Associating factors of suicide and repetition following self-harm: A systematic review and meta-analysis of longitudinal studies

Background Longitudinal evidence for sociodemographic and clinic factors deviating risk for suicide and repetition following SH (self-harm) varied greatly. Methods A comprehensive search of PubMed, Web of Science, EMBASE, and PsycINFO was conducted from January 1st, 2010 to April 5th, 2022. Longitudinal studies focusing on examining associating factors for suicide and repetition following SH were included. PROSPERO registration CRD42021248695. Findings The present meta-analysis synthesized data from 62 studies published from Jan. 1st, 2010. The associating factors of SH repetition included female gender (RR, 95%CI: 1.11, 1.04–1.18, I2=82.8%), the elderly (compared with adolescents and young adults, RR, 95%CI: 0.67, 0.52–0.87, I2=86.3%), multiple episodes of SH (RR, 95%CI: 1.97, 1.51–2.57, I2=94.3%), diagnosis (RR, 95%CI: 1.60, 1.27–2.02, I2=92.7%) and treatment (RR, 95%CI: 1.59, 1.40–1.80, I2=93.3%) of psychiatric disorder. Male gender (RR, 95%CI: 2.03, 1.80–2.28, I2=83.8%), middle-aged adults (compared with adolescents and young adults, RR, 95%CI: 2.40, 1.87–3.08, I2=74.4%), the elderly (compared with adolescents and young adults, RR, 95%CI: 4.38, 2.98–6.44, I2=76.8%), physical illness (RR, 95%CI: 1.95, 1.56–2.43, I2=0), multiple episodes of SH (RR, 95%CI: 2.02, 1.58–2.58, I2=87.4%), diagnosis (RR, 95%CI: 2.13, 1.67–2.71, I2=90.9%) and treatment (RR, 95%CI: 1.36, 1.16–1.58, I2=58.6%) of psychiatric disorder were associated with increased risk of suicide following SH. Interpretation Due to the substantial heterogeneity for clinic factors of suicide and repetition following SH, these results need to be interpreted with caution. Clinics should pay more attention to the cases with SH repetition, especially with poor physical and psychiatric conditions