Measuring Child Welfare Service Providers' Use of Trauma-Informed Practices: The Child Welfare Trauma-Informed Individual Assessment Tool

Background and Purpose: Limited research exists on measuring the use or effectiveness of trauma-informed care services, particularly in child welfare settings (Chadwick Trauma-Informed Systems Project, 2012). Despite the prevalence of trauma among children and families referred to the child welfare system, many agencies have only recently implemented training efforts to help administrators and front-line staff identify and respond to the trauma needs of their clients. This article describes the development and initial validation of an instrument designed to measure the extent to which front-line child welfare service providers employ trauma-informed practices with the children and families that they serve.

Methods:Data were collected between September 2012 and May 2013 from state child welfare workers attending university-sponsored child welfare trainings. Participants consisted of 213 child welfare workers, supervisors, and administrators from a state child welfare agency. Participants were largely female (83.1%), white/Caucasian (53.5%), and had earned at least a Bachelor’s level of education (97.2%). The Child Welfare Trauma-Informed Individual Assessment Tool included 17 items (5 point Likert scales) that addressed front-line providers’ beliefs about their work with parents and other caregivers, their overall understanding of trauma, and their knowledge of trauma-informed community resources. Items were developed, piloted and reviewed by child welfare practitioners with expertise in trauma and trauma informed care methods  Because the Child Welfare Trauma-Informed Individual Assessment Tool is early in its development, an exploratory factor analysis with WLSMV extraction and Geomin rotation was conducted using Mplus 7.3 to uncover the underlying factor structure.

Results:  From the initial pool of 17 items, a total of 6 items were removed from the analysis due to high error variance (>.70) and significant cross loading of an item across multiple factors (multiple factor loadings >.350). After removal of the 6 items from consideration in the exploratory process, a four-factor model was the most parsimonious solution to demonstrate model goodness of fit (c²=20.75, df=17, p=.238). Additionally, the four-factor model demonstrated a significant improvement in fit over the three-factor solution (difference c²=74.29, df=8, p<.001). Other indices of model fit supported the four-factor solution with RMSEA = .034 (90% CI =.001, .077), CFI=.999, TFI=.966, SRMR=.016. Preliminary analyses indicate that the overall instrument has good internal reliability (Cronbach’s α=.859).

Conclusions and Implications: In light of the recent research highlighting the damaging impact of trauma across the lifespan and the renewed focus on practice and systemic support of child welfare practice to enhance child, youth, and family well-being, including integration of trauma-informed practice (Samuels, 2012), child welfare organizations must be trauma-informed. The development of the Child Welfare Trauma-Informed Individual Assessment Tool represents an important step in measuring the incorporation of trauma-informed practices into child welfare workers’ daily work with clients. The effectiveness of these efforts partially hinges on their ability to understand and recognize the impact of trauma on a child’s development and behavior. The Child Welfare Trauma-Informed Individual Assessment Tool offers a critical first step in measuring the degree to which workers and staff understand trauma-informed care.