Characteristics Associated with Outcome in a Community Mental Health Partial Hospitalization Program.

This study examined characteristics of patients in a community mental health partial hospitalization program to determine what, if any, relationship existed between those for whom treatment was considered relatively successful and those for whom treatment was considered less successful. The study sought to improve the utilization of the partial hospitalization modality by establishing relationships that could be used to predict treatment outcome and program suitability for individual patients. The sample was 519 admissions to a partial hospitalization program in a large Midwestern community mental health center. An associational one group pretest-posttest design was employed to determine if significant associations existed for seventeen patient characteristics and eight treatment outcome groups. Successful outcome groups included those patients with a Goal Attainment Scale at or above the program average; an improvement on the Kincannon, Mini-Mult profile at discharge; or no state hospitalization within two years following discharge, in circumstances both with and without ongoing treatment. Less successful outcome groups included those patients who dropped out of treatment in less than thirty days, returned to the state hospital within two years, obtained negative Goal Attainment scores, or did not have an improved Kincannon profile. Descriptive, chi-square, and stepwise logistic regression statistics were employed in the computer analysis of the data. The study found twelve of seventeen characteristics singularly associated with treatment outcome. Strongest associations occurred for the variables previous state hospitalizations, age at admission to the program, age at onset of mental illness, intelligence, and mother/father mentally ill. Unique patterns of variables were found for each outcome group which, when considered interactively, discriminated successful from less successful patients. The most success occurred among patients who had clearly identifiable precipitants, were diagnosed schizophrenic, had an average or above intelligence, a longer length of stay in the program, were White females with no previous psychiatric hospitalizations, had no parental incidence of mental illness or substance abuse, were older at age of admission to the program and at age of onset. Finally, indices for predicting success in partial hospitalization programming were developed.Ph.D.Social sciences educationUniversity of Michiganhttp://deepblue.lib.umich.edu/bitstream/2027.42/160049/1/8412262.pd