Date of Award
Spring 2005
Document Type
Dissertation - Restricted
Degree Name
Doctor of Philosophy (PhD)
Department
Psychology
First Advisor
Saunders, Stephen M.
Second Advisor
Wierzbicki, Michael
Third Advisor
Grych, John
Abstract
Attrition from psychotherapy with children and adolescents is a significant problem. Models addressing attrition were presented, including the Service Utilization Model (SUM). Youths with Severe Emotional Disturbance (SED) are a significant concern when considering attrition. Continuum-of-care programs, such as the Wraparound Milwaukee Program, have been created to address the high needs of SED youths. This study examined attrition from the Wraparound Milwaukee Program. Univariate relationships between dropping out and a number of demographic and clinical indicators were performed. A logistic regression analysis examined the effects that three types of predictor variables have upon attrition as predicted by the Service Utilization Model. A second set of logistic regression analyses examined whether the relationships that the predictor variables had with attrition would be moderated by the youths' early treatment response. The attrition rate for youths from the Wraparound Milwaukee Program was 29%. The family's race/ethnicity and the receipt of outpatient or in-home therapy during the first 6 months were found to be significant predictors of later attrition. When youths' early treatment response was included in the logistic regression analyses, it did not alter the impact that the predictor variables had upon attrition. The results showed that youths who dropped out of the program were more often from a minority group. They received less day treatment during the first 6 months. Conversely, youths who were placed in a group home or a residential treatment program in the first 6 months showed a higher likelihood of dropout. Early treatment response on externalizing symptoms did not interact with the other predictor variables regardless of whether it was examined as a continuous or categorical variable. The limitations and implications of this study were discussed. Recommendations for future studies of the Wraparound Milwaukee Program and attrition from continuum-of-care programs as a whole were made.