Meta-Analysis of Interventions to Prevent Suicide
According to the World Health Organization (WHO), more than 800,000 people die by suicide each year, and WHO has emphasized the critical need to identify interventions with proven efficacy for suicide prevention. This study conducted a meta-analysis of randomized controlled trials (RCTs) that compared the efficacy of various interventions versus control to prevent death by suicide among adults. Investigators searched several databases (i.e., EMBASE and MEDLINE) from inception through December 31, 2015, including RCTs that compared prevention strategies with control, focusing on data related to demographics, methods, outcomes, and risk of bias. After reviewing more than 8,000 citations, 72 RCTs and 6 pooled analyses met inclusion criteria, with more than half being published in the last 10 years. While most trials studied interventions targeted at individuals with known risk factors for suicide, some trials evaluated interventions in a general or primary care population. Of the studies included in this review, 29 RCTs reported on complex psychosocial interventions, with 3 reporting on the WHO Brief Intervention and Contact (BIC) intervention. The WHO BIC intervention includes an educational session on suicide prevention followed by regular contact with a trained provider (phone or in-person) for up to 18 months. Other interventions that were examined included trials of lithium and cognitive behavioral therapy (CBT), among others.
- The WHO BIC intervention was associated with significantly lower odds of death by suicide. No other suicide prevention intervention showed a statistically significant effect in reducing death by suicide.
- Studies of CBT and lithium, while suggestive of comparable effects in reducing death by suicide, enrolled fewer patients and results did not reach statistical significance.
- Two national VA studies involving suicide prevention interventions are currently underway:
- Randomized study (via the Cooperative Studies Program) of the protective effects of lithium, which is planning to enroll nearly 2,000 participants with bipolar disorder or depression who have survived a recent suicide attempt or were hospitalized specifically to prevent suicide.
- REACH-Vet national implementation study of prediction models and care management for suicide prevention.
- Many studies were at risk for bias in their assessment of suicide, potentially obscuring the effect.
- For many interventions there were insufficient data (e.g., limited trials and/or small sample sizes) to draw any definitive conclusions about their efficacy.
- Because this review did not include patient-level data, investigators were unable to explore potential moderators or mediators of the efficacy of suicide prevention interventions.
This work was supported by the VA National Center of Patient Safety, Center of Inquiry Program. Dr. Shiner is supported by an HSR&D Career Development Award. All authors are part of the VAMC in White River Junction, VT.
Riblet N, Shiner B, Young-Xu Y, and Watts B. Strategies to Prevent Death by Suicide: A Meta-Analysis of Randomized Controlled Trials. British Journal of Psychiatry. June 2017;210(6):396-402.