The Functions of Nonsuicidal Self-Injury in an Adolescent Clinical Sample: Frequency, Structure, and Psychological Correlates

Abstract

Nonsuicidal self-injury (NSSI; e.g., cutting and burning) is most prevalent among adolescents, especially in clinical populations. Understanding the functions of NSSI can help clarify the behavior's etiology and treatment. Although research has begun to examine common functions of NSSI, there have been three major shortcomings to this literature: (a) inadequate assessment of the full range of NSSI functions, (b) use of non-validated assessment instruments, and (c) a limited examination of the psychological correlates of NSSI functions. The current study addressed these limitations through the use of a valid, comprehensive measure of NSSI functions and thorough measurement of diagnostic and clinical correlates of NSSI functions. In addition to supporting previous findings on the frequency and factor structure of NSSI functions, a central goal of the current project was to examine how NSSI functional endorsement varies for self-injurers with externalizing versus internalizing psychopathology, and for injurers with versus without borderline personality disorder (BPD). Consistent with previous research, the current study found that the most common NSSI functions were affect regulation and self-punishment. In addition, results support a two-factor structure of NSSI functions: (a) intrapersonal - self-reinforcing (e.g., affect regulation) and (b) interpersonal - other-reinforcing (e.g., interpersonal influence). In regard to psychological correlates, greater endorsement of intrapersonal functions was associated with internalizing disorders (i.e., anxiety disorders, depressive disorders, and bulimia) and suicidal ideation, as well as with emotion and personality correlates (i.e., negative emotionality, impulsive urgency, loneliness, and self-derogation), even when controlling for NSSI severity (i.e., lifetime NSSI methods and frequency). Greater endorsement of interpersonal functions was associated with distress disorders (i.e., generalized anxiety disorder, depressive disorders, and posttraumatic stress disorder) and BPD, even when controlling for NSSI severity. Treatment implications of this research are discussed.