Predictors of psychiatric hospitalization: a multivariate analysis.

Abstract

Inpatient treatment continues to be the most expensive form of mental health service. This study sought to improve the methodological weaknesses, e.g., poor statistical controls, in the literature by using multivariate statistics to predict hospitalization. Results revealed that aftercare, i.e., outpatient treatment, is an important factor in reducing the utilization of inpatient resources, even when controlling for demographic and psychiatric history variables. Further, background characteristics, while easily measured, are not important predictors of hospitalization.