The Effect of a Statewide, Trauma-Informed Intervention in Child Welfare on Child Well-Being

Background: Childhood trauma, including maltreatment, is a serious public health issue in the U.S. In 2015, approximately four million referrals for abuse and neglect were made to child protective services involving over 7 million children (USDHHS, 2017). Children in child welfare (CW) often experience complex trauma–multiple traumatic events that begin early in life and involve primary caregivers (Kisiel et al., 2009).  Evidence-based trauma treatment can be key to their recovery, but studies rarely compare child outcomes of different treatment models.  This study examines the effectiveness of Attachment, Self-Regulation and Competency (ARC; Kinniburgh et al., 2005) Child-Parent Psychotherapy (CPP; Lieberman & VanHorn, 2004/2016), and Trauma-Focused Cognitive Behavioral Therapy (TF-CBT; Cohen et al., 2006) in an evaluation of a statewide, trauma-informed CW initiative. Specifically, we examined effects on child behavior problems and symptoms of posttraumatic stress.

Method: A total of 842 children in the evaluation received treatment: 44.89% (n=378) ARC, 35.99% (n=303) TF-CBT, and 18.76% (n=158) CPP. Clinicians (n=323) recruited eligible children: birth-18 years; English/Spanish speaking; involved in the CW system within a year of referral. Children averaged 9.14 years at baseline; 4.35% were Hispanic, 70.31% were White, 18.65% were African-American; and 38.12% were in state custody. Children averaged 5.2 types of trauma. Clinicians administered assessments at baseline, 6, 12, and 18 months. Measures included the UCLA PTSD Index-RI (Pynoos et al., 1998) and the Young Child PTSD Checklist (Scheeringa, 2010) to assess posttraumatic stress, and the Child Behavior Checklist (CBCL; Achenbach & Rescorla) to assess behavior problems. We used four-level regression models to account for non-independence of observations (repeated child assessments), and among clinicians and agencies. Covariates included child age and sex, total trauma types, psychotropic medication use, and custody status; we addressed missing data using maximum likelihood estimation.

Results: Youth reported that the severity of their posttraumatic symptoms decreased significantly from baseline to 6 months (B=-6.725, p<.001, ES=0.497) and 12 months (B=-8.294, p<.001, ES=0.614), as did their parents (B =-0.799, p<.05, ES=0.165 and B=-2.188, p<.001, ES=0.452). Arousal, Re-experiencing, and Avoidance/Numbing symptoms also decreased, except for Avoidance/Numbing at 12 months. Young children had reduced symptoms, except at 12 months. We found significant improvements in children’s behavior problems from baseline to 6 months (B=-4.428, p<.001, ES=0.424) and 12 months (B=-5.482, p<.001, ES=0.525). However, outcomes differed by model and child age, with the most positive outcomes for ARC and TF-CBT, and the fewest benefits for younger children and those in CPP.

Conclusion: Positive child outcomes suggest that trauma treatment is an effective means of improving the developmental trajectories of children with complex trauma, but model effectiveness varies. Young children may not have improved as much as their older peers due to limited treatment dosage (e.g., few children completed CPP’s 18-month treatment), due to poor measurement options for young children, and/or limited treatment efficacy. We will engage the audience in a discussion of study challenges and successes.