Suicide Prevention & Digital Interventions: A Meta-Analysis and Clinical Implications

Background and Purpose: Over the past decade there has been a proliferation of digital-based mobile interventions in suicide prevention. These interventions hold significant promise in addressing known barriers for individuals experiencing suicidality in receiving treatment, including affordable cost, wider availability, and less stigma. Prior meta-analyses in this field are limited and the results have been mixed. The current systematic review and meta-analysis aimed to provide an up-to-date summary of the effectiveness of digital interventions designed to address suicidality as well as provide key subgroup analyses relating to variables of age, gender, and control group type.

Methods: A search for English-language peer-reviewed articles was conducted using five health-related databases. The searches were restricted to research articles published before January 1, 2022. Randomized controlled trials of the effectiveness of suicide prevention interventions that used digital-based mobile technology (apps/online programs) were included. Interventions needed to have been designed or adapted to specifically address some aspect of suicidality, and a primary outcome of a suicide variable needed to be reported. PRISMA reporting guidelines were adhered to and the review’s protocol was preregistered with PROSPERO (CRD42021230901). Three researchers extracted data using a custom spreadsheet, and risk of bias within selected studies was assessed using the Cochrane risk-of bias tool for randomized trials.

Extracted data were analyzed using random-effects Restricted Maximum Likelihood model in Stata 17. Suicidal ideation was used as the primary outcome and the analysis focused on the time-point of post treatment. Subgroup analyses were planned a priori and included comparisons of the age category of study participants (adolescents vs. adults), missing data handling (ITT vs. non-ITT), and control group (TAU vs. active control). A meta-regression analysis was performed to examine the relation between intervention effect and gender.

Results: The search yielded 4317 articles. After deletion of duplicates, and screening according to the inclusion and exclusion criteria, 16 articles were identified. Risk of bias results found studies to largely be moderate-to-low quality. The random-effects model indicated a small but significant effect of treatment upon suicidal ideation, k=16, g=0.11 (95% CI: 0-0.23), p=.049. The test of heterogeneity showed that the effect sizes among studies varied significantly. Subgroup analyses found the interventions to have a significant effect on adults (g=0.15, 95% CI: 0.03, 0.28, p=.01) but not adolescents (g=-0.11, 95% CI: -0.26, 0.04, p=.17). The online interventions showed better effects compared to waitlist control (g=0.17, 95% CI: 0.04, 0.30, p=.01) but no significant difference compared to active control (g=-0.04, 95% CI: -0.11, 0.04, p=.36). No significant relation was observed between the proportion of gender (male) and the effect size between studies.

Conclusions and Implications: This study significantly contributes to the evidence base confirming the effectiveness of digital interventions for suicide prevention. This presentation will present recommendations for clinicians and policy makers to consider the adoption of these digital interventions in practice. In addition, research recommendations and needs to examine the effectiveness of these interventions across vulnerable subgroups and guide further intervention development in this field will be explored.