Risk Model-Guided Clinical Decision Support for Suicide Screening: A Randomized Clinical Trial

Area of Science:

• Clinical Informatics
• Psychiatry
• Public Health

Background:

• Effective suicide prevention necessitates robust risk identification, intervention, and follow-up protocols.
• Current risk identification methods, including self-reporting and clinical screening, have limitations.
• Statistical risk models offer potential but face integration challenges within clinical workflows, particularly concerning alert burden and clinical decision support (CDS) effectiveness.

Purpose of the Study:

• To evaluate the effectiveness of risk model-driven clinical decision support (CDS) in prompting suicide risk assessments.
• To compare interruptive versus noninterruptive CDS alert systems for suicide risk assessment.

Main Methods:

• A comparative effectiveness randomized clinical trial involving 561 patients across neurology divisions at an academic medical center.
• Patients were randomized to either interruptive or noninterruptive CDS systems designed to prompt further suicide risk assessment using a real-time, validated statistical suicide attempt risk model.
• Suicide risk assessment documentation was manually reviewed to determine the main outcome (decision to assess risk in person) and secondary outcomes (rates of suicidal ideation and attempts).

Main Results:

• Interruptive CDS resulted in significantly more decisions to screen for suicide risk (42%) compared to noninterruptive CDS (4%) and the prior year's baseline rate (8%).
• The odds ratio for decisions to screen with interruptive CDS versus noninterruptive CDS was 17.70 (95% CI, 6.42-48.79; P < .001).
• No documented episodes of suicidal ideation or attempts were observed in either treatment arm during the study period.

Conclusions:

• Interruptive clinical decision support (CDS) significantly increases the rate of documented suicide risk assessments.
• The findings support the use of interruptive CDS for prompting in-person suicide risk evaluations.
• Further large-scale trials comparing interruptive CDS with standard care are recommended to assess its effectiveness in reducing suicidal self-harm.