Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Health Outcomes and Policy Research

Research Advisor

Teresa M. Waters, Ph.D.


Richard P. Barth, Ph.D. Theora A. Evans, Ph.D. Song Hee Hong, Ph.D. David M. Mirvis, Ph.D.


Adolescents, Behavioral health, Hierarchical linear modeling, Multisystemic therapy, Outcomes


Intensive in-home services (IIHS) work with families toward building skills and resources to better manage their children with behavioral difficulties. Factors that impact long-term outcomes following IIHS are not well understood. This study examined the relative importance of youth demographic and clinical characteristics, family history variables, characteristics of program participation, and organizational-level factors in explaining placement stability, educational progress, and contact with legal authorities one year following discharge from IIHS.

The sample included all youth who received IIHS, were discharged between January 1, 2001 and December 31, 2006, and completed a follow up at one-year post-discharge (n = 2,649) from a large provider of IIHS. Hierarchical linear modeling (HLM) was employed to test the relationship between case characteristics and outcomes, as youth were nested within offices. The non-experimental nature of the study demanded close attention to the issue of response bias. Logistic regression was used to model the probability of responding to the postdischarge survey, with the predicted values used in the HLM analysis to correct for response bias.

Adolescent males were found to have significantly lower odds of placement with their family, higher odds of contact with legal authorities and out of home placement, and lower odds of a positive composite outcome than adolescent females. Youth of both genders with antisocial behavior had significantly lower odds of positive outcomes. Length of service was significantly associated with higher odds of negative outcomes for all dependent measures except contact with legal authorities. This last finding was unexpected and may be a function of unmeasured risk factors that affect both length of service and long-term outcome. Office-level characteristics were not found to be significant predictors in most of the models.

Recommendations for improvement of the intensive in-home services program included focusing resources on high-risk clinical characteristics rather than changing organizational attributes such as size and turnover. Although a high rate of turnover potentially has other negative consequences, finding more effective ways to treat antisocial behavior, particularly in males ages 10 to 16, may be the activity most likely to secure higher odds of positive outcomes for youth served by the program.




One year campus-only embargo expired January 2010