TY - JOUR
T1 - Exploration of the use of non-census indicators for prediction of mental health admissions
AU - Royse, D. D.
N1 - Copyright:
Copyright 2017 Elsevier B.V., All rights reserved.
PY - 1980
Y1 - 1980
N2 - The purpose of this exploratory study was to determine whether variables generally collected by state agencies could be used to predict admissions to state inpatient psychiatric facilities and outpatient admissions to community mental health centers. With the county as the unit of analysis this research retrospectively examined 20 independent variables for each of Ohio's 88 counties during a five-year period, 1973-1977. Some of the independent variables employed in this study were retail alcohol sales, average weekly earnings, suicides, divorces, delinquency cases, dependency and neglect cases, and arraignments. Factor analysis of the independent variables consistently produced three unnamed factors in each year. When multiple regression was conducted with these three factors an average of 34 percent of the variation in inpatient admissions and five percent of the variation in outpatient admissions were explained. A stepwise buildup regression of the variables produced five predictors that accounted for an average of 46 percent of the variation in inpatient admissions. These predictors and the average variation explained were aid to dependent children (28 percent), death (seven percent), unemployment (five percent), dropouts (three percent), and divorces (three percent). Aid to dependent children and unemployment accounted for an average of 37 percent of the inpatient admissions variance during the five-year study period. No consistent pattern of variables explained any sizeable amount of variation in outpatient admissions. It was concluded that while the large amount of explained variation in inpatient admissions was impressive from a theoretical perspective, the results were not good enough in a practical sense to be used by planners as a model in planning for future admissions.
AB - The purpose of this exploratory study was to determine whether variables generally collected by state agencies could be used to predict admissions to state inpatient psychiatric facilities and outpatient admissions to community mental health centers. With the county as the unit of analysis this research retrospectively examined 20 independent variables for each of Ohio's 88 counties during a five-year period, 1973-1977. Some of the independent variables employed in this study were retail alcohol sales, average weekly earnings, suicides, divorces, delinquency cases, dependency and neglect cases, and arraignments. Factor analysis of the independent variables consistently produced three unnamed factors in each year. When multiple regression was conducted with these three factors an average of 34 percent of the variation in inpatient admissions and five percent of the variation in outpatient admissions were explained. A stepwise buildup regression of the variables produced five predictors that accounted for an average of 46 percent of the variation in inpatient admissions. These predictors and the average variation explained were aid to dependent children (28 percent), death (seven percent), unemployment (five percent), dropouts (three percent), and divorces (three percent). Aid to dependent children and unemployment accounted for an average of 37 percent of the inpatient admissions variance during the five-year study period. No consistent pattern of variables explained any sizeable amount of variation in outpatient admissions. It was concluded that while the large amount of explained variation in inpatient admissions was impressive from a theoretical perspective, the results were not good enough in a practical sense to be used by planners as a model in planning for future admissions.
UR - http://www.scopus.com/inward/record.url?scp=85040278073&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85040278073&partnerID=8YFLogxK
M3 - Article
AN - SCOPUS:85040278073
JO - [No source information available]
JF - [No source information available]
ER -