Child Welfare and the Treatment of Substance Use Disorders: A Preliminary Analysis of Predictors of Treatment Completion and Family Reunification

Background and Purpose. The intersection of child maltreatment and substance abuse is ubiquitous. Alcohol and drug abuse in families fundamentally impacts the well-being of children. For families involved in child welfare and substance abuse treatment systems, risk is compounded. The practice community is becoming better equipped to meet the needs of families involved in multiple systems; however, limited research has examined the impact of substance abuse treatment for child welfare involved families with respect to family reunification. The purpose of this study is to identify client-based factors contributing to substance abuse treatment completion and family reunification.

 Methods. Data for these analyses are derived from a statewide substance abuse treatment program provided to parents accused of child maltreatment, for whom substance abuse is considered by child welfare personnel to be a contributing factor. Data included treatment-provider clinical administrative information, child welfare agency maltreatment investigation and child removal information, behavioral health claims and encounter data. Data from 4,020 adults who were referred to the program between July 1, 2011 through June 29, 2012 were analyzed.  Women comprised 64% of the sample. Fifty percent of the individuals were white, and 8% were American Indian, African American, or other race. Alcohol (15%), Marijuana (10%), and Methamphetamine (10%) represented the more frequently self-reported substances of abuse.   

 Results. Thirty percent of the sample completed treatment and 20% of the sample achieved family reunification. Two logistic regression models were run. The first model (Chi2(4) = 28.67, p<.001 ) demonstrated that alcohol use (OR = 1.37; p<.01) and marijuana use (OR = 1.65; p<.001), compared to all other drug use, were associated with treatment completion. The second logistic regression model (Chi2(5) = 71.62, p<.001) demonstrated that being female (OR = 1.23; p<.05) and alcohol abuse at assessment (OR = 1.30; p<.02) were associated with greater likelihood of family reunification. In the second model, treatment completion was significantly associated with family reunification (OR = 1.90; p<.01) controlling for gender and substance abuse.

Implication. Preliminary results suggest a need for further inquiry. Future efforts will examine multiple potential confounders through linked “big data” collected by Arizona child welfare administrators. Early findings suggest that treatment providers are especially effective at treating problematic alcohol use and particularly responsive to the needs of women. Practice implications point to a need for gender responsiveness for men and women alike as well as individuals who abuse methamphetamine, opioids, and prescription medications.