muehlenkamp, 2005). However, others argue that this lack of research is an indication that the DSM V should include NSSI as a way to increase consistent understanding of NSSI and make researchers more likely to use rigorous research methods to study the behavior (Muehlenkamp, 2005). Another reason researchers have proposed the addition of a separate N SSI disorder is the number of self injuring individuals who do not fall into any of the current Axis I or Axis II diagnostic categories. Muehlenkamp (2005) argues that the impulse disorder, not otherwise specified (NOS) is currently the best fit for indivi duals engaging in NSSI, although it is far from sufficient. Currently, the only guideline for determining diagnostic validity is that of Feighner et al., (1972) who describes five criteria for the inclusion of a new diagnostic category (Muehlenkamp, 2005) These criteria include; clinical
65 description, laboratory studies, delimitation from other disorders ( i.e exclusionary criteria) follow up studies, and family studies to determine heritability factors (Muehlenkamp, 2005). Using this model to measure the necessity of including NSSI as a distinct clinical disorder shows that NSSI Specifically, clinical descri ptions of NSSI are well defined, and current research has found some physiological associations with NSSI, however, more longitudinal and follow up studies are needed, as well as family studies to determine heritability factors (Muehlenkamp, 2005). As a r esult of the increasing prevalence and negative psychological implications of NSSI, two DSM V workgroups (i.e. Child and Adolescent Disorders and Mood Disorders) have been created that propose classifying NSSI as a distinct DSM diagnosis (Glenn & Klonsky, 2010). According to the American Psychiatric Association DSM V Development website, the proposed diagnostic classification for NSSI will be : A. In the last year, the individual has, on 5 or more days, engaged in intentional self inflicted damage to the surfac e of his or her body, of a sort likely to induce bleeding or bruising or pain (e.g., cutting, burning, stabbing, hitting, excessive rubbing), for purposes not socially sanctioned (e.g., body piercing, tattooing, etc.), but performed with the expectation th at the injury will lead to only minor or moderate physical harm. The absence of suicidal intent is either reported by the patient or can be inferred by frequent use of methods that the patient knows, by experience, not to have lethal potential. (When uncer tain, code with NOS 2.) The behavior is not of a common and trivial nature, such as picking at a wound or nail biting. B. The intentional injury is associated with at least 2 of the following:
66 1. Negative feelings or thoughts, such as depression, anxiety, tensi on, anger, generalized distress, or self criticism, occurring in the period immediately prior to the self injurious act. 2. Prior to engaging in the act, a period of preoccupation with the intended behavior that is difficult to resist. 3. The urge to engage in s elf injury occurs frequently, although it might not be acted upon. 4. The activity is engaged in with a purpose; this might be relief from a negative feeling/cognitive state or interpersonal difficulty or induction of a positive feeling state. The patient ant icipates these will occur either during or immediately following the self injury. C. The behavior and its consequences cause clinically significant distress or impairment in interpersonal, academic, or other important areas of functioning. D. The behavior does not occur exclusively during states of psychosis, delirium, or intoxication. In individuals with a developmental disorder, the behavior is not part of a pattern of repetitive stereotopies. The behavior cannot be accounted for by another mental or medical disorder (i.e., psychotic disorder, pervasive developmental disorder, mental retardation, Lesch Nyhan Syndrome) (APA, 2010, p. 4). The proposed addition of this diagnostic category will likely reduce problems associated with the lack of consensus among m ental health professionals regarding the definition of NSSI (Wilkinson & Goodyer, 2011). It may also improve treatment options for individuals with NSSI, as well as streamline and improve research efforts in the area of self injury (Wilkinson & Goodyer, 20 11). If the diagnostic category is adopted, subtyping is also suggested to further clarify definitional issues surrounding NSSI. Specifically, a Non Suicidal Self Injury, Not Otherwise Specified (NOS) Type 1: Subthreshold was proposed to include individual injured himself or herself fewer than five times in the past month. This can
67 include individuals who, despite a low frequency of the behavior, frequently think Jacobson, 2009, p. 4). The second proposed subtype was identified as Non Suicidal Self Injury Disorder, Not Otherwise Specified (NOS), Type 2, Intent Uncertain. This subtype is meant to hat in addition The Proposal to the DSM V Childhood Disorder and Mood Disorder Work Groups to Include Non Suicidal Self Injury (NSSI) as a DSM V Disorder was published by Shaffe r and Jacobson (2009) as a discussion of the benefits of adding NSSI to the list of DSM V modifications. The proposal was suggested, problems of a public health and clinical nature that arise because of a lack of clarity about its meaning and significance that we feel could be remediated by discrepancies in NSSI research and treatment represented by the broa d notion of diagnostic criterion for NSSI creates significant difficulty for clinicians, researchers and individuals engaging in NSSI. The concept of subclinical levels of se lf injury has also added to some of the definitional confusion surrounding NSSI. While it is important to distinguish between clinical levels of NSSI and more minor types of self harm (e.g. skin picking, scratching, interfering with wound healing), it shou ld be noted that these behaviors are still cause for concern. Many of these actions are also referred to
68 as body focused repetitive behaviors, and are often associated with increased anxiety, depression, obsessive compulsive spectrum disorder (OCD) and bod y dysmorphic disorder (BDD) (Croyle & Waltz, 2007). A study conducted by Croyle & Waltz (2007) attempted to examine self injury as a behavior lying on a continuum. They found that 68% of their total sample engaged in at least one type of mildly self injur ious behavior in their lifetime, including skin picking, nail biting, and wound interference. While many of these students did not endorse any negative affect associated with the behavior, as many as 20% of individuals in the low self harm group endorsed s ome levels of negative affect associated with the behavior. Additionally, thirty five percent of participants fell in the moderately injurious self harm group. These individuals had higher levels of somatic symptoms, impulsivity, OCD characteristics, disor dered eating behaviors, higher levels of shame and were more likely to have a history of abuse (Croyle & Waltz, 2007). This information suggests that NSSI lies on a continuum, and there is a positive correlation between NSSI, psychopathology, and overall n egative outcomes (Croyle & Waltz, 2007). The definition is further complicated by the continuing belief that NSSI is a were the result of cutting or piercing (National Center for Inj ury Prevention and Control, 2012 ). The lack of differentiation between suicidal attempts and NSSI may also contribute to an overestimation of reported suicide attempts in adolescents. For example, in a study conducted by Kumar, Pepe, & Steer (2004),
69 approximately 80% of the sample noted that their self injurious behavior had been misconstrued as a suicidal attempt. A study conducted by Cheng, Mallinckrodt, Soet, & Sevig (2010) used the Counseling Center Assessment of Psychological Symptoms, 62, (CCA PS 62) to determine if there were any psychological symptoms that strongly correlated with NSSI, and could be used as an NSSI warning sign for clinicians when conducting intake sessions with new counseling center patients. Archival data was obtained from 2 ,184 college students using the CCAPS 62, a self injury screening item, and a 16 item self injury measure. While no specific subscales were found to be significantly associated with NSSI, 11 specific items on the CCAPS 62 were found to moderately predict t he presence of NSSI (Cheng et al., 2010). These items were found to tap symptoms of depression, self loathing, anxiety, anger, and feelings of losing control (Cheng et al., 2010). While the aforementioned study is similar to the present study there are s everal important differences that should be noted. First, Cheng et al., (2010) used a threshold of five total NSSI incidents to determine NSSI grouping to self injurious incident. However, there was no specific time frame in which these five incidents must have occurred. The current study will utilize the criterion of five incidents within the past year to determine clinical levels of NSSI; a threshold that is more consist ent with the proposed DSM V criteria. Additionally, Cheng et al., (2010) sought to determine if any of the items on the CCAPS 62 could be used as a screening question for NSSI behavior. To accomplish this, the eight
70 subscale items of the CCAPS 62 were exam ined in relation to two groups; individuals who engaged in NSSI (with five or more incidents) and individuals who did not engage in NSSI. As a result, individuals engaging in subclinical fected the results of the study. The current study examine s NSSI in a way that is different from prior studies because it separates individuals who engaged in NSSI in the past from those who currently self injure and investigates the differences between th ese two groups. This distinction is important because there may be different psychosocial outcomes for those who continue to engage in the behavior, versus the individuals who no longer self injure. Problem Statement As indicated by the literature review, mental health issues among college students have become an increasingly more complex and significant problem within the past decade (Gallagher, 2009) The increase in the number of college students, combined with the vulnerable developmental period for em ergent mental health issues, make college students an important age group for research and intervention. Additionally, research has indicated that college students are using NSSI more often as a coping mechanism during times of stress (Gallagher, 2009). Sp ecifically, prevalence rates for adolescents and young adults within the community population appear to be increasing at an alarming rate (Brausch & Gutierrez, 2010). This shift in population corresponds with an increase in research targeted at nonclinical samples with less known psychopathology. Recent estimates indicate that somewhere between 14 to 36% of college age students have engaged in NSSI behaviors at some point in their lives, (Gollust,
71 Eisenberg, & Golberstein, 2008; Laye Gindhu & Schonert Reich l, 2005), with approximately 7% self injuring within the past year (Gollust, Eisenberg, & Golberstein, 2008). NSSI has been found to be a known risk factor for suicidal behavior, although it is recognized as being distinct from suicidal thoughts and actio ns life, the functions of NSSI are complicated and not well understood. Several theoretical models attempt to determine the functionality of NSSI, with varying degrees of res earch and support. Currently, the four function model (FFM) represents the most comprehensive theoretical representation of the functions of self injury (Nock & Cha, 2009). This model proposes that the functions of self injury fall along two dichotomous ca tegories: reinforcement that is either positive or negative, and contingencies that are either automatic or social (Nock & Cha, 2009). Other theories, such as the emotional regulation theory, (Prinstein, Guerry, Browne, & Rancourt, 2009) anxi ety reduction model, hostility model (Ross & Heath, 2003) and the experiential avoidance theory (Chapman, Gratz, & Brown, 2006) also posit explanations for the functions of self injury. Although there has been a substantial increase in NSSI scholarship, significant gaps in the literature s till exist. Despite the increase in prevalence rates, NSSI still does not exist as a separate diagnostic category in the DSM. This creates diagnostic and treatment complications, as NSSI has only been distinguished as a symptom of Borde rline Personality Disorder (BPD), despite the fact that the majority of self injurers do not meet diagnostic criteria for BPD.
72 As a result, NSSI has been proposed as a separate diagnostic category for the updated DSM V (American Psychiatric Association, 20 10) which will publish on May 18 th 2013 If included, t his addition will not only distinguish NSSI as a distinct disorder, but the operational definition will also help alleviate discrepancies in prevalence rates. The large discrepancies within prevalence rates present one of the most significant gaps in current NSSI scholarship. It has been difficult for researchers to accurately determine the true rate of self injury among the young adult population, primarily because of the differences in NSSI measures and definitions. Including NSSI in the DSM V will significantly decrease the confusion surrounding the definition of self injury and streamline empirical studies. Partly because of the conflicting definitions of NSSI, research has not been conducted to de termine whether significant differences exist among individuals who do not self injure, those who have engaged in NSSI at some point in their lives, and those who currently self injure. Understanding the psychosocial outcomes for these three groups can hav e substantial clinical implications. If individuals currently engaging in NSSI are found to have significantly worse psychosocial outcomes than the other two groups, it can lend further support to the notion that NSSI should be a separate, diagnostic categ ory within the DSM V. Additionally, comparing the psychosocial outcomes of th ose with no NSSI behaviors, with individuals who have engaged in NSSI either currently or in the past, can help determine the necessity of treatment aimed specifically at self inj ury. The findings of this proposed study may also help
73 develop additional treatment options for individuals engaging in NSSI behaviors. To address identified gaps in the literature on NSSI, the primary research question of this study wa s: 1. Is there a signif icant difference in levels of symptomology as measured by the CCAPS 62 for students who currently engage in NSSI, those who have engaged in NSSI in the past, and those who have never engaged in NSSI? Table 1 1. Four Function Model of NSSI P ositive Reinforcement Negative Reinforcement Automatic (A) APR Creates a desirable physiological state (a means of feeling generation) ANR Reduces tension or other affective state(s) Social (S) SPR Provides attention from others SNR Offers escape from interpersonal tasks or demands
74 Figure 1 1. The experiential avoidance model (EAM) of deliberate self harm *In this model, the term DSH (deliberate self harm) is used to describe self injury* (Chapman, Gratz, & Brown, 2006, p. 373) (Author perm ission was obtained prior to using this diagram)
75 CHAPTER 2 METHODS AND PROCEDUR ES Participants A random sampling of 5,000 undergraduate students at the University of Florida was recruited through an email listserv provided by the Office of Institutional Research. Out of the 5,000 students that were recruited for the study, a sample size of 1,131 students compl eted the surve y, giving a 22.6 % response rate. This response rate was somewhat lower that expected as electronic response rates have been found to b e between 30 35% (Shannon & Bradshaw, 2002 Shih & Fan, 2009). According to the survey results, a total of 42 students (3.4%) had self injured in the past twelve months, and 142 students (11.4%) have self injured throughout their lives. The demographic inf ormation derived from this study was compared to the 2010 Healthy Gators Student Survey, as well as the University of Florida undergraduate population as a whole. The Healthy Gators Student Survey was used as a comparison because it is one of the most larg e scale, comprehensive surveys conducted on University of Florida students and includes questions regarding physical and mental health The majority of the demographic information from the current study was similar to both the Healthy Gators Survey as wel l as the University population, and can be found in Table 2 1 Gender was not as closely aligned with the University of Florida demographics, though it was almost identical to the gender breakdown from the Healthy Gator Survey (Table 2 2) This is not sur prising, as survey literature
76 suggests that females are generally more likely to fill out surveys than males (Porter & Umbach, 2006). A total of 386 (52.7%) of the survey respondents were between the ages of 17 20, 325 (44.3%) were between the ages of 21 29, and 22 (3%) were 30 or older. Students were also asked their current yea r in school, and the results were as follows; 158 (21.5%) were first year students, 130, (17.7%) were in their second year, 216 (29.5%) were third year students, 208 (28.4%) were i n their sexual orientation were as follows; 680 (92.8%) identified as heterosexual, 13 (1.8%) identified as gay, 8 (1.1%) identified as lesbian, 23 (3.1%) identified as b Measures Participants completed three measures as part of the online survey for this study. The first measure, (CCAPS 62) was a brief assessment to determine ychosocial functioning across several domains. Author permission was obtained prior to using the CCAPS 62. A copy of written author permission can be found in Appendix B. Next, participants were given a screening question to determine the presence of self injurious behaviors (Appendix C). Participants then filled out the NSSI Questionnaire to determine their current or past self injury (Appendix C). Finally, participants were asked to provide demographic information including gender, ethnicity, marital sta tus and sexual orientation (Appendix D).
