Building Protection In Interventions For High Risk Adolescents With Serious Emotional Disturbances And Aggression

Natasha K. Bowen
School of Social Work, CB 3550
University of North Carolina
Chapel Hill NC 27705
919 403 2976
FAX: 919 962 7557
nbowen@email.unc.edu
Purpose: Adolescents with serious emotional disturbances (SED) and violent behaviors are costly and difficult to serve in public mental health settings. Many also become involved with the legal system, increasing the costs of serving them and decreasing their chances of successful adult functioning. Using data collected by practitioners in an intensive intervention program in North Carolina, this study sought to determine if (a) building protection was associated with decreases in violent behavior, and (b) program effects differed for adolescents with different levels of initial protection.
 
Methods: Annual assessment/treatment-planning data collected by case managers in the program include over 60 established risk and protective factors, and indicators of violent behavior.  Changes over time in mean behavior ratings, and mean levels of protection (total and 2 subcategories) were examined for 388 adolescents (ages 13-17) with three waves of data.  The relationship between changes in protection and behavior was then examined using linear regression and HLM. Analyses were conducted for the full sample and three subgroups with different entry levels of protection.
 
Results: Improvements in behavior and three protection composites were observed for the full sample. Subgroup experiences differed, however, with the high protection group often showing no improvement. After controlling for demographics and initial risk levels, increased protection was predictive of improved behavior. HLM analyses revealed homogeneous rates of change within subgroups, higher rates of change for the low protection group, and the importance of home social skills to behavioral improvement.
 
Implications for Practice: Results validated the intervention goal of building protection in high risk adolescents with SED and aggression; suggested that home social skills are a key protective factor to target in efforts to reduce violent behavior; and indicated that adolescents with high, medium and low levels of initial protection respond differently to interventions and therefore require different intervention strategies.