Effectiveness and Implementation of a Clinician Decision Support System to Prevent Suicidal Behaviors

NCT05671133 | Recruiting DMC FDA Device

• 1 other identifier: 2022P002689
• Study Type: interventional
• Target: 4,000
• Locations: 1 country
• Sites: 1

Brief Summary

The primary aim of this project are to evaluate a comprehensive, practice-ready, and deployment-focused strategy for improving the prediction and prevention of suicide attempts among a sample of 4,000 patients presenting to an ED with a psychiatric concern. The first aim is to evaluate the effects of providing information about risk of patient suicidal behavior to ED clinicians. The investigators hypothesize that patients randomly assigned to have their clinician receive their risk score will have a lower rate of suicide attempts during 6-month follow-up and that this effect will be mediated by changes in clinician decision-making.

Conditions

Suicide; Suicide, Attempted

Outcome Measures

Primary Outcomes (1)

• Suicide attempt

  The primary endpoint is the occurrence (0,1) of a suicide attempt during the 6 months following randomization. The outcome will be measured as a single binary variable, indicating whether a participant experienced at least one suicide attempt during the follow-up period. This outcome will be ascertained using participant self-report and EHR documentation. Self-reported suicide attempts will be assessed using questions from the Self-Injurious Thoughts and Behaviors Interview (SITBI), a widely used measure of suicidal thoughts and behaviors with strong reliability and validity and one of the measures recommended in the PhenX Toolkit. EHR review will further identify documented suicide attempts during follow-up. Information from both sources will be used to determine a single outcome value for each participant. The singular outcome measure is the occurrence of at least one suicide attempt during follow-up (Yes/No; 0,1) as identified by either source (self-report or EHR).

  6-months

Secondary Outcomes (1)

• Suicide attempt

  1-month

Study Arms (2)

Experimental

EXPERIMENTAL

Investigators will use previously developed EHR-derived machine- learning-based model to generate predicted probability risk scores of 1-month suicide attempt for each patient. The EHR-derived risk score will be combined with the patient self-report data (initially from the 20-question screener) to generate composite predicted risk scores (based on the most updated ML model developed by investigators via the ensemble ML Super Learner \[SL\] method) for the patient within minutes of the patient completing the self-report survey. Risk scores will be provided to the clinician while the patient is in the ED. This feedback includes information about the patient's relative risk compared to other psychiatric patients seen in the ED and the risk factors contributing to this level of risk.

Diagnostic Test: Clinician Decision Support Tool

Control

NO INTERVENTION

Patient's clinician is not given Clinician Decision Support Tool (care as usual)

Interventions

Clinician Decision Support Tool DIAGNOSTIC_TEST

Clinician Decision Support Tool that provides information about patient's statistical probability of suicide attempt in next 1 month

Experimental

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Adult status (≥18 years-old);
• Willing to provide an email address;
• Presentation at the APS

You may not qualify if:

• Inability to understand the study procedures and provide informed consent such as those with gross cognitive impairment (including florid psychosis), intellectual disability, dementia, acute intoxication, or the presence of extremely agitated or violent behavior

Sponsors & Collaborators

• Massachusetts General Hospital: lead

Study Sites (1)

Massachusetts General Hospital

Boston, Massachusetts, 02114, United States

RECRUITING

MeSH Terms

Conditions

Suicide; Suicide, Attempted

Condition Hierarchy (Ancestors)

Self-Injurious Behavior; Behavioral Symptoms; Behavior

Central Study Contacts

Matthew Nock, PhD

CONTACT

617-496-4484; nock@wjh.harvard.edu

Amy Ahn, PhD

CONTACT

yahn@fas.harvard.edu

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: PARTICIPANT
• Purpose: PREVENTION
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Research Scientist

Study Record Dates

• First Submitted: December 16, 2022
• First Posted: January 4, 2023
• Study Start: February 25, 2025
• Primary Completion (Estimated): May 31, 2028
• Study Completion (Estimated): September 28, 2028
• Last Updated: June 22, 2026

Record last verified: 2026-06

Data Sharing

• IPD Sharing: Will share
• Shared Documents: STUDY PROTOCOL, SAP
• Time Frame: By study end

Locations

US (1)