77 The Counseling Center Assessment of Psychological Symptoms 62 (CCAPS 62) The Counseling Center Assessment of Psychological Symptoms (CCAPS 62) was developed in 2009 as a measure to assess common issues associated w ith the mental health of college aged students. It is a global assessment of mental health functioning that assesses eight different subscales, including, depression, (13 items) eating concerns, (9 items) substance use, (6 items) generalized anxiety, (9 it ems) hostility, (7 items) social anxiety, (7 items) family distress, (6 items) and academic distress (5 items). It was designed as a standardized assessment for university counseling centers working specifically with individuals ages 18 25 (Locke et al., 2 011). The assessment may be used as a screening instrument or for initial and post assessments, and takes 7 10 minutes to complete (Center for Collegiate Mental Health, 2010). It may also be used for progress monitoring purposes, as the measure is sensitiv e to change (Center for Collegiate Mental Health, 2010). The full CCAPS 62 can be found in Appendix A of this manuscript. The CCAPS 62 includes a total of 62 statements describing thoughts, feelings and experiences that college students may have. Students are asked to consider how well each statement describes them during the past two weeks. A Likert type scale is used with answer choices ranging from 1 (not at all like me) to provided, and administrators have the option of scoring the assessment by hand or through the use of the Titanium Schedule program (Center for Collegiate
78 Mental Health, 2010) When scoring by hand, items are grouped by subscale and the answers given (0 5) are added up. Any items that are reverse scored are indicated in the manual. This total is then divided by the number of items in the subscale, yielding a mean raw score for each subscale. These raw scores can be converted to percentiles using the percentile table found in the CCAPS 62 scoring manual. For the purposes of this study, the CCAPS 62 was scored using a CCAPS Excel Scoring Program provided by the Center for Collegia te Mental Health. In order to score the measures, individual answers to each question were imputed into an excel spreadsheet which converted the answers into raw scores, normalized clinical scores (NCS), and percentile scores. For the purposes of this anal ysis, NCS scores were utilized for each participant. The 2009 normative sample used for the CCAPS 62 includes college students at 52 different American institutions (N=19,247) presenting for treatment at college mental health centers. That the normative s ample was derived from a clinical setting makes it different from the present study, which used the CCAPS 62 with a random sample of college students, yielding primarily nonclinical students. Thus, it is important to note that the distribution of scores fr om the normative sample cannot be considered representative of the general population. As a result, even scores falling in the average range may still indicate participant distress in certain areas. The normative sample used for the CCAPS 62 included ages 18 63 years with a modal age of 19 and a mean age of 22.6 (SD=5.07). The population was 64.2% female, 35.4% male, and .2% transgender. 72.6% identified as White/Caucasian, 7.0% as Black/African American, 6.0% as
79 Asian/Asian American, 4.9% as Hispanic/Latin o, 3.1% as Multiracial, 2.7% preferred not to answer, 2.5% as other, .5% as Native American, and .3% as Native Hawaiian or Pacific Islander. Finally, among the normative sample, 18.1% were identified as first year students, 19.7% as sophomores, 22.1% as ju niors, 22.8% as seniors, and 14.9% as graduate students (Center for Collegiate Mental Health, 2010). The authors of the CCAPS 62 argue that many college age assessments have been developed and normed without accounting for significant features of college et al., 2011). Locke et al., (2011) argue that ignoring some of these issues detracts from the validity of measures assessing the psychological symptoms of college students. The CCAPS 62 also includes questions that measure academic (Locke et al., 2011). Two different factor analyses of the original CCAPS (which included 101 items) were completed during th e initial stages of development. The factor structure remained largely consistent over the two studies completed, with a few minor changes within the measure. The first completed factor analysis included a sample of 2,155 clients from one counseling center and revealed 20 factors, with 64.4% of the variance accounted for by those 20 factors (Locke et al., 2011). From this first study, 14 scales were identified on the measure. When the clinical team further reviewed the factors, three subscales were removed due to overlap or lack of clinical utility (Locke et al., 2011). An additional subscale was removed
80 due to its poor performance with clients. The final factor analysis in this study determined nine subscales with five additional freestanding items that did not load on any of the subscales. These five items were deemed important enough to stand alone, and included items regarding dissociative symptoms, cultural/ethnic identity, violent thoughts and history of abuse (Locke et al., 2011). The final set of item s in the first study included a minimum factor loading of .32 for each item and an item total correlation of .3 or above (Locke et al., 2011). This initial factor analysis yielded a measure with 70 items and nine different subscales. Following the factor a nalysis, the measure became the CCAPS 70. The second factor analysis was conducted on the CCAPS 70 with subsample of 11,106 clients from 52 institutions and also yielded nine subscales; however, the decision was made to eliminate the Spirituality subscale. While this subscale loaded cleanly and did not cross load with other factors, researchers decided to remove it because of its lack of clarity in defining religion and spirituality, and its relative lack of clinical utility. Thus, the second factor analysi s yielded 62 questions and 8 subscales (Locke et al., 2011). Specifically, the primary factor loading range for each subscale was; Depression=. 37 .74, Eating Concerns=. 59 .89, Substance Use=. 52 .94, Generalized anxiety= .4 .84, Hostility= .45 .92, Socia l Anxiety=. 55 .8, Family Distress=. 55 .9, and Academic Distress=. 56 .83 (Locke et al., 2011). The consistency of factor loadings across the two studies suggests a robust factor structure (Locke et al., 2011). The second factor analysis yielded the final CCAPS 62, which will be utilized in this study.
81 Overall, the CCAPS 62 demonstrates internal consistency that ranged be; Depression subscale=. 913, Eating Concerns=. 883, Substa nce Use=. 853, Generalized Anxiety=. 846, Hostility=. 863, Social Anxiety=. 823, Family Distress=. 811, and Academic Distress=. 781 (Locke et al., 2011). Additionally, two week test retest reliability was conducted with 117 undergraduate students and found to be significant (p<. 001) for all CCAPS 62 subscales (Locke et al., 2011). The correlation coefficients for each subscale can be found in Table 2 3. Convergent validity was established with several other measures examining constructs on the CCAPS 62 (L ocke et al., 2011). Pearson product moment correlations coefficients were determined between each subscale on the CCAPS 62 and another referent measure. The measures included in this analysis were the Beck Depression Inventory (BDI); Eating Attitudes Test (EAT); Alcohol Use Disorders Identification Test (AUDIT); Beck Anxiety Inventory (BAI); Social Phobia Diagnostic Questionnaire (SPDQ); Self Report Family Inventory total score (SFI); the Academic Adjustment subscale of the Student Adaption to College Quest ionnaire (A.A); the Trait Anger subscales of the State Trait Anger Expression Inventory 2; and the Marlowe Crowne Social Desirability Scale (MCSD). The Pearson product moment correlation coefficients between the CCAPS 62 subscales and referent measures can be found in Table 2 4. There is literature linking the majority of the constructs on the CCAPS 62 with increased rates of NSSI behaviors in the undergraduate population (Ballard, Bosk, & Pao, 2010; Kress & Hoffman, 2008). The relationships between NSSI
82 a nd depression, eating concerns, substance use, generalized anxiety, social anxiety, family distress, academic distress and hostility (Nock, Teper, & Hollander, 2007) suggest that the CCAPS 62 is a acceptable measure to use when studying NSSI behaviors. The CCAPS 62 is a free measure that is available to all college counseling centers. As a result, a large number of universities utilize the CCAPS 62. The information obtained by the CCAPS 62 is often used to assess the needs of students in university counseli ng centers, as well as examine trends over time. Additionally, the CCAPS (both the current version [CCAPS 62] and the prior version [CCAPS 70]) have been used in several studies seeking to examine the psychological functioning of college students. Ma (2006 ) used the CCAPS 70 with a nonclinical sample of 1,056 students in a large Midwestern public university to determine some of the most salient psychological issues facing college students. NSSI Questionnaire The NSSI questionnaire utilized in this study wa s created by the principal investigator using V childhood disorder and mood disorder work groups to include Non suicidal Self inj ury (NSSI) as a DSM The decision to create a questionnaire rather than use a pre existing measure was twofold. First, there was a lack of instruments that provided all of the specific diagnostic criteria included in the new DSM V proposed revision of NSSI. Several measures provided most of the information, but were considered incomplete. Additionally, some of these measures did not use items that were consistent with the wording
83 of the proposed criteria. A complete copy of this measure may be found in Appendix C. Secondly, there are no m based on several small studies and do not have robust reliability or validity information (Gutierrez, Osman, Barrios, & Kopper, 2001). Fur ther, the majority of the measures yield an unmanageable combination of quantitative and qualitative answers. Generally, these measures do not provide scoring information, which produces a data set that is without any criteria or set cut off points (Gutier rez, Osman, Barrios, & Kopper, 2001; Heath, Schaub, Holly, & Nixon, 2009; Kopper, Nixon & Cloutier, 2005) Thus, it would be exceedingly difficult to effectively analyze the results of the surveys in any valuable manner. When creating this NSSI questionnai re, the wording of the questions was kept as close to the actual proposed DSM V criteria as possible. The primary cha and the criteria was formatted into question design. Additionally, two qu estions were lifetime, as well as the frequency of that behavior. The wording was kept the same as the question assessing NSSI within the past year, except the wording participants who engaged in NSSI during the past. The complete questionnaire can be found in the Appendix C. Individuals who endorsed self injury on the injured more
84 past year. Demographic Information A ll participants were asked to complete a demographic information questionnaire The demographic ques tions for this questionnaire were adapted from the 2010 Healthy Gators Student Survey Report. Healthy Gators sought to provide insight into ten key health and health risk related areas impacting University of Florida students. In order to provide the most comprehensive answer choices to questions such as ethnicity and sexual preference, the wording and choices from the 2010 survey were utilized. The demographics included in the questionnaire were: gender, age, current year in school, ethnicity and sexual orientation. A copy of the demographics informa tion sheet can be found in A ppendix D. Procedures The procedures and protocol for this study were submitted to the A copy of the IRB appr oval can be found in Appendix E Following IRB approval, the Principal Investigator (PI) contacted the assistant Vice President of Student Affairs at the University of Florida (UF) who, in turn, provided a sample of 5 000 students selected randomly by the University Registrar. All students included in the study were currently enrolled as undergraduates. Data collection began in February 2012 and was completed in March 2012. Student Voice, an online survey generator and manager used by the UF Office of Ins titutional Research was utilized to prepare the questionnaires in an
85 online format and collect the data. One week prior to the release of the survey, t he Assistant Vice President of Student Affairs initially sent an email to the 5,000 randomly selected stu dents encouraging their participation in the study. Following this contact Student Voice sent an email to all students in the sample with the link to the questionnaire and measures. Each ema il provide d students with a un ique link to access the survey, ens uring students complete d the survey only once, and making all responses independent. Students were informed that they were being asked to participate in a study examining the psychosocial outcomes of college students engaging in different degrees of self i njurious behaviors. Additionally, students were informed that they were not selected for the survey due to any knowledge of past self injurious behaviors or counseling center records. This was added to ensure that students understood their selection was co mpletely random and not due to any other factor. Data collection for the study consisted of three measures and took students appr oximately 20 minutes to complete online. When students click ed on the link in their email, they were directed to the study pag e hosted by Student Voice. This page included the IRB protocol, informed consent form, (Appendix E ) and a brief explanation of the purpose of the study Following the IRB and informed consent, students were asked to chec k a box indicating that they read th e informed consent and agreed to participate. They were also informed th at they were free to stop completing the survey a t any time. If participants cho se to stop wh ile completing the survey, incomplete answers were reported to the PI. Incomplete surveys w ere not used in the data analysis of this study.
