Predictive Validity of the Columbia-Suicide Severity Rating Scale among a Cohort of At-risk Veterans

Objective: To examine the psychometric properties of the Columbia- Suicide Severity Rating Scale (C-SSRS) among a cohort of veterans identified to be at risk for suicide. Method: Convergent, divergent, and predictive validity of the C-SSRS were examined using secondary data from a study of veterans who presented to the psychiatric emergency room (n = 237). Data were collected 1 week, 1 month, 3 months, and 6 months following emergency room discharge. Results: Results demonstrated good evidence for convergent and divergent validity. Baseline intensity subscale scores predicted actual and interrupted attempts, and any behavior for all available data and for the subsample who had 6-month follow-up data, and also predicted preparatory behavior in the full sample. Baseline severity subscale scores predicted preparatory behavior and any behavior for the full sample and for those with 6 months of follow-up, and actual attempts for the full sample. Severity of ideation cutoff scores was significantly associated with actual attempts and any behavior in both samples. Adding a prior behavior criterion demonstrated similar results. Conclusions: Findings suggest that the C-SSRS is a psychometrically sound measure that can be used to augment suicide risk assessment with veterans who are already identified to be at risk for suicide.