Predicting time-to-independent-release from current level of functioning for psychiatric inpatients: a "survivor" analysis.

Non-optimal psychiatric discharges occur frequently and result in high costs. The Time-Sample Behavioral Checklist (TSBC) has been demonstrated to be the best method for determining successful independent discharge within 2 weeks of assessment for adult inpatients. This study examined the extent to which TSBC indexes and perceived dangerousness predict staff independent discharge-readiness judgments up to 6 months after initial assessment. Data from 22 acute and chronic/mixed units (N = 362) were analyzed using Cox proportional hazard regression. TSBC appropriate interpersonal interaction and appropriate behavior variability predicted shorter time-to-independent-release (TTIR). TSBC bizarre facial expressions and verbalizations predicted longer TTIR. Post hoc analyses suggest that acutely admitted inpatients perceived to be dangerous were discharged sooner than others-a finding that is likely attributable to differential psychotropic medication responsiveness. Implications are discussed for TSBC implementation for earlier identification of discharge-ready inpatients and for tailoring interventions to target behavior that predicts independent discharge success.