86 Participants began the study by completing an online version of the Counseling Center Assessment of Psychological Symptoms ( CCAPS 62 ) (Locke et al., 2011) This measure took approximately five m inutes to com plete, and measured psychosocial functioning in eight distinct domains. Following completion o f the CCAPS 62, students were given an NSSI screening question (Appendix C) This question was deliberately broad, and meant to identify anyone who ha d ever engaged in any type of self injurious behavior. Following this question, participants were asked to endorse the types of self injury they ha d engaged in (e.g. scratching, cutting, burning, stabbing, hitting, and i nterfering with wound healing) either currently, or in the past. According to the literature, providing a checklist of self injurious behaviors generally elicits higher numbers of incident reporting (Heath, Toste, Nedecheva, & Charlebois, 2008), as some in dividuals who engage in behaviors such as interfering with wound healing do not always identify as self injurers. This question was constructed specifically to identify a broad group of self injurers, including those that do not meet DSM V proposed criteri a for diagnosis. Individuals who answer to this question and did not endo rse any type of self injury were directed to a demographic measure and were then inst ructed that they had finished the survey. However, individuals who answer ed to this question and/or endorse d any type of self injury were directed to a NSSI questionnaire prior to filli ng out demographic information. This screening question was designed to eliminate individuals who ha d no experience whatsoever with self injury.
87 Followin g the screening question, the NSSI measure was presented, and was i njury. This measure took approximately 5 10 minutes for students to com plete. Finally, these students were also asked several demographic q uestions to determine their age, gender, race/ethnicity, educational status, and sexual preference. At the conclusion of the survey, participants were d Wellness homepage, and were provided with the Alac hua County Crisis Center h otline. These resources were provided in case participants experience d any undue stress while participating in the study. Following the initial email contact, all students who had not yet completed the survey were sent a weekly r eminder email encouraging them to do so for a period of four weeks. Because of the unique link s ent to each student, it was possible for Student Voice to only send re minders out to students who had not yet participated in the study.
88 Table 2 1. Ethnicity Breakdown of Sample White, Non Hispanic Black, African American Hispanic or Latino/a Asian/ Pacific Islander Multiracial American Indian or Alaskan Native Other Current Study 69.8% 6% 13.2% 6.4% 2.9% 0% 1.9% Healthy Gators 2010 Survey 7 2% 8.2% 17.3% 14.4% 4.2% 1.1% Not specified University of Florida undergraduate population* (2010 enrollment) 58.5% 9.4% 16.5% 8.3% 1.7% .4% Not specified *(National Center for Education Statistics, 2010). Table 2 2. Gender Breakdown of Sample Male Female Current Study 32.7% 67.3% Healthy Gators 2010 Survey 32.3% 67.8% University of Florida undergraduate population* (2010 enrollment) 45% 55% *(National Center for Education Statistics, 2010)
89 Table 2 3. CCAPS 62 Correlation Coefficients Su bscale Correlation coefficient Depression .917 Eating Concerns .896 Substance Use .900 Generalized Anxiety .842 Hostility .834 Social Anxiety .888 Family Distress .914 Academic Distress .759 Table 2 4. Pearson Product Moment Correlation Coeffici ents of the CCAPS 62 Referent Measure CCAPS 62 Subscale Correlation Coefficient BDI Depression .721** EAT 26 Eating Concerns .648** AUDIT Substance Use .811** BAI Generalized Anxiety .643** SPDQ Social Anxiety .733** SFI Family Distress .648** A.A Academic Distress .680** Trait Anger Hostility .566** ** Indicates p<.001
90 Table 2 5. Data collection timeline First Stage Second Stage Third Stage What participants received Participants received a preliminary email from the Vice President of the Of fice of Student Affairs informing them of the upcoming study Participants received an email from Student Voice with a unique link to access the online survey Students who had not completed the survey received weekly follow up emails for a period of four we eks after the survey had been disseminated Date of receipt One week prior to survey dissemination occurred during the first week of February 2012 Second week of February 2012 Weekly; until the second week of March 2012
91 CHAPTER 3 RESULTS The current stu dy examined the psychosocial functioning of college students in relation to nonsuicidal self injurious (NSSI) behaviors. Specifically, it sought to determine whether there were significant differences in psychosocial functioning among individuals that do n ot self injure, individuals who have engaged in self injury in the past, and individuals who currently self injure. The goal of creating three distinct groups was to determine if psychosocial outcomes are significantly different between groups and to exami ne NSSI in accordance with the proposed DSM V revisions that categorize NSSI as a separate diagnostic category. The eight subscales from the CCAPS 62 were used as independent variables for the first research question in this study, and as the dependent va riables in the subsequent research questions. These variables were identified as depression, eating concerns, substance use, generalized anxiety, hostility, social anxiety, family distress, and academic distress. The dependent variables for the first resea rch question were identified as past NSSI, present NSSI, and no NSSI. These variables were used as independent variables for the remaining research questions. The research questions examined in this study were: 1. Which of the eight subscales on the CCAPS 62 are most strongly associated with NSSI behavior (either past or present behavior)? 2. Will individuals reporting present NSSI behaviors have significantly different levels of depression than those who report past NSSI behaviors, or no NSSI behaviors? 3. Will ind ividuals reporting present NSSI behaviors have significantly different levels of eating concerns than those who report past NSSI behaviors, or no NSSI behaviors?
92 4. Will individuals reporting present NSSI behaviors have significantly different levels of subst ance use than those who report past NSSI behaviors or no NSSI behaviors? 5. Will individuals reporting present NSSI behaviors have significantly different levels of generalized anxiety than those who report past NSSI behaviors or no NSSI behaviors? 6. Will indiv iduals reporting present NSSI behaviors have significantly different levels of hostility than those who report past NSSI behaviors or no NSSI behaviors? 7. Will individuals reporting present NSSI behaviors have significantly different levels of social anxiety than those who report past NSSI behaviors or no NSSI behaviors? 8. Will individuals reporting present NSSI behaviors have significantly different levels of family distress than those who report past NSSI behaviors or no NSSI behaviors? 9. Will individuals repor ting present NSSI behaviors have significantly different levels of academic distress than those who report past NSSI behaviors or no NSSI behaviors? This study sought to determine which of the constructs on the CCAPS 62 are most strongly associated with cu rrent levels of NSSI. Examining this relationship may help practitioners and researchers prioritize and develop treatments for some of the more commonly co occurring symptoms of NSSI. Identifying these relationships will also assist practitioners in recogn izing the common risk factors and disorders associated with NSSI, which may make it easier to identify individuals who self injure. For example, if an individual presents for counseling with a disorder that was found to be significantly correlated with NSS I, it can be helpful for the patient to be assessed for NSSI as well as the presenting concern. This study hypothesized that individuals who currently engage in NSSI will show significantly higher levels of depression, eating concerns, substance use, gener alized and social anxiety, hostility, family
93 distress and academic distress than individuals who have engaged in NSSI in the past, or those who have never engaged in NSSI behaviors. Of the 5,000 survey requests that were sent out, 1,131 (22.6%) students r esponded to the request. This total number produced a sample size of 733 completed surveys, yielding an overall response rate of 14.6%. Students who began surveys and then abandoned them, or submitted surveys with missing data were excluded from the final analysis. The majority of those who did not complete surveys stopped while completing the NSSI Questionnaire. Because the surveys were anonymous and answers were not associated with names or ID numbers, it was not possible to follow up with participants to further explore the missing data. the self total of 54 respondents (7.4%) reported engaging in NSSI in the past year (present NSSI group), and 118 (16.1%) indicated that they have engaged in NSSI at some time in the past, but not within the past year. Descriptive statistics were also calculated for gender and ethnicity of each of the groups (past NSSI, present NSSI, and no NSSI) and can be found in Tables 3 1 and 3 2. Males and females who engaged in NSSI in the past showed no significant differences on any of the eight CCAPS 62 variables (Table 3 3 ). The present NSSI group showed statistically significant dif ferences on the variables Depression, t(51.285)= 2.271 p=.004 and Academic Distress, t(52)= 2.260 p=.012 (Table 3 4 ). Finally, there were no statistically significant differences
94 found among males and females in the non NSSI group on any of the CCAPS 62 va riables (Table 3 5 ). The data from the first research question was analyzed using multinomial logistic regression to determine which of the eight CCAPS domains were most strongly associated with NSSI behaviors. Eight one way ANOVA tests were used to analy ze the data for the remaining research questions to determine if there were any significant differences between the eight independent variables identified on the CCAPS 62, and the three levels of NSSI identified on the self injury questionnaire (no NSSI/pa st NSSI/present NSSI). ANOVA tests were chosen because of their increased ability to demonstrate differences between the eight independent variables. The first research question posed in this study is: 1. Which of the eight subscales on the CCAPS 62 are the m ost strongly associated with NSSI behavior (either past or present behavior)? Multinomial logistic regression was used to analyze the data associated with this research question. Regression coefficients are reported in Tables 3 6 for prediction of membersh ip in the present NSSI group versus the no NSSI group and in Table 3 7 for prediction of membership in the past NSSI group versus the no NSSI group. A chi square test was used to determine if there was a significant relationship between the dependent varia ble (past, present, no NSSI) and the combination of 8 independent variables. The chi square statistic was significant, 2(16; n = 733) =124.164 p = .000, indicating that one or more of the 8 independent variables was significantly associated with classific ation in one of the three groups (i.e., past NSSI, present NSSI, or no NSSI groups ) Anxiety, 2(2; n = 733)= 13.933, p =.001 and hostility 2(2; n = 733) =12.765, p =.002
95 significantly predicted group membership. More specifically, hostility significantly discriminated between membership in the present and No NSSI groups, 2(1; n =773) = 5.584, p = .018, and membership in the past and No NSSI groups, 2(1; n =773) = 10.286 p =.001. In addition, a nxiety discriminated between the past and No NSSI groups,: 2(1; n =773) = 14.056, p =.000 In this study, hostility was found to be the strongest predictor of NSSI behaviors overall. Anxiety was the second strongest predictor. Table 3 8 presents means and standard deviations for the eight variables from the CCAP S 62 for each of the three NSSI groups. Research questions 2 8 asked whether there were differences among the groups on the means. To address these research questions, one way Welch ANOVA tests w ere conducted to determine if there were significant differen ces among the means of the three groups of self injury identified on the self injury questionnaire (no NSSI, past NSSI behaviors, and present NSSI behaviors) on any of the eight variables from the CCAPS 62. The Welch ANOVA test was used because the alterna tive, the ANOVA test, is known to be negatively impacted by unequal variances across the groups being compared, particularly when the groups are dramatically unequal in size, as they are in this study. As an additional check on the necessity of the Welch test, a Levene test comparing the variance was conducted for each of the eight variables to determine if equal variance assumptions are met Significant differences among the variances for the three NSSI groups were found for all variables, except Social Anxiety and Academic Distress. Because difference among the variances was prevalent, the Welch
96 ANOVA was used for all variables. Pairwise comparisons of the three NSSI groups were conducted following a significant Welch ANOVA. Pairwise comparisons were conducted using the Games Howell test, which is also designed for use when variances are not equal for the groups being compared. 2. Will individuals reporting present NSSI behaviors endorse significantly different levels of depression than those who report past NSSI behaviors, or no NSSI behaviors? For the depression variable, a statistically significant difference between the three groups was determined by the one way Welch ANOVA, F (2,111.365) = 24.225, p =.000. Results of the Games Howell post hoc test (s ee Table 3 8 ) indicated that mean depression was significantly different for all pairs of NSSI groups. Sample means indicate that individuals who currently engage in NSSI had the highest mean depression and individuals who do not engage in NSSI have the lo west mean depression. 3. Will individuals reporting present NSSI behaviors have significantly different levels of eating concerns than those who report past NSSI behaviors, or no NSSI behaviors? A statistically significant difference was found between the t hree groups as determined by a one way Welch ANOVA, F (2, 115.968) = 5.649, p =.005). The Games Howell test (see Table 3 8 ) indicated that there was no significant difference in mean eating concerns between those who engaged in NSSI in the past versus those currently engaging in NSSI. However, the mean for individuals who engaged in NSSI in the past was significantly different from the mean individuals who have never engaged in NSSI. The mean for individuals who currently engage in NSSI was not significantl y different from that of individuals who never engaged in NSSI. Sample means indicate that individuals who
97 currently engage in NSSI had the highest mean eating concerns, while individuals who have never engaged in NSSI had the lowest mean eating concerns. 4. Will individuals reporting present NSSI behaviors have significantly different levels of substance use than those who report past NSSI behaviors or no NSSI behaviors? A non statistically significant difference was found between the three groups as determi ned by a one way Welch ANOVA ( F (2, 109.836) = 2.970, p =.055). 5. Will individuals reporting present NSSI behaviors have significantly different levels of generalized anxiety than those who report past NSSI behaviors or no NSSI behaviors? A statistically sig nificant difference was found between the three groups as determined by a one way Welch ANOVA test ( F (2, 108.367) = 25.578, p =.000). The Games Howell post hoc test (see Table 3 8 ) found no significant mean anxiety difference found among individuals who en gaged in NSSI in the past, and those who currently engage in NSSI. However, there was a significant mean anxiety difference between individuals who engaged in NSSI in the past and those who have never engaged in NSSI. There also was a significant mean diff erence between those individuals who currently engage in NSSI and those who have never engaged in NSSI behavior. Sample means indicate that individuals who currently engage in NSSI have the highest generalized anxiety mean, while individuals who have never engaged in NSSI have the lowest generalized anxiety mean. 6. Will individuals reporting present NSSI behaviors have significantly different levels of hostility than those who report past NSSI behaviors or no NSSI behaviors?
