Evaluation of Clinical Prognostic Models for Suicide Attempts After a Major Depressive Episode

Using data from a 2-year prospective study of 304 depressed subjects, a series of Cox proportional hazard regression models were developed to predict future suicide attempt. The models were evaluated in terms of discrimination (the ability to rank subjects in order of risk), calibration (accuracy of predicted probabilities of attempt), & sensitivity & specificity of risk group stratification based on cross-validated predicted probabilities. Although an additive model with past attempt, smoking status, & suicide ideation achieved 75% sensitivity & specificity, models that performed best in terms of discrimination included interactions between predictor variables. (35 refs.)