Factors associated with unplanned discharge from psychiatric day treatment programs. A multicenter study

This study describes and evaluates the adequacy of sociodemographic and clinical descriptors as potential predictors of unplanned discharge from and completion of psychiatric day treatment programs. A 2-year retrospective chart review was completed on all patients (N = 327) attending three university-affiliated day treatment programs. Statistical comparisons were made between those patients who completed and those who had unplanned discharge. Logistic regression was used to generate a predictive model of unplanned discharge using identified variables. The rate of unplanned discharge was 54%. Factors associated with program completion were diagnoses of major depression or posttraumatic stress disorder, a history of completing a prior day treatment program, and higher education levels. Active substance abuse and a history of three or more inpatient admissions were associated with unplanned discharge. The predictive model was able to correctly classify 71% of patients completing day treatment programs and 43% of patients with unplanned discharges. Traditional demographic and clinical variables contribute differentially to program completion/noncompletion. Given the relatively poor predictability of unplanned discharge in this study, patient selection practices based on these factors alone may be somewhat limiting. Interactive effects of patient and program characteristics need to be addressed to improve program outcome.