98 A statistically significant diffe rence was found between the three groups as determined by a one way Welch ANOVA test (F(2, 111.559)=33.002, p=.000). The Games Howell post hoc test (see Table 3 8 ) determined that there was no significant mean difference in hostility between individuals th at engaged in NSSI in the past and those currently engaging in NSSI. A significant mean difference in hostility was found between individuals who engaged in NSSI in the past and those individuals who never engaged in NSSI. Additionally, there was a signifi cant mean difference between individuals who currently engage in NSSI and those who had never engaged in NSSI. Sample means indicate that those who currently engage in NSSI behaviors have the highest hostility mean, and those who have never engaged in NSSI have the lowest. 7. Will individuals reporting present NSSI behaviors have significantly different levels of social anxiety than those endorsed past NSSI behaviors or no NSSI behaviors? A statistically significant difference was found between the three grou ps as determined by the one way Welch ANOVA test (F(2, 118.306)=29.902, p=.000) The Games Howell post hoc test (see Table 3 8 ) determined that individuals who engaged in NSSI in the past did not show a significant difference in social anxiety then those wh o are presently engaging in NSSI behaviors. However, there was a significant mean difference between individuals who engaged in NSSI behaviors in the past and those who had never engaged in NSSI. There was also a significant mean difference between individ uals who currently engage in NSSI behaviors and those who have never engaged in NSSI. Sample means indicate that individuals who currently engage in NSSI have the highest social anxiety mean, while individuals who have never engaged in NSSI have the lowest
99 8. Will individuals reporting present NSSI behaviors have significantly different levels of family distress than those who report past NSSI behaviors or no NSSI behaviors? A statistically significant difference was found between the groups as determined by a one way Welch ANOVA test (F(2, 111.598) = 12.946, p =.000). The Game Howell post hoc test (see Table 3 8 ) found that there was no significant mean difference between individuals who engaged in NSSI in the past and those who currently engage in NSSI. A s ignificant mean difference was found between individuals who engaged in NSSI in the past and those who had never engaged in NSSI. Additionally, there was a significant mean difference found between those individuals currently engaging in NSSI and those who had never engaged in NSSI. Sample family distress means were highest for individuals who are currently engaging in NSSI and lowest for those who have never engaged in NSSI. 9. Will individuals reporting present NSSI behaviors have significantly different le vels of academic distress than those who endorsed past NSSI behaviors or no NSSI behaviors? A statistically significant difference was found between the groups as determined by a one way Welch ANOVA test ( F (2, 116.305)=25.519, p=.000). The Games Howell pos t hoc test (see Table 3 8 ) determined that mean academic distress was significantly different for all pairs of NSSI groups. Sample means indicate that individuals who currently engage in NSSI had the highest academic distress mean, while those who had neve r engaged in NSSI had the lowest.
100 Table 3 1. Gender Breakdown by Group Table 3 2. Ethnic Br eakdown by Group White, Non Hispanic Black, African American Hispanic or Latino/a Asian/ Pacific Islander Multiracial American Indian/ Alaskan Native Others Past NSSI 65.3% n=77 5.9% n=7 12.7% n=15 7.6% n=9 5.9% n=7 0% n=0 2.5% n=3 Present NSSI 74.1% n =40 3.7% n =2 9.3% n =5 7.4% n =4 3.7% n =2 0% n =0 1.9% n =1 No NSSI 69.5% n =390 6.2% n =35 13.4% n =75 6.6% n =37 2.7% n =15 0% n =0 1.6% n =9 Total 69.8% 6% 13.2% 6.4% 2.9% 0% 1.9% n=507 n =44 n =95 n =50 n =24 n =0 n =13 Male Female Past NSSI 30.5% n=36 69.5% n=82 Present NSSI 35.2% n=19 64.8% n=35 No NSSI 32.3% n=181 67.7% n=380 Overall Sample 32.7% n=236 67.3% n=497
101 Table 3 3. Means and Standard Deviations by Gender and CCAPS 62 Scale for the Past NSSI Group Male Female Depression .141 (1.14) .22 (1.41) Eating Concerns 3.00 (3.62) 1.97 (3.29) Sub stance Use .44 (2.60) 1.20 (4.49) Anxiety 2.21 (3.46) 1.67 (3.12) Hostility .06 (1.06) .02 (.89) Social Anxiety .21 (1.00) .22 (.96) Family Distress .12 (1.08) .12 (.82) Academic Distress .19 (.87) .42 (.94) Note. In each cell means are sho wn without parentheses and standard deviations are shown in parentheses.
102 Table 3 4. Means and Standard Deviations by Gender and CCAPS 62 Scale for the Present NSSI Group Male Female Depression .03 (1.30) 1.09 (2.14) Eating Concerns 2.40 (3.56) 2.76 (3.99) Substance Use 2.09 (4.90) 3.36 (6.69) Anxiety 1.86 (3.66) 2.33 (3.44) Hostility .04 (.90) .33 (.77) Social Anxiety .23 (.92) .64 (.87) Family Distress .10 (.99) .06 (.99) Academic Distress .23 (.71) .38 (.87) Note. In each cell means are shown without parentheses and standard deviations are shown in parentheses.
103 Table 3 5. Means and Standard Deviations by Gender and CCAPS 62 Scale for the non NSSI Group Male Female Depression .83 (1.20) .79 (1.24) Eating Concerns 1.59 (3.49) 1.27 (3.18) Substance Use 1.23 (4.08) .67 (3.23) Anxiety .04 (2.40) .02 (2.23) Hostility .57 (.63) .52 (.70) Social Anxiety .28 (.89) .31 (.90) Family Distress .42 (.77) .44 (.73) Academic Distress .64 (.84) .66 (.80) Note. In each c ell means are shown without parentheses and standard deviations are shown in parentheses.
104 Table 3 6. Present NSSI compared to no NSSI Regression Coefficient Standard Error Wald Chi Square Statistic p value Depression .175 .111 2.503 .114 Eating Conce rns .009 .044 .040 .841 Substance Use .045 .031 2.162 .141 Anxiety .083 .053 2.437 .119 Hostility .479 .203 5.584 .018* Social Anxiety .375 .202 3.456 .063 Family Distress .043 .205 .043 .836 Academic Distress .362 .214 2.852 .091 p < .05*, p < .0 01** Table 3 7. Past NSSI compared to no NSSI Regression Coefficient Standard Error Wald Chi Square Statistic p value Depression .013 .102 .017 .896 Eating Concerns .001 .032 .001 .975 Substance Use .018 .030 .377 .539 Anxiety .151 .040 14.046 .00 0* Hostility .495 .154 10.286 .001* Social Anxiety .231 .138 2.789 .095 Family Distress .169 .144 1.389 .238 Academic Distress .116 .158 .539 .463 p < .05*, p < .001**
105 Table 3 8. Means and Standard Deviations for Each Variable Past NSSI ( n = 118 ) Present NSSI ( n = 54) No NSSI ( n = 561 ) Depression .20 a (1.33) .72 b (1.94) .80 c (1.23) Eating Concerns 2.28 a (3.41) 2.63 ab (3.81) 1.37 b (3.28) Substance Use .97 a (4.01) 2.91 a (6.10) .85 b (3.54) Anxiety 1.83 a (3.22) 2.16 a (3.50) .00 b (2.29) Hostility .028 a (.94) .20 a (.83) .54 b (.68) Social Anxiety .22 a (.97) .49 a (.90) .30 b (.90) Family Distress .05 a (.91) .00 a (.98) .43 b (.74) Academic Distress .35 a (.93) .17 b (.87) .65 c (.81) Note. In each cell means are shown wit hout parentheses and standard deviations are shown in parentheses. Means within a row that share a common superscript are not significantly different by the Games Howell test
106 CHAPTER 4 DISCUSSION Late adolescence and early adulthood represents a developme ntally challenging transition for many colleg e students (Hunt & Eisenberg, 2010). Mental health problems are highly prevalent among this population, and the onset of most major mental illnesses begins prior to the age of 24 ( Eisenberg, Downs, Golberstein, & Zivin, 2009). Additionally, college counseling centers have reported an increase in the c omplexity and severity of young adults seen for treatmen t in the past decade (Blanco et al., 2008). Of concern is also the number of college age individuals using no nsucidal self injury (NSSI) as a means of coping with stressful events or mental illness. It has been estimated that approximately 12 17% of students in college populations have engaged in NSSI at least once in their lives ( Whitlock, Eels, Cummings, & Puri ngton, 2009). The proportion of college age individuals engaging in NSSI necessitates the need for additional research; however, the scholarship in this area con tinues to have significant gaps. Limitations in the literature have caused confusion among rese archers regarding the definition and prevalence rates of NSSI, and has made it difficult to effectively identify and treat individuals engaging in the behavior. The most current edition of the Diagnostic and Statistical Manual for Mental Disorders, Fourth Edition Text Revision (DSM IV TR ) only includes NSSI as a symptom of Borderline Personality Disorder (BPD). However, research indicates that a large proportion of individuals engaging in NSSI may not fall into any of the identified Axis I or Axis II cate gories (Muehlenkamp, 2005). This discrepancy has led many
107 researchers to argue that NSSI should be identified as a separate diagnostic category. Additionally, the lack of organized diagnostic information regarding NSSI has made it difficult for researchers to study its co morbidity and psychological outcomes. The overall lack of consensus regarding definition, prevalence, and outcomes have complicated and diluted the understanding of NSSI and the best ways to identify and treat the behavior. This study exam ined whether there were significant differences in the psychosocial functioning among college students who currently engage in NSSI, those who engaged in NSSI in the past, and those who have never engaged in NSSI behaviors. The first aim of this study was to determine whether there were any specific areas of mental health functioning strongly associated with NSSI behaviors. This question was posed in an effort to determine whether certain deficits in psychological functioning could help identify if an indi vidual was more likely to self injure. The second aim of this study was to determine whether individuals who presently engage in NSSI, or engaged in NSSI in the past have significantly higher levels of psychopathology than those who have never self injured Another major purpose of this study was to consider the newly proposed diagnostic criteria in the DSM V and issues associated with these proposed changes. This chapter will discuss the evolution of the research question and analysis, highlight key findin gs from this study and consider how these findings relate and add to the current NSSI literature, including the proposed DSM V addition of NSSI. Additionally, limitations of the current study will be discussed, as well as future directions for research.
