A Risk Calculator to Predict Suicide Attempts Among Individuals With Early-Onset Bipolar Disorder.

Area of Science:

• Psychiatry
• Clinical Psychology
• Epidemiology

Background:

• Bipolar disorder in youth is associated with a high risk of suicide attempts.
• Accurate prediction of suicide risk is crucial for timely intervention in early-onset bipolar disorder.

Purpose of the Study:

• To develop and validate a one-year risk calculator (RC) for predicting individualized suicide attempt risk in early-onset bipolar disorder.
• To identify key clinical predictors readily available in practice for suicide risk assessment.

Main Methods:

• Utilized data from 394 youth in the longitudinal Course and Outcome of Bipolar Youth (COBY) study.
• Trained a risk calculator using boosted multinomial classification trees and calibrated predictions with Platt scaling.
• Validated the RC on an independent half of the sample, assessing suicide attempts using the A-LIFE scale.

Main Results:

• The final model included ten predictors, with age of mood disorder onset, non-suicidal self-injurious behavior, and current age being most influential.
• The risk calculator achieved an AUC of 0.82, indicating good predictive performance.
• Recent (past 6 months) presence of risk factors reliably predicted higher suicide attempt risk compared to lifetime history.

Conclusions:

• The developed risk calculator shows promise as a clinical and research tool for identifying individuals at high risk for suicide attempts in early-onset bipolar disorder.
• This tool can aid in prospective identification of high-risk periods, enabling targeted interventions.
• Further validation and implementation in clinical settings are warranted.