Psychotic-like experiences and non-suicidal self-injury in the community population: a systematic review and...

The aim of the study was to systematically examine the association between psychotic-like experiences (PLEs) and non-suicidal self-injury (NSSI) in community populations through meta-analysis, with a focus on prospective evidence regarding this association and potential psychosocial correlates.

A comprehensive search was conducted in PubMed, PsycINFO, and Embase from inception to August 2025. Studies reporting quantitative data on the association between PLEs and NSSI in community populations were included. A random-effects meta-analysis was performed to estimate the pooled odds ratio (OR) with 95% confidence interval (CI). This association was further examined through a preliminary prospective meta-analysis of two longitudinal studies reporting adjusted odds ratios for the prediction of incident NSSI by baseline PLEs. Potential sources of heterogeneity were explored by exploratory subgroup analyses across study design, geographical region, gender composition, NSSI assessment methods, and assessment time window match between PLEs and NSSI.

Eight studies involving 19,719 participants were included in the meta-analysis, and an additional eleven studies were included in the narrative synthesis. Individuals with PLEs had a significantly increased odds of NSSI compared to those without PLEs (OR = 3.09, 95% CI [2.05-4.66], p < 0.001). A prospective analysis of two longitudinal studies indicated that baseline PLEs significantly predicted incident NSSI at follow-up (OR = 2.17, 95% CI [1.56-3.02]). The estimated prevalence of PLEs and NSSI in community populations was 11.0% (95% CI [7.9%-15.2%]) and 6.9% (95% CI [4.3%-11.1%]), respectively. Narrative synthesis suggested that depressive symptoms, childhood adversity, bullying victimization, and pandemic-related stress were associated with higher odds of NSSI, whereas social support was associated with lower odds.

The findings demonstrate a robust association between PLEs and NSSI in community populations. Preliminary prospective evidence from a limited number of longitudinal studies further suggests that baseline PLEs may predict incident NSSI at follow-up, although this finding warrants confirmation in future research.