108 Evolution of R e search Question and A nalysis The present study was originally designed to compare the psychosocial functioning of individuals who met the clinical criteria for NSSI with those who self injured but did not meet clinical criteria, and those wh o had never self injured. The information derived from this study was intended to align with the proposed DSM V criteria for a NSSI diagnosis. Prior to conducting this study, the NSSI questionnaire was developed based on the proposed DSM V criteria, with the projected groups of, Clinical NSSI, Subclinical NSSI, and No NSSI. If a participant endorsed all of the criteria required in making a diagnosis of NSSI; inflicted damage to the s urface of your body, (e.g., cutting, burning, stabbing, hitting, Association, 2010) two out of four statements associated with the functions of NSSI stated in criteria B, cri he behavior and its consequences cause clinically significant distress or impairment in interpersonal, academic, or other lusively during states of psychosis, would be placed in the Clinical group. Participants who endorsed criteria A and D, or reported engaging in two or more self injurious episo des over their lifetime (but not necessarily in the past year) would be placed in the Subclinical group. Finally, any participant that
109 f injury would be placed in the No NSSI group. During data analysis, several issues arose that necessitated the revision of the original research questions. Specifically, when scoring the NSSI measure, it was determined that only 2 out of 172 participants fell into the Clinical category. Upon further examination, it was found that an additional 9 participants were one criterion short of falling into Clinical category. The most predominant criterion not injurious behavior cause significant distress or It is possible that the participants did not feel that their self injurious behaviors significantly interfered with their functioning, or caused signif icant distress, although a clinician might disagree with this statement. Because it was not possible own functioning wa s accurate, this question may have yielded inaccurate responses. These resul ts made it impossible to examin e the data in relation to the Clinical and Subclinical groups, as a group of 2 cannot yield reliable statistical information. Consequently, it was necessary to reorganize the data into more appropriate groups. It was determin ed you engaged in intentional, self inflicted damage to the surface of your body, (e.g., cutting, burning, stabbing, hitting, excessive rubbing), for purposes not socially sanctioned (e.g., body piercing, tattooin determ ine whether participants could be classified as current self injurers or p ast self injurers.
110 The revision of the NSSI groups and subsequent research questions has important implications for the proposed DSM V addition of NS SI. It is possible that the diagnostic criterion for the proposed NSSI disorder is too stringent, thus making it difficult for clinicians to diagnose self injuring individuals with an NSSI disorder. If this is the case, it once again makes it difficult to accurately measure and determine the incidence and prevalence of NSSI in both the community and clinical populations. An underestimation of individuals with NSSI may also contribute to an untreated pop ulation of individuals who are engaging in NSSI but do not meet the formal criteria to be diagnosed with a disorder. There has been some preliminary data collected for the DSM V, and kappa statistics for several of the new or revised disorders have been published in the American Journal of Psychiatry. These k appa statistics were derived from the test retest reliability of clinicians in diagnosing the disorders in clinical field trials A kappa value of 0.6 0.8 is considered excellent, 0.4 0.6 is considered good, and 0.2 0.4 may be acceptable. Any kappa statist ics below 0.2 are considered unacceptable. The initial field trials for the NSSI diagnosis yielded a kappa statistic of .03, which can be considered unacceptable in terms of test retest reliability (Freedman, et al., 2013). Borderline Personality Disorde r, which includes in its symptoms the only reference to NSSI in the DSM IV, had a kappa statistic of 0.54, which represents stronger diagnostic reliability. The final revision of the DSM V, which will be released in May 2013, does not include a diagnostic category of NSSI. While this diagnosis was ultimately not added to the DSM V, it still remains an important behavior worthy of further clinical research.
111 Study Implications Cheng, Mallinckrodt, Soet, & Sevig (2010) hypothesized that college students prese nting at a counseling center would be unlikely to disclose self injurious behaviors at the time of intake due to stigma and anonymity concerns It has also been suggested that presenting a self injury measure at time of intake would result in inaccurate an swers and lengthy intakes. Additionally, a 2006 study conducted by Whitlock et al., found that 36% of respondents reported that no one else wa s aware of their NSSI behavior. Because m any college counseling centers already utilize the CCAPS 62 as a screenin g tool for new intakes, it would be extremely useful for clinicians to know if there are any subscales that could In this study, high levels of hostility and anxiety were found to be the best predictors of NSSI in col lege students. I ndividuals who engage in NSSI behaviors were found to have significantly higher levels of hostility than those who had never engaged in NSSI. Individuals who engaged in NSSI in the past were also found to have higher levels of hostility tha n those who had never engaged in NSSI. However, individuals who currently engage d in NSSI behaviors did not demonstrate significantly different levels of hostility than those who engaged in NSSI in the past. disposition consisting of chronic suspicion, mistrust and cynicism, the frequent experience of angry thoughts and emotions, and a tendency for easily evoked anger and aggressive re search has found hostility to be an important indicator for a wide range of
112 phys ical and psychosocial outcomes (Maier, et al., 2009). Hostility as a predictor of self injurious behavior aligns with one prior study examining the hostility model of NSSI (Ro ss & Heath, 2003). This model proposed that individuals engage in NSSI when they are unable to openly express feelings of anger and hostility (Ross & Heath, 2003). Essentially, self injury provides a cathartic outlet for individuals who are unable to adequ ately express these feelings. ( 2003 ) study was the only study found that attempted to find a relationship between hostility and NSSI. However, the current study provides additional evidence that there is a significant link between hostility traits and NSSI. This finding may be especially useful to college counseling centers hoping to identify or critical marker for NSSI behaviors. It is important to note that this finding cannot be used to inform diagnostic decision making as it is only one piece of understanding an However, it can be used as a tool to prompt clinicians to complete a follow up assessment in this area. This finding also provides suppo rt to criteria B.1 on the DSM V proposed NSSI diagnosis generalized distress, or self criticism, occurring in the period immediately prior to the self injurious act. s a predictor of NSSI also supports the four function model (FFM), wherein NSSI serves an automatic negative function that reduces hostile feelings in an individual. Anxiety was also identified as a predictor of NSSI in this study. Historically, literatur e has determined that anxiety i s the psychological
113 characteristic perhaps most strongly associated with NSSI (Andover, Pepper, Ryabchenko, Orrico, & Gibb, 2005). The current study supports the link between anxiety and NSSI behaviors by finding that individ uals currently engaging in NSSI had significantly higher levels of generalized and social anxiety than those who had never engaged in NSSI. Additionally, those who had engaged in NSSI in the past were also found to have significantly higher levels of socia l and overall anxiety than those who had never engaged in NSSI. This finding also supports the automatic negative reinforcement (ANR) function of the four function model of NSSI. ANR is generally understood as a way to moderate anxiety (reduce a negativ e feeling or state) (Lloyd Richardson, Nock, & Prinstein, 2009). NSSI may function as a way for individuals to experience immediate relief from anxious thoughts or feelings (Favazza, 1998). This finding is also consistent with the psychological precipitant criterion that notes that psychological distress such as depression or anxiety accompanies the NSSI behaviors (American Psychiatric Association, 2012). The second goal of this study was to examine each of the eight subscale domains on the CCAPS 62 in re lation to NSSI. Specifically, the research questions sought to determine if individuals engaging in NSSI behaviors (either presently or in the past) showed more psychological symptoms than individuals who have never engaged in NSSI. The subscale domains id entified in the CCAPS 62 were; depression, eating concerns, substance use, generalized anxiety, hostility, social anxiety, academic distress, and family distress. The
114 below discussion does not include anxiety or hostility, as they have already been mention ed in relation to research question one Depression The current study found that individuals who engaged in NSSI behaviors (either presently or in the past) showed significantly more depressive symptoms than those who had never engaged in NSSI. Furthermor e i ndividuals who currently engage in NSSI demonstrated significantly more depressive symptoms than those who engaged in NSSI more than one year ago. According to the DSM V proposed revisions, one of the criteria required to diagnose NSSI is the psycholog feelings or thoughts, such as depression, anxiety, tension, anger, generalized distress, or self criticism, occurring in the period immediately prior to the self 2012). This finding provides support for the inclusion of the psychological precipitant, specifically depression, in the diagnostic criteria. As the research in this area has been mixed, the current study adds support to the notion that there is a link be tween individuals who engage in NSSI and depression. A recent study conducted by Wilcox, Arria, Caldeira, Vincent, diagnosis of depression predicted past year NSSI and had independen t It is important to note, however, that there was also questionable reliability with the kappa statistic associated with Major Depressive Disorder (MDD) in the upcoming DSM V which was 0.28 (Freedman et al., 2013 ). The diagnostic criteria for MDD remains unchanged from the DSM IV TR calling into question the reli ability of the current diagnostic
1 15 definition It is possible that the parameters of depression are still not well defined, which may affect how it relat es to NSSI symptoms. Eating C oncerns This study also found a significant correlation between eating concerns and individuals who have engaged in NSSI, both currently and in the past. Specifically, it was found that individuals engaging in NSSI ( presently or in the past ) have significantly more eating concerns than those who have never engaged in NSSI. This finding aligns with past research indicating that there is a significant relationship between disordered eating and NSSI behaviors (Claes, Vanderecycke n, & Vertommen, 2005; Gollust, Eisenberg, & Golberstein, 2008; MacLaren & Best, 2010; Ross, Heath, & Toste, 2009). Studies have found that prevalence rates for eating disordered patients who also engage in NSSI are between 25.4% and 55.2% (Muehlenkamp, Cl aes, Smits, Peat, & Vandereycken, 2011; Svirko & Hawton, 2007). Research suggests that trait impulsivity is a common characteristic of both dis ordered eating and NSSI, (MacLaren & Best, 20 10; Ross, Heath, & Toste, 2009) as well as child abuse, low self est eem, and childhood traumatic experiences. (Muehlenkamp et al., 2011). Substance A buse A non statistically significant difference was found between the three groups and the substance use variable. According to research, ther e is significant evidence of co m orbidity between the substance abuse and nonsuicidal self injury. NSSI is common among clinical samples of substance dependent individuals, with prevalence rates reported between 34 50% (Serras, Saules, Cranford, & Eisenberg, 2010), and drug use has been f ound to be a risk
116 factor for self injury among adolescents (Serras, Saules, Cranford, & Eisenberg, 2010). It is important to note that in previous studies, drug use has been found to be more predictive of NSSI behaviors than alcohol use (Serras, Saules, Cr anford, & Eisenberg, 2010). The distinction could not be made in this study, however, because drug and alcohol abuse were measured as one variable. Family D istress Family distress also had a significant relationship with NSSI. Specifically, individuals wh o endorsed NSSI behaviors (either presently or in the past) had significantly higher rates of self reported family distress than those who had never engaged in NSSI. However, there was no significant difference in family distress found between individuals who currently engage in NSSI, and those who engaged in NSSI in the past. Research has suggested that individuals who self injure often experience family distress, such as parental criticism, family incohesiveness, parental overprotection or over involvemen t, and poor parental communic ation (Tulloch, Blizzard & Pink us, 1997; Wedig & Nock, 2007, Wong, Stewart, Ho, & Lam, 2007; You & Leung 2012). A recent study conducted by Claes, Houben, Vandereycken, Bijttebier, & Muehlenkamp, (2010) found that adolescents w ho engage in NSSI tend to have less positive relationships with their parents than their non self injuring peers. Additionally, the DSM V proposed revision for NSSI states that the behavior must cause significant impairment in interpersonal, academic, or o ther areas of functioning (American Psychiatric Associat ion, 2012). The above finding supports the idea that individuals engaging in NSSI have more family distress and interpersonal difficulties than those who do not engage in NSSI.
