Integrating and Sustaining Evidence-Based Mental Health Services in Child Welfare

Background and Purpose: Children who are abused and neglected are at an elevated risk of mental health disturbances, but they seldom receive evidence-based mental health services after entering the child welfare system. Hence, there is a need to increase routine access to validated mental health interventions in the child welfare system. Parent training models such as parent-child interaction therapy (PCIT) are among the most effective interventions for young children with emotional and behavioral disturbances. This study examined data from a randomized trial of a PCIT model that was adapted for foster parent-child dyads. Variable- and person-centered approaches were used to examine the impact of PCIT on child symptom profiles.

 

Methods: Eligible children in foster care (n=128), ages 3-6, were randomly assigned to a waitlist control group that received usual care or to one of two treatment groups that received different doses of PCIT. One group of foster parent-child dyads was assigned to receive two full-day group trainings coupled with eight weeks of telephone consultation and homework. The second group was assigned to receive three trainings plus 14 weeks of consultation and homework. At baseline and 14 weeks post-baseline, foster parents rated children’s externalizing and internalizing symptoms via the Child Behavior Checklist and emotional lability via the Emotion Regulation Checklist. Independent t-tests were used to assess mean differences in child outcomes between treatment and control groups. A latent profile analysis was performed to expose different classes of child symptom profiles. Finally, a multinomial logistic regression was conducted to test whether treatment effects varied between latent classes. For this analysis, the two PCIT groups were combined into one group because research indicates that the two treatments did not produce differential effects (Author, 2016).

 

Results: Extending prior evidence indicating that the intervention reduced child externalizing and internalizing symptoms, results showed that children who received PCIT were also rated as having significantly lower emotional lability post-baseline (p = .004). A latent profile analysis of externalizing, internalizing, and emotional lability profiles revealed that a three-class model best fit the data according to BIC and entropy values as well as significant Vuong-Lo-Mendell-Rubin and bootstrapped likelihood ratio tests. Based on their aggregate levels of disturbance, children were classified as presenting with mild problems (30.7%), moderate problems (50.0%), or severe problems (19.2%). Multinomial logistic regressions showed that children who received PCIT were significantly less likely than controls to have severe problems and more likely to have mild problems (OR = 3.6, p = .029) and moderate problems (OR = 4.9, p = .009).

 

Conclusions and Implications: The