Determinants of psychiatric rehospitalization: a social area analysis.

The purpose of this paper is to develop an analytic framework and methodology for estimating the demand for psychiatric hospitalization in state facilities, either new admissions or readmissions of clients from the Community Services Boards (CSBs) of Virginia. The combination of community and organizational factors with inpatient characteristics for the CSBs can help identify pertinent predictors of psychiatric hospitalization. Review of the literature on psychiatric epidemiology shows that the predictors of psychiatric hospitalization can be classified into four major dimensions including community resources, socio-demographic factors, CSB client characteristics, and CSB organizational factors. It is postulated that the rates of first admissions and readmissions to state facilities are influenced by four dimensions. Furthermore, the variation in readmissions rates is contingent upon previous first admission rates and the average length of stays in state institutions while other conditions-such as community resources and demographics, CSB client mix, and CSB organizational factors-are simultaneously considered. The data were compiled from multiple sources. A path analytic model, using a linear structural relations program (LISREL VI), was validated. After the model was carefully fitted, the estimation equations for the rates of first admissions and readmissions were finalized. The estimates for psychiatric hospitalization, both first- and re-admissions, were made for each of the study CSBs.