117 Academic D istress Fina lly, academic distress was found to have a significant relationship with NSSI. Individuals who self injured both presently and in the past were found to have higher levels of academic distress than their non self injuring peers. Additionally, individuals w ho engaged in self injury in the past exhibited significantly less academic distress than those who still self injure While there have been relatively few studies conducted related to NSSI and academic distress, Claes, et al., (2010) found that adolescent s with NSSI scored lower in four academic areas (Verbal, Mathematics, Problem Solving, and General Academics) than their non self injuring peers. The concerns associated with performing poorly in school appear to cause significant distress in th ese adolesc ents engaging in NSSI behaviors. While examining this variable, it is important to note that all the students surv eyed in this sample were from a Research One (R1) institution with rigorous academic expectations. It is possible that a sample from a communi ty college, or a non college sample would yield a different set of outcomes. Summary of Implications The findings in this study indicate that, as hypothesized, individuals engaging in NSSI experience significantly more negative psychosocial outcomes than their non self injuring peers. Specifically, those who engaged in NSSI in the past year were significantly different on all of the psychosocial variables examined, with the exception of substance abuse. Those who had endorsed at least two episodes of self injury in their lifetimes (but not in the past year) had significantly different rates of depression, eating concerns, anxiety, family
118 distress, and academic distress than those who never engaged in NSSI. The only variable t hat was not associated with past NSSI behaviors was substance abuse. Hostility was also found to be the most strongly correlated with NSSI The information derived from this study provides further support to the four func tion model (FFM) proposed by Lloyd Richardson, Nock, & Prinstein (2009) Specifically, the results support the Automatic Negative Reinforcement (ANR) function, which posits that NSSI is used as a means to alleviate a negative feeling or state. While it is not certain whether the individuals using NSSI are doing it to alleviate feelings of depression, anxiety, or hostility, these feelings have generally been found to precede the behavior (American Psychiatric Association, 2012). Limitations Although this st udy was beneficial in contributing to the current scholarship in the area of NSSI in college students, several limitations can be identified. The following will describe the se limitations in terms of the internal and external validity of the results. Inte rnal V alidity This study was conducted as an online survey completed by undergraduate students. Self selection bias is always a problem when conducting online survey research, as students have the opportunity to opt in or out of the survey based on any num ber of mitigating factors. While the ultimate sample size was relatively large (733), it was a small proportion considering the original number of survey requests sent out (5,000). Perhaps the biggest threat
119 to the internal validity of this study was the 3 98 students who began filling out the survey and discontinued it before finishing Because of the anonymity of online survey research, there was no way to follow up with these students to determine why they decided to withdraw their participation from the study. It is impossible to know whether these students discontinued their participation at random, or due to sp ecific circumstances. I t appeared that many students terminated their participation in the study after answering the preliminary questions regard ing self injury. Specifically, participants often stopped when asked about questions regarding their thoughts, feelings and behaviors during and surrounding the time of the self injurious act. It is possible that the students who abandoned the survey would have been the ones falling into the clinical group, but there was no way to determine what that pattern may be under the design of the current study. The norming sample used in the CCAPS 62 also represents a threat to CCAPS 62 is used as an assessment tool for students presenting to college counseling centers, thus, a clinical sample was population as a whole; rather it is indicative of college students who present for counseling services. As a result, students who fell in the average range on this assessment may still be experiencing distress in some of the areas assessed. If the study was conducted with a measure from a community sample, more statistically significant results may have been found.
120 External V alidity While this study represents a contribution to the field of NSSI research, there are threats to external validity that should be mentioned. While the sample size can be considered r elatively large, it was restricted to one public university in the southeastern United States. From the results of this study, it is not possible to determine if students in another part of the country or from a different school would have similar experien ces regarding psychological health and NSSI. As mentioned earlier, University of Florida is a highly ranked and academically rigorous institution, which may also have influenced any of the variables examined. The age and education level of the students fil ling out this survey also restricts the application of the study, as the survey was limited to college students. Individuals who are college aged but not attending a university may have different experiences with NSSI. Future Directions Research suggests t hat college student mental health is rapidly becoming a more complex and significant issue deserving of further study. Additionally, an increase among individuals engaging in NSSI has necessitated further research and intervention in this area as well. Thi s study was designed to c ontribute to the research on NSSI and provide support to the p roposed DSM V criteria for n on suicidal self i njury. This study added to the current research base on NSSI, and also lent support to several of the diagnostic criteria p roposed for the DSM V NSSI diagnosis. Specifically, the current findings strengthened the proposed criteria stating psychological precipitants and significant interpersonal distress.
121 The sample size of this study did not provide support for the existence of a clini cal category for NSSI, as only 2 participants fell into the Clinical category. As a result, it was impossible to say whether there were significant psychological differences between individuals fitting the clinical criteria, and those falling in to a subclinical category. It would be beneficial for future studies on this subject to increase the sample size to better capture the true nature of the clinical diagnosis. It is possible that a larger sample size across several universities would have y ielded a clinical group that could be statistically analyzed. Future studies in this area might also utilize other methods of data collection. If a follow up study were planned, it would be useful to use a data collection method other than online survey d ata. This method of data collection made it impossible to determine why 398 participants abandoned the survey in the middle. If it was possible to follow up with these 398 participants, it may have strengthened the results of this study. The results of th is study supports the adoption of a NSSI Not Otherwise Specified (NOS) category in which individuals engage in NSSI, but the severity of the behavior does not reach clinical levels. Future research examining this issue will be useful in light of the decisi on not to include NSSI to the DSM V. It will be imperative for future studies to determine the most appropriate way to measure NSSI behaviors.
122 APPENDIX A IRB AND INFORMED CONSENT
126 APPENDIX B CCAPS 62
128 APPENDIX C CCAPS 62 AUTHOR APP ROVAL
129 APPENDIX D NSSI SCREENING QUEST ION & QUESTIONNAIRE any of the following that you have engaged in during your lifetime. Cutting Inte rfering with wound healing (picking a scab until it bleeds) Burning Hitting self Scratching In the past year, have you engaged in intentional, self inflicted damage to the surface of your body, (e.g., cutting, burning, stabbing, hitting, exce ssive rubbing), for purposes not socially sanctioned (e.g., body piercing, tattooing, etc.) Please note; this does not include common or trivial behaviors such as nail biting or picking at a wound. Yes No How many times have you engaged in t his behavior in the past year? Once 2 4 times 5 or more times Have you engaged in the intentional, self inflicted damage to the surface of your body, (e.g., cutting, burning, stabbing, hitting, excessive rubbing), for purposes not socially sanct ioned (e.g., body piercing, tattooing, etc.) at ANY TIME in your life? Yes No How many times have you engaged in this behavior in your lifetime ? Once 2 4 times 5 10 times More than 10 times Was this behavior done with suicidal intent?
130 Ye s No Please indicate whether these statements apply to you by checking yes or no for each item: Negative feelings or thoughts, such as depression, anxiety, tension, anger, generalized distress, or self criticism, occurred in the period immediatel y prior to injuring myself Yes No Prior to engaging in the act, I experience a period of preoccupation with the intended behavior that was difficult to resist Yes No The urge to engage in self injury occurs frequently, although I might not act on it Yes No The act is engaged in with a purpose; this might be relief from a negative feeling/cognitive state, interpersonal difficulty or increase of a positive feeling state I generally anticipate that these feelings will occur eith er during or immediately following the self injury Yes No Does the self injurious behavior cause significant distress or impairment in interpersonal, academic, or other important areas of functioning? Yes No Do the episode(s) of self inj ury occur only when you are intoxicated with one or more substances? Yes No
131 APPENDIX E DEMOGRAPHIC INFORMAT ION We would like to gather additional information about you to better inform the results of this study. Please answer all questions indicated. 1. Please indicate your gender Male Female 2. How old are you? 17 20 21 29 30 or older 3. What is your current year of school? 1 st year 2 nd year 3 rd year 4 th year 5 th year or more 4. What is the ethnicity you most clo sely identify with? Hispanic or Latino/a White Black Asian or Pacific Islander Multiracial American Indian or Alaska Native 5. What is the sexual orientation you most closely identify with? Heterosexual Gay Lesbian Bisexual/pansexual/ queer Unsure
132 LIST OF REFERENCES Adrian, M., Zeman, J., Erdley, C., Lisa, L., & Sim, L. (2011). Emotional dysregulation and interpersonal difficulties as risk factors nonsuicidal self injury in adolescent girls. Journal of Abnormal Child Psychology, 39, 389 400. Alexander, N., & Clare, L. (2004). You still feel different: The experience and injury in the context of a lesbian or bisexual identity. Journal of Community and Applied Social Psychology, 14, 70 84. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4 th ed., Rev.) Washington, DC: Author. American Psychiatric Association. (2012). Proposed Revision: Nonsuicidal Self Injury Retrieved from: http://www.dsm5.org/ProposedRevisions /Pages/proposedrevision.aspx?ri d=443# Andover, M. S., Pepper, C. M., Ryabchenko, K. A., Orrico, E. G., & Gibb, B .E. (2005). Self mutilation and symptoms of depression, anxiety and borderline personality disorder. Suicide and Life Threatening Behavior, 35 581 591. Ballard, E., Bosk, A., & Pao, M. (2010). Invited commentary: Understanding brain suicidal self injury. Journal of Youth and Adolescence, 39, 327 334. Benton, S. A., Robertson, J. M., Tseng, W., Newton, F. B., & Benton, S. L. (2003). Changes in counseling center client problems across 13 years. Professional Psychology: Research and Practice, 34, 66 72. Blanco, C., Okuda, M., Wright, C., Hasin, D. S., Grant, B. F., Liu, S., & Olfson, M. (2008). M ental health of college students and their non college attending peers: Results from the national epidemiologic study on alcohol and related conditions. Archives of General Psychiatry, 65, 1429 1437. Bleiberg, E. (2001). Treating personality disorders in children and adolescents: A relational approach. New York, NY: Guilford Press. Bohus, M., Limberger, M., Ebner, U., Glocker, F. X., Schwarz, B., Wernz, M., & Lieb, K. (2000). Pain perception during self reported distress and calmness in patients with borde rline personality disorder and self mutilating behavior. Psychiatry Research, 95, 251 260.
133 Borrill, J., Fox, P., Flynn, M., & Roger, D. (2009). Students who self harm: Coping style, rumination and alexithymia. Counseling Psychology Quarterly, 22, 361 372. Brausch, A. M., & Gutierrez, P. M. (2010). Differences in non suicidal self injury and suicide attempts in adolescents. Journal of Youth and Adolescent, 39, 233 242. Brown, M. Z., Comtois, K. A., & Linehan, M. M. (2002). Reasons for suicide attempts and nonsuicidal self injury in women with borderline personality disorder. Journal of Abnormal Psychology, 111, 198 202. Bureau, J. F., Martin, J., Freynet, N., Poirier, A., Lafontaine, M. F., & Cloutier, P. (2010). Perceived dimensions of parenting and non suicidal self injury in young adults. Journal of Youth and Adolescence, 39, 484 494. Carlson, L., DeGeer, S. of self cutting behavior among the adolescent population. PRAXIS, 5, 22 29. Cavanaugh, R. M. (2002). Self mutilation as a manifestation of sexual abuse in adolescent girls. Journal of Pediatric Adolescent Gynecology, 15, 97 100. Center for Collegiate Mental Health (2010). CCAPS 2010 user manual. University Park, PA: Chapman, A .L., Gratz, K. L., & Brown, M. Z. (2006). Solving the puzzle of deliberate self harm: The experiential avoidance model. Behaviour Research and Therapy, 44, 371 394. Cheng, H., Mallinckrodt, B., Soet, J., & Sevig, T. (2010). Developing a screening instrument and at risk profile for nonsuicidal self injurious behavior in college women and men. Journal of Counseling Psychology, 57, 128 139. Claes, L., Houben, A., Vanderecycken, W., Bijttebier, P., & Muehlenkamp, J. (2010). Brief report: The association between non suicidal self injury, self concept, and acquaintance with self injurious peers in a sample of adolescents. Journal of Adolescence, 33, 775 778. Claes, L., Vanderecycken, W., & Vertommen, H. (2005). Self care versus self harm: Piercing, tattooing and self injuring in eating disorders. European Eating Disorders Review, 13, 11 18. Connors, R. (1996). Self injury in trauma survivors: Functions and meanings. American Journal of Orthopsychiatry, 66, 197 206.
134 Cornish, J. A., Kominars, K. D., Riva, M. T., McIntosh, S., & Henderson, M. C. (2000). Perceived distress in university counseling center clients across a six year period. Journal of College Student Development, 41, 104 109. Cross, L. W. (1993). Body and self in feminine development: Implications for eating disorde rs and delicate self mutilation. Bulletin of the Menninger Clinic, 57, 41 67. Crowell, S. E., Beauchaine, T. P., McCauley, E., Smith, C. J., Vasilev, C. A., & Stevens, A. L. (2008). Parent child interactions, peripheral serotonin, and self inflicted injur y in adolescents. Journal of Consulting and Clinical Psychology, 76, 15 21. Croyle, K. L., & Waltz, J. (2007). Subclinical self harm: Range of behaviors, extent, and associated characteristics. American Journal of Orthopsychiatry, 77, 332 342. De Leo, 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. Dougherty, D. M., Mathias, C. W., Marsh Richard, D. M., Prevette, K. N., Dawes, M. A., Hatzis, E. O. (2009). Impulsivity and clinical symptoms among adolescents with non suicidal self injury with or without attempted suicide. Psychiatry Research,. 169, 22 27. Eisenberg, D. E., Downs, M. F., Goldberstein, E., & Zivin, K. (2009). Stigma and help seeki ng for mental health among college students. Medical Care Research and Review, 66, 522 541. Eisenberg, D. E., Golberstein, E., & Gollust, S. E. (2007). Help seeking and access to mental health care in a university student population. Medical Care, 45, 594 601. Favaro, A., & Santonastaso, P. (2000). Self injurious behavior in anorexia nervosa. Journal of Nervous and Mental Disease, 188, 537 542. Favazza, A. R. (1996). Bodies under siege: Self mutilation and body modification in culture and psychiatry. Bal timore, MD: The Johns Hopkins University Press. Favazza, A. R. (1998). The coming of age of self mutilation. Journal of Nervous and Mental Disease, 186, 259 268.
