Do Complex Trauma and Gender Moderate the Relation between Treatment Conditions and Outcomes of Youth in Residential Care?

Background:  Youth in residential care report higher incidences of physical and sexual abuse than youth in community-based care (Collin-Vézina, Coleman, Milne, Sell, & Daigneault, 2011; Zelechoski et al., 2013); with higher rates of complex trauma (CT) among youth in residential (92%) than in community-based care (77%; Brigg et al., 2012).  Research shows gender variances in reported abuse (Collin-Vézina et al., 2011) and psychological symptoms (Maschi et al., 2008) among youth in care. Together, these findings support the need for trauma-informed and gender responsive approaches.  Treatment factors, such as crisis response and length of stay in treatment represent longstanding concerns in residential care that contribute to treatment outcomes.  Although crisis response interventions, such as restraint or seclusion, pose a risk to positive treatment outcomes (Sunseri, 2001; 2004) they are used in both traditional and trauma-informed residential programs to manage youth crises (Muskett, 2014).  Findings on the effects of length of stay on treatment outcomes are mixed (Chow, 2014).  Further examination of the effects of client gender and prior trauma experiences on the relation between treatment conditions and outcomes is warranted.

Methods: Data for this study was drawn from a program evaluation of a Midwest youth psychiatric residential facility that transitioned from a traditional care model to a trauma-informed care model.  Measures of client characteristics were gender and CT.  A measure of CT was created as a continuous scale based on the sum of types of abuse a youth had experienced (ICC Cronbach’s α = .72).  Treatment conditions included model of care, crisis response, and length of stay. Change in impairment was measured using the Child and Adolescent Functional Assessment Scale (CAFAS). A three-step hierarchical regression was used to test the main effects of model of care, gender, CT (step 1), length of stay, and crisis response on change in impairment (step 2) and the moderating effects of gender and CT (step 3). 

Results: The sample included 206 youth in either the traditional (n = 104) or trauma-informed (n = 102) model.  Fifty females (58.82%) and 67 males (55.40%) met the qualification for CT.  Forty-one females (48.2%) and 50 males (41.3%) experienced six or more crisis interventions during treatment. The overall model explained 30.2% of the variance in impairment change (R² = 30.2). Trauma-informed treatment and longer lengths of stay predicted greater improvements while experiencing six or more physical restraints was associated with less improvement (R² = 22.4).  The final model with interactions was statistically significant (R²Δ = 7.8). Specifically, the interaction between gender and length of stay in treatment indicated that longer stays in treatment were associated with less change in functional impairment for girls than boys on average.

Implications: The findings support that youth gender is an important factor to consider in determining appropriate lengths of stay in therapeutic residential treatment. Our results also support trauma-informed care models and the need for improved progress monitoring during treatment for optimal outcomes.  Future research using alternative measures of CT may yield additional findings with implications for treatment approaches in therapeutic residential care.