Improving Screening and Detection of Suicide Risk: Results from the Emergency Department Safety Assessment and Follow-up Evaluation (ED-SAFE) Effectiveness Trial

Background: A fundamental step in building evidence-based clinical guidelines promoting suicide prevention in the emergency department setting is to establish if universal suicide risk screening is feasible and effective at improving detection of risk during routine care.

Objective: To test whether implementing universal suicide risk screening protocols in the emergency department significantly increases the percentage of patients screened during routine care and improves suicide risk detection.

Methods: The Emergency Department Safety Assessment and Follow-up Evaluation (ED-SAFE) was conducted at eight diverse emergency departments in the continental United States. It used an interrupted time series design with three sequential phases: Treatment as Usual (Phase 1), Universal Screening (Phase 2), and Universal Screening + Intervention (Phase 3). In Phase 2, universal suicide risk screening was implemented during routine clinical care using performance improvement methods. In Phase 3, universal screening was continued and a telephone intervention delivered after the emergency department visit was added. Screening for any intentional self-harm ideation/behavior and detection of intentional self-harm ideation/behavior were the primary outcomes. Chart reviews and patient interviews established whether a screening for intentional self-harm was documented (screening), whether the individual endorsed intentional self-harm ideation or behavior (detection), and whether the intentional self-harm was suicidal in nature. Chi-square tests and generalized estimating equations were performed.

Results: Documented screenings for intentional self-harm rose from 26% (Phase 1) to 85% (Phases 2 and 3) of all patients, χ² (2, N=236,789)=71,000, P<0.001. Detection of self-harm ideation/behavior rose from 2.9% to 5.7%, χ² (2, N=236,789)=902, P<0.001. Interviews confirmed that 75% of intentional self-harm detected was suicidal in nature. Time series plots revealed significant site heterogeneity.

Conclusions: Using a performance improvement approach, a dramatic improvement in suicide risk screening and detection during routine care was feasible, though considerable site variability existed. Further exploration into factors associated with screening success and whether the intervention introduced in Phase 3 led to reduced suicidal behavior is planned.