135 Fliege, H., Lee, J. R., Grimm, A., & Klapp, F. B. (2009). Risk factors and correlates of del iberate self harm behavior: A systematic review. Journal of Psychosomatic Research, 66, 477 493. Freedman, R., Lewis, D. A., Michels, R., Pine, D. S., Schultz, S. K., Tamminga, 5: New blooms and old th orns. American Journal of Psychiatry, 17 0, 1 5. Gallagher, R. P. (2009). National survey of counseling center directors. The International Association of Counseling Services, Monograph Series Number 8R, 1 48. Glassman, L. H., Weierich, M. R., Hooley, J. M., Deliberto, T. L., & Nock, M. K. (2007). Child maltreatment, nonsuicidal self injury, and the mediating role of self criticism. Behavior Research and Therapy, 45, 2483 2490. Glenn, C. R., & Klonsky, E. D. (2009). Social context during non suicidal self injury indicates suicide risk. Personality and Individual Differences. 46, 25 29. Glenn, C. R., & Klonsky, E .D. (2010). A multi method analysis of impulsivity in nonsuicidal self injury. Personality Disorders: Theory, Research, and Treatment, 1, 67 75. Gollust, S. E., Eisenberg, D., & Golberstein, E. (2008). Prevalence and correlates of self injury among university students. Journal of American College Health, 56, 491 498. Granic, I., & Dishion, T. J. (2003). Deviant talk in adolescent friendships: A st ep toward measuring a patheogenic attractor process. Social Development, 12 (3), 314 333. Gratz, K. L. (2006). Risk factors for deliberate self harm among female college students: The role and interaction of childhood maltreatment, emotional inexpressivity, and affect intensity/reactivity. American Journal of Orthopsychiatry, 76 238 250. Gratz, K. L. (2007). Targeting emotion dysregulation in the treatment of self injury. Journal of Clinical Psychology: In Session, 63, 1091 1103. Guerry, J. D., & Prinstein M. J. (2010). Longitudinal prediction of adolescent nonsuicidal self injury: Examination of a cognitive vulnerability stress model. Journal of Clinical Child and Adolescent Psychology, 39, 77 89.
136 Gutierrez, P. M., Osman, A., Barrios, F. X., & Kopper, B. A. (2001). Development and initial validation of the self harm behavior questionnaire. Journal of Personality Assessment, 77 475 490. Hawton, K., Arensman, E., Townsend, E., Bremner, S., Feldman, E., Goldney, Bendz, L. (1998). Deliberate self harm: Systematic review of efficacy of psychosocial and pharmacological treatments in preventing repetition. British Medical Journal, 317, 441 447. Hawton, K., Rodham, K., Evans, E., & Weatherall, R. (2002). Deliberate self harm in adolescents: Sel f report survey in schools in England. British Medical Journal, 325, 1207 1211. Hayes, S., Wilson, K. G., Gifford, E. V., Follette, V. M., & Strosahl, K. (1996). Experiential avoidance and behavioral disorders: A functional dimensional approach to diagnos is and treatment. Journal of Consulting and Clinical Psychology, 64, 1152 1168. Healthy Gators Student Survey Report. (2010). University of Florida. Retrieved from http://healthygators.ufsa.ufl.edu/wp content/uploads/2011/03/HG Student Survey 2010 Revised .pdf Heath, N. L., Baxter, A. L., Toste, J. willingness to access school based support for nonsuicidal self injury. Canadian Journal of School Psychology, 25, 260 276. Heath, N. L., Schaub, K., Holly, S., & Nixon, M. K. (2009). Self injury today: Review of population and clinical studies in adolescents. In M. K. Nixon & N. L. Heath (Eds.) Self injury in youth: The essential guide to assessment and intervention, (pp. 9 27). New York, NY: Routledge. Heath, N. L., Tost e, J. R., & Beettam, E. injury. Canadian Journal of School Psychology, 21, 1/2, 73 92. Heath, N. L., Toste, J. R., Nedecheva, T., & Charlebois, A. (2008). An examination of nons uicidal self injury among college students. Journal of Mental Health Counseling, 30, 137 156. Heilbron, N., & Prinstein, M.J. (2008). Peer influence and adolescent nonsuicidal self injury: A theoretical review of mechanisms and moderators. Applied and Pre ventive Psychology, 12, 169 177.
137 Herpertz, S., Sass, H., & Favazza, A. (1997). Impulsivity in self mutilative behavior: Psychometric and biological findings. Journal of Psychiatric Research, 31, 451 465. 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. Hilt, L.M., Nock, M.K., Lloyd Richardson, E.E., & Prinstein, M.J. (2008). Longitudinal study of nonsuicidal self injury among young adolescents. Journal of Early Adolescence, 28 (3), 455 469. Hoff, E.R., & Muehlenkamp, J.J. (2009). Nonsuicidal self injury in college students: The role of perfectionism and rumination. Suicide and Life Threatening Beh avior, 39, 576 587. Hunt, J. & Eisenberg, D. (2010). Mental health problems and help seeking behavior among college students. Journal of Adolescent Health, 46, 3 10. International Society for the Study of Self injury (2007). Definitional issues surroundin g our understanding of self injury. Conference proceedings from the annual meeting. Montreal, Canada. Janis, I.B., & Nock, M.K. (2009). Are self injurers impulsive?: Results from two behavioral laboratory studies. Psychiatry Research, 169, 261 267. Kanan, L.M., Finger, J., & Plog, A. (2008). Self injury and youth: Best practices for school intervention. School Psychology Forum: Research in Practice, 2, 67 79. Kibler, J. (2009). Self injury in the schools: An exploratory analysis of Midwest school counselor North American Journal of Psychology, 11, 309 322. Kisch, J., Leino, V., & Silverman, M. M. (2005). Aspects of suicidal behavior, depression and treatment in college students: Results from the spring 2000 national college heal th assessment survey. Suicide and Life Threatening Behavior, 35, 3 13. Kitzrow, M. Challenges and recommendations. National Association of School Personnel Administrators Journal, 41, 167 181 Klonsky, E. D. (2007). The functions of deliberate self injury: A review of the evidence. Clinical Psychology Review, 27, 226 239.
138 Klonsky, E. D., & Glenn, C. R. (2009). Psychosocial risk and protective factors. In M. K. Nixon & N. L. Heath (Eds.), S elf injury in youth: The essential guide to assessment and intervention, (pp. 45 58). New York, NY: Routledge. Klonsky, E. D., & Moyer, A. (2008). Childhood sexual abuse and non suicidal self injury: Meta analysis. The British Journal of Psychiatry, 192, 166 170. Kraft, D. P. (2009). Mens Sana: The growth of mental health in the American College Health Association. Journal of American College Health, 58, 267 275. Kress, V. E., & Hoffman, R. M. (2008). Non suicidal self injury and motivational interviewin g: Enhancing readiness to change. Journal of Mental Health Counseling, 30, 311 329. Kumar, G., Pepe, D., & Steer, R. self reported reasons for cutting themselves. Journal of Nervous and Mental Disease, 192, 83 0 836. Latzman R. D., Gratz, K. L., Young, J., Heiden, L. J., Damon, J. D., & Hight, T. L. (2010). Self injurious thoughts and behaviors among youth in an underserved area of the southern United States: Exploring the moderating roles of gender, racial/eth nic background and school level. Journal of Youth and Adolescence, 39, 270 280. Laye Gindhu, A., & Schonert Reichl, K.A. (2005). Nonsuicidal self harm among harm. Journal of Youth and Ado lescence, 34 (5), 447 457. Levesque, C., Lafontiane, M. F., Bureau, J. F., Cloutier, P., & Dandurand, C. (2010). The influence of romantic attachment and intimate partner violence on non suicidal self injury in young adults. Journal of Youth and Adolescence 39, 474 483. Linehan, M. (1993). Cognitive behavioral therapy for borderline personality disorder. New York: Guilford Press. Lloyd Richardson, E. E., Nock, M. K. & Prinstein, M. K. (2008). Functions of adolescent non suicidal self injury. In M. K. Nixo n & N. Heath (Eds.), Self injury in youth: Essential guide to assessment and intervention New York NY : Routledge.
139 Lloyd Richardson, E. E., Perrine, N., Dierker, L., & Kelley, M. L. (2007). Characteristics and functions of non suicidal self injury in a c ommunity sample of adolescents. Psychological Medicine, 37, 1183 1192. Locke, B. D., Buzolitz, J. S., Lei, P. W., Boswell, J. F., McAleavey, A. A., Sevig, T. D... Hayes, J. A. (2011). Development of the counseling Center Assessment of Psychological Sympto ms 62 (CCAPS 62). Journal of Counseling Psychology, 58, 97 109. Lofthouse, N., & Yager Schweller, J. (2009). Nonsuicidal self injury and suicide risk among adolescents. Current Opinion in Pediatrics, 21, 641 645. Low, G., Jones, D., MacLeod, A., Power, M. & Duggan, C. (2000). Childhood trauma, dissociation, and self harming behavior: A pilot study. British Journal of Medical Psychology, 73, 269 278. Lynam, D. R., Miller, J. D., Miller, D. J., Bornovalova, M. A., & Lejuez, C. W. (2011). Testing the relati ons between impulsivity related traits, suicidality and nonsuicidal self injury: A test of the incremental validity of the UPPS model. Personality Disorders: Theory, Research, and Treatment, 2, 151 160. Lynch, T. R., & Cozza, C. (2009). Behavior therapy f or nonsuicidal self injury. In M. K. Nock, (Ed.), Understanding nonsuicidal self injury: Origins, assessments and treatment (pp. 221 250). Washington, DC: American Psychological Association. MacAniff Zila, L., & Kiselica, M. S., (2001). Understanding and counseling self mutilation in female adolescents and young adults. Journal of Counseling and Development, 79, 46 52. MacLaren, V. V. & Best, L. A. (2010). Nonsuicidal self injury, potentially addictive behaviors, and the five factor model in undergraduate s. Personality and Individual Differences, 49, 521 525. Maier, K. J., Goble, L. A., Neumann, S. A., Giggey, P. P., Suarez, E. C., & Waldstein, S. R. (2009). Dimensions across measures of dispositional hostility, expressive style and depression show some v ariation by race/ethnicity and gender in young adults. Journal of Social and Clinical Psychology, 28, 1199 1225. Muehlenkamp, J. J. (2005). Self injurious behavior as a separate clinical syndrome. American Journal of Orthopsychiatry, 75, 324 333.
140 Muehlen kamp, J. J. (2006). Empirically supported treatments and general therapy guidelines for non suicidal self injury. Journal of Mental Health Counseling, 28, 166 185. Muehlenkamp, J. J., Claes, L., Smits, D., Peat, C. M., & Vandereycken, W. (2011). Non suici dal self injury in eating disordered patients: A test of a conceptual model. Psychiatry Research, 188, 102 108. Muehlenkamp, J. J., Engel, S. G., Wadeson, A., Crosby, R. D., Wonderlich, S. A., Simonich, H., & Mitchell, J. E. (2009). Emotional states prece ding and following acts of non suicidal self injury in bulimia nervosa patients. Behaviour Research and Therapy, 47, 83 87. 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., Bradley, A. R., & Larsen, M. A. (2010). Abuse subtypes and nonsuicidal self injury: Preliminary evidence of complex emotion regulation patterns Journal of Nervous and Mental Disease, 198, 258 263. Najmi, S., Wegner, D. M., & Nock, M. K. (2007). Thought suppression and self injurious thoughts and behaviors. Behaviour Research and Therapy, 45, 1957 1965. National Center for Disease Control and Pr evention. (2012). Injury center: Violence prevention. Centers for Disease Control. Atlanta: GA. New Freedom Commission on Mental Health (2003). Achieving the Promise: Transforming Mental Health Care in America. Final Report. DHHS Pub. No. SMA 03 3832. Roc kville, MD Newman, C. F. (2009). Cognitive therapy for nonsuicidal self injury. In M. K. Nock (Ed.), Understanding nonsuicidal self injury: Origins, assessments and treatment, (pp. 201 219). Washington, DC: American Psychological Association. Nixon, M. K ., & Cloutier, P. F. (2005). Ottawa Self Injury Inventory. Unpublished manuscript, Centre for Youth and Society, University of Victoria, Victoria, Canada. Nixon, M. K., Cloutier, P. F., & Aggarwal, S. (2002). Affect regulation and addictive aspects of repe titive self injury in hospitalized adolescents.
