Mortality in adolescents after therapeutic intervention for self-harm: A systematic review and meta-analysis

Self-harm in adolescents is an international concern. Evidence highlights that therapeutic intervention (TI), such as cognitive behaviour therapy informed treatments, after self-harm leads to reduced self-harm repetition. However, there is no prior literature about the effects of TI on future mortality in adolescents. We examined the effect of TI on mortality rates in adolescents across RCTs.

This review was reported in accordance with PRISMA guidance. MEDLINE, EMBASE, PsycINFO, and Cochrane Library were searched to 19 June 2024. Two authors independently screened titles, abstracts, and full texts against pre-defined criteria. RCTs were included if they compared a TI versus a comparator in adolescents up to 18 years with at least one prior self-harm episode. There was no lower age limit. For the pooled effect size of mortality, the DerSimonian-Laird method was used, and a random effects model for self-harm and suicide attempts. The primary outcome was intra or post-trial mortality in adolescent post TI, and the effect of TIs on self-harm including attempted suicide episodes were secondary outcomes. Analyses were done in Stata.

Twenty-four trials of TIs consisting of 3470 randomised adolescents were included. The pooled risk difference for mortality of participants in the TI group was 0.002 (95% CI -0.003 to 0.008, p = 0.42). There were 6 deaths in the TI group compared to 15 deaths in the comparator group. The pooled risk difference for TI on repeat self-harm was -0.07 (95% CI -0.132 to -0.007, p = 0.028), and -0.05 (95% CI -0.086 to -0.007, p = 0.022) for suicide attempts compared to comparator.

This review found no significant impact of TIs on future mortality in adolescents. We also demonstrated that TIs can reduce suicide attempts which can lead to substantial benefits for adolescents, families, and clinical services.