Scholars@Duke publication: A scoping review of psychosocial suicide prevention interventions among adults: Clinical trial factors, predictors, mediators, and moderators in randomized controlled trials.

A scoping review of psychosocial suicide prevention interventions among adults: Clinical trial factors, predictors, mediators, and moderators in randomized controlled trials.

Publication , Journal Article

Kearns, JC; Polizzi, CP; Naganuma-Carreras, J; Bourassa, KJ; Halverson, TF; Perskaudas, R; Kimbrel, NA; Interian, A; Bossarte, RM; DeBeer, BB ...

Published in: Clin Psychol Rev

August 2025

Published version (DOI) Link to item

OBJECTIVE: There has been a proliferation of psychosocial interventions to prevent suicide over the past several decades with varying levels of benefit. We conducted a comprehensive scoping review to synthesize our knowledge of the effectiveness, strengths, and limitations of these interventions among adults. We focused on important aspects of randomized controlled trials (RCTs), including internal and external validity, outcome measurement, and exclusion criteria. We examined predictors, mediators, and moderators associated with suicide outcomes. METHODS: A systematic search was conducted for RCTs of psychosocial suicide prevention interventions with adults, following PRISMA-ScR guidelines. RESULTS: We identified 141 publications and 131 were unique RCTs. Publications had some concerns with risk of bias (internal validity); publications had moderate or high external validity. A quarter of RCTs excluded older adults (60 years or older) and individuals with psychotic symptoms/features and bipolar disorder/symptoms despite the fact that these groups are at elevated risk for suicide death. Forty-one publications (29.1 % of total publications) examined predictors, mediators, and moderators of suicide outcomes. Sixteen (11.3 % of total publications) publications included predictors, two (1.4 %) included mediators, 11 (7.8 %) included posttreatment mediators, and 16 (11.3 %) included moderators. Participant gender/sex, marital status, and race did not moderate treatment effects. Mixed results may be due to insufficient statistical power to test predictor, mediator, and moderator hypotheses. CONCLUSION: This review highlighted RCT aspects (e.g., exclusion of older adults and individuals with serious mental illness, improving suicide outcome measurement) that may be beneficial to consider in future RCTs. Additional research on predictors, mediators, and moderators is needed.