141 Journal of American Academy of Child and Adolescent Psychiatry, 41, 1333 1341. Nixon, M. K. & Heath, N. L. (2009). Self injury in youth: The essential guide to assessment and intervention. New York NY : Routl edge. Nock, M. K. (2009). Why do people hurt themselves? New insights into the nature and functions of self injury. Current Directions in Psychological Science, 18 78 83. Nock, M. K., & Cha, C. B. (2009). Psychological models of nonsuicidal self injury. In M. K. Nock (Ed.) Understanding non suicidal self injury: Origins, assessment and treatment (pp. 65 77 ). Washington, DC: American Psychological Association. Nock, M. K., & Favazza, A. R. (2009). Nonsuicidal self injury: Definition and classification. In M. K Nock (Ed.), Understanding nonsuicidal self injury: Origins, assessments and treatment, (pp. 79 98). Washington, DC: American Psychological Association. Nock, M.K ., Joiner, T. E., Gordon, K. H., Lloyd Richardson, E., & Prinstein, M J. (2006). Non su icidal self injury among adolescents: Diagnostic correlates and relation to suicide attempts. Psychiatry Research, 144, 65 72. Nock, M. K., & Mendes, W. B. (2008). Physiological arousal, distress tolerance, and social problem solving deficits among adoles cent self injurers. Journal of Consulting and Clinical Psychology, 76, 28 38. Nock, M. K., & Prinstein, M. J. (2004). A functional approach to the assessment of self mutilative behavior. Journal of Counseling and Clinical Psychology, 72, 885 890. Nock, M K., & Prinstein, M. J. (2005). Contextual features and behavioral functions of self mutilation among adolescents. Journal of Abnormal Psychology, 114, 140 146. Nock, M. K., Prinstein, M. J., & Sterba, S. K. (2009). Revealing the form and function of self injurious thoughts and behaviors: A real time ecological assessment study among adolescents and young adults. Journal of Abnormal Psychology, 118, 816 827. Nock, M. K., Teper, R., & Hollander, M. (2007). Psychological treatment of self injury among adole scents. Journal of Clinical Psychology, 63, 1081 1089.
142 Paivio, S., & McCulloch, C. R. (2004). Alexithymia as a mediator between childhood trauma and self injurious behaviors. Child Abuse and Neglect, 28, 339 354. Pledge, D. S., Lapan, R. T., Heppner, P. P., Kivlighan, D., & Roehlke, H. J. (1998) Stability and severity of presenting problems at a university counseling center: A 6 year analysis. Professional Psychology: Research and Practice, 29, 386 389. Plener, P. L., Libal, G., Keller, F., Fegert, J. M., & Muehlenkamp, J. J. (2009). An international comparison of adolescent non suicidal self injury (NSSI) and suicide attempts: Germany and the USA. Psychological Medicine, 39, 1549 1558. Polk, E., & Liss, M. (2007). Psychological characteristics of self in jurious behavior. Personality and Individual Differences, 43, 567 577. Polk, E., & Liss, M. (2009). Exploring the motivations behind self injury. Counseling Psychology Quarterly, 22, 233 241. Porter, S. R., & Umbach, P. D. (2006). Student survey response rates across institutions: Why do they vary? Research in Higher Education, 77, 169 192. Prinstein, M. health risk behavior: Factors that alter or add to peer influence. Journal of Pe diatric Psychology, 26 (5), p. 287 295. Prinstein, M. J., Guerry, J. D., Browne, C. B., & Rancourt, D. (2009). Interpersonal models of nonsuicidal self injury. In M. K. Nock (Ed.) Understanding nonsuicidal self injury: Origins, assessments and treatment (pp 79 98). Washington, DC: American Psychological Association. Robins, C. J., & Chapman, A. L. (2004). Dialectical behavior therapy: Current status, recent developments, and future directions. Journal of Personality Disorders, 18, 73 89. Rodham, K., & Haw ton, K. (2009). Epidemiology and phenomenology of nonsuicidal self injury. In M. K Nock (Ed.), Understanding nonsuicidal self injury: Origins, assessments and treatment, (pp. 37 62). Washington, DC: American Psychological Association.
143 Ross, S., & Heath, N (2002). A study of the frequency of self mutilation in a community sample of adolescents. Journal of Youth and Adolescence, 31, 67 77. Ross, S., & Heath, N. (2003). Two models of adolescent self mutilation. Suicide and Other Life Threatening Behavior, 3 3, 277 287. Ross, S., Heath, N., & Toste, J. R. (2009). Non suicidal self injury and eating pathology in high school students. American Journal of Orthopsychiatry, 79, 83 92. Schaffer, D., & Jacobson, C. (2009). Proposal to the DSM V childhood disorder a nd mood disorder work groups to include non suicidal self injury (NSSI) as a DSM V disorder. Retrieved from http://www.dsm5.org/ProposedRevisions/Pages/proposedrevisio n.aspx?ri d=443 Schroeder, S. R., Oster Granite, M. L., Berkson, G., Bodfish, J.W., Breese, G. R... Wong, D. F. (2001). Self injurious behavior: Gene brain behavior relationships. Mental Retardation and Developmental Disabilities, 7, 3 12. Selby, E. A., A nestis, M. D., Bender, T. W., & Joiner, T. E. (2009). An exploration of the emotional cascade model in borderline personality disorder. Journal of Abnormal Psychology, 118, 375 387. Serras, A., Saules, K. K., Cranford, J. A., & Eisenberg, D. (2010). Self injury, substance use, and associated risk factors in a multi campus probability sample of college students. Psychology of Addictive Behaviors, 24, 119 128. Shannon, D. M., & Bradshaw C. C. (2002). A comparison of response rate, response time, and costs of mail and electronic surveys. The Journal of Experimental Education, 70, 179 192. Shenk, C. E., Noll, J. G., & Cassarly, J. A. (2010). A multiple meditational test of the relationship between childhood maltreatment and non suicidal self injury. Journal of Youth and Adolescence, 39, 335 342. Sher, L., & Stanley, B. (2009). Biological models of nonsuicidal self injury. In M. K Nock (Ed.), Understanding nonsuicidal self injury: Origins, assessments and treatment (pp., 99 116). Washington, DC: American Psy chological Association.
144 Shih, T H., & Fan, X. (2009). Comparing response rates in e mail and paper surveys: A meta analysis. Educational Research Review 4, 26 40. Slee, N., Arensman, E., Garnefski, N., & Spinhoven, P. (2007). Cognitive behavioral therap y for deliberate self harm. Crisis, 28, 175 182. Soet, J., & Sevig, T. (2006). Mental health issues facing a diverse sample of college students: Results from the college student mental health survey. National Association of Student Personnel Administrator s Journal, 43, 410 431. Stanford, S., & Jones, M. P. (2009). Psychological subtyping finds pathological, harm. Child Psychology and Psychiatry, 50, 807 815. Stanley, B., Sher, L., Wilson, S., Ekman R., Huang, Y., & Mann, J. J. (2009). Non suicidal self injurious behavior, endogenous opioids and monoamine neurotransmitters. Journal of Affective Disorders, 124, 134 140. Strong, M. (1998). The bright red scream: Self mutilation and the language of pai n. New York, NY: Viking Press. Svirko, E., & Hawton, K. (2007). Self injurious behaviors and eating disorders: The extent and nature of the association. Suicide and Life Threatening Behavior, 37, 409 421. Townsend, E., Hawton, K., Altman, D. G., Arensman, E., Gunnell, D., Hazell, P., House, A., & Van Heeringen, K. (2001). The efficacy of problem solving treatments after deliberate self harm: Meta analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Psychological Medicine, 31, 979 988. Trupin, E. W., Stewart, D. G., Beach, B., & Boesky, L. (2002). Effectiveness of a dialectical behavior therapy program for incarcerated female juvenile offenders. Child and Adolescent Mental Health, 3, 121 127. Tulloch A. L., Blizzard, L., & Pinkus, Z. (1997). Adolescent parent communication in self harm. Journal of Adolescent Health, 21, 267 275. U.S. Department of Education, National Center for Education Statistics. (2011). Digest of Education Statistics, 2010 (NCES 2011 015). Walsh, B. (2007). Clinical assessment of self injury: A practical guide. Journal of Clinical Psychology: In Session, 63, 1057 1068.
145 Walsh, B. W., & Rosen, P. (1985). Self mutilation and contagion: An empirical test. American Journal of Psychi atry, 142 119 120. Wedig, M. M. & Nock, M. N. (2007). Parental expressed emotion and adolescent self injury. Journal of American Academy of Child and Adolescent Psychiatry, 46, 1171 1178. Weismoore, J. T., & Esposito Smythers, C. (2010). The role of cog nitive distortion in the relationship between abuse, assault, and non suicidal self injury. Journal of Youth and Adolescence, 39, 281 290. Weitzman, E. R. (2004). Poor mental health, depression, and associations with alcohol consumption, harm, and abuse in a national sample of young adults in college. Journal of Nervous and Mental Disease, 192, 269 277. White, V. E., Trepal Wollenzier, H., & Nolan, J. M. (2002). College students and self injury: Intervention strategies for counselors. Journal of College C ounseling, 5, 105 114. Whitlock, J. (2010). Self injurious behavior in adolescents. PLoS Medicine, 7, 1 4. Whitlock, J., Eels, G., Cummings, N., & Purdington, A. (2009). Nonsuicidal self injury in college populations: Mental health provider assessment of prevalence and need. Journal of College Student Psychotherapy, 23, 172 183. Whitlock, J., Eckenrode, J., & Silverman, D. (2006). Self injurious behaviors in a college population. Pediatrics, 117, 1939 1948. Whitlock, J., & Knox, K. L. (2007). The relation ship between a self injurious behavior and suicide in a young adult population. Archives of Pediatric Adolescent Medicine, 7, 634 640. Whitlock, J., Purington, A., & Gershkovich, M. (2009). Media and the internet and non suicidal self injury. In M. K. Noc k (Ed.), Understanding nonsuicidal self injury: Origins, assessment and treatment (pp. 139 155). Washington, DC: American Psychological Association. Wilcox, H. C., Arria, A. M., Caldeira, K. K., Vincent, K. B., Pinchevsky, G. M., & Longitudinal predictors of past year non suicidal self injury and motives among college students. Psychological Medicine, 42, 717 726.
146 Wilkinson, P., & Goodyer, I. (2011). Non suicidal self injury. European Child and Adolescent Psychiatry, 20, 103 108. W ong, J. P. S., Stewart, S. M., Ho, S. Y., & Lam, T. H. (2007). Risk factors associated with suicide attempts and other self injury among Hong Kong adolescents. Suicide and Life Threatening Behavior, 37, 453 466. Yates, T. M. (2004). The developmental psyc hopathology of self injurious behavior: Compensatory regulation in posttraumatic adaptation. Clinical Psychology Review, 24, 35 74. Yates, T. M., Carlson, E. A., & Egeland, B. (2008). A prospective study of child maltreatment and self injurious behavior i n a community sample. Developmental Psychopathology, 20, 651 671. Yorgason, J. B., Linville, D., & Zitzman, B. (2008). Mental health among college students: Do those who need services know about them and use them? Journal of American College Health, 57, 1 73 181. You, J., & Leung, F. (2012). The role of depressive symptoms, family invalidation and behavioral impulsivity in the occurrence and repetition of non suicidal self injury in Chinese adolescents: A 2 year follow up study. Journal of Adolescence, 35, 385 395. Zivin, K., Eisenberg, D., Gollust, S. E., & Golberstein, E. (2009). Persistence of mental health problems and needs in a college student population. Journal of Affective Disorders, 3, 180 185.
147 BIOGRAPHICAL SKETCH Stacey Rice was born i n 1983 in Wayne, New Jersey. She spent her childhood in Wayne, and graduated from Wayne Valley High School in 2001. Following high school graduation, she moved to N ew Brunswick, NJ to obtain her b s University. While in her junior year of college, she had the opportunity to study at Macquarie University in Sydney, Australia. For three years following college graduation, Stacey worked as a mental health counselor in both inpatient and outpatient psyc hiatric facilities. In 2007, Stacey returned to school to obtain a m advocacy from Montclair State University. This multidisciplinary degree furthered he alth needs. Following graduation, Stacey was accepted to the University of rogram, and obtained her Master of Education in August 2011. While studying at the University of Florida, Stacey had the opportunity to work both i n the school system and outpatient mental health clinics. Stacey plans to complete her Ph.D in August 2013, after completion of her internship with the Baltimore City Public School System.