Non-suicidal self-injury and other key predictors of suicide deaths among children and adolescents in England: a population-based case-control study

Abstract Background Suicide is a leading cause of mortality in young people globally. A history of prior self-harm, with or without suicidal intent, is associated with increased risks of future suicide attempts (SAs) or suicides. While non-suicidal self-injury (NSSI) is likely a predictor of SA, its particular association to suicide deaths remains unexplored. Methods This is a population-based case-control study in England. Cases were all individuals who died by suicide before their 18th birthday ( n = 240), between April 2019 and March 2021. Controls ( n = 240) were age- and sex-matched individuals who died from other unexpected causes. Conditional logistic regression was used to estimate odds ratios for self-harm history and other risk factors, adjusting for ethnicity and socioeconomic deprivation. A fully adjusted multivariable model was formed to estimate associations between each risk factor and suicide death, adjusting for all measured factors. Results A history of NSSI was strongly associated with subsequent suicide death (partially adjusted OR, 27.8; 95% CI, 9.5–81.0; fully adjusted OR, 9.2; 95% CI, 2.1–39.5). The association remained significant in sensitivity analyses restricted to those with only NSSI history without SA (partially adjusted OR, 7.7; 95% CI, 3.6–16.5). History of SA was also significant (adjusted OR, 64.2; 95% CI, 8.8–468; fully adjusted OR, 12.3; 95% CI, 0.4–357). Notably, 42.6% of suicides occurred in individuals without previous self-harm, and 71% of suicides occurred on the first documented act of intentional self-injury or self-poisoning. MDMA was recorded in 14 suicide cases, and alcohol use in 34 suicide cases, regardless of whether the intake resulted in overdose. Fully adjusted model found that known NSSI history, known suicidal ideation history, and bullying victimization remained independently associated with suicide death. Conclusions NSSI and SA are both major predictors for suicide in young people. Risk assessments should aim to include all significant factors. These findings highlight the need for holistic clinical screening for all forms of self-harm regardless of intent, indicated suicide prevention intervention for those identified, and enhanced access to mental health services.