Psychological subtyping finds pathological, impulsive, and 'normal' groups among adolescents who self-harm

Background: Research to date suggests that as many as 12-15% of young people engage in self-harm behaviour; however, the current understanding of the psychological basis of adolescent self-harm is limited. The objective was to determine whether adolescents who self-harm are a psychologically homogenous group. It was hypothesised that psychological subtypes would exist and these groups would report different rates of self-harm. Method: Nine hundred and forty-four school students aged 11 to 19 and 166 first-year psychology students aged 21 or younger completed a self-report questionnaire. Participants were aged 11 to 21 (mean = 15.4, SD = 2.1). Sixty-two percent of the sample were female (n = 692). Students were allocated to psychologically distinct groups. Rates of self-harm were compared for the psychological subtypes of self-harmers. Results: Two hundred and thirty-four participants reported lifetime self-harm (21.1%; 95% CI 19-23%) and 78 reported recent self-harm (7.0%; 95% CI 6.7-7.3%). The present study identified three psychologically quite distinct groups of adolescents within those who reported self-harm - a psychologically pathological group, a psychologically 'normal' group, and an impulsive group. The pathological group reported the highest rate of recent self-harm (50.9%); the psychologically 'normal' and impulsive groups reported similar rates of self-harm (28.7% and 24.6%, respectively). Conclusions: Adolescents who self-harm are not a psychologically homogenous group. One pathological subtype of self-harmers appears to most closely reflect a number of the psychological and social factors previously associated with self-harm. However, a large proportion of the sample was allocated the psychologically 'normal' subtype. This finding highlights the importance of psychological screening of adolescents presenting for treatment for self-harm as subtypes of self-harmers may require disparate strategies for intervention. Further research is required in order to identify appropriate treatment strategies for each subtype.