The Swenson and colleagues (2010) study was a randomized trial of treatments for physically abused and neglected children and their families in South Carolina. The participants were referred by Child Protective Services (CPS) between November 2000 and October 2003 and consisted of 86 youths and the custodial parents implicated in a CPS report of physical abuse. Participants were included if physical abuse was present (as determined by CPS), if the youths were between 10 and 17 years old, if the family resided within Charleston County, and if the family had open cases within the previous 90 days. Participants were excluded if they had previous experience with Multisystemic Therapy (MST), if reunification was inappropriate or unsafe, or when parents with psychosis were present.
Participants were randomized to a Multisystemic Therapy for Child Abuse and Neglect (MST–CAN) condition (n= 44) and an Enhanced Outpatient Treatment (EOT) condition (n= 42). Both treatments were carried out by a local mental health center. The EOT participants were given the standard services for physically abused youths (substance use, parent training, individual therapy, family therapy, parent–child sessions), but also enhanced services encouraging family engagement, plus a parenting program called Systematic Training for Effective Parenting of Teens. MST–CAN participants received on average of 88 hours of treatment over 7.6 months, while EOT participants received 76 hours of treatment over 4.0 months.
The mean age of the youths in the sample was 13.88 years; 55.8 percent were female; 68.6 percent were African American, and 22.1 percent white. The mean age of parents (reported for abuse) was 41.79 years; 65.1 percent were female, and 58.1 percent were single parents. Measurements were taken at baseline and at 2, 4, 10, and 16 months past the baseline. Numerous methods and sources were used to measure family functioning, child and parent functioning, parenting behavior, social support, reabuse, and out-of-home placement. The instruments used were the Child Behavior Checklist, the Trauma Symptom Checklist for Children, the Global Severity Index and the Positive Symptom Total Scale of the Brief Symptom Inventory, the Conflict Tactics Scale, the Interpersonal Support Evaluation List, and data on reabuse and out-of-home placement from CPS.
Data in the study was analyzed using latent growth curve modeling and effect sizes (Cohen’s d) were calculated for the models in an intent-to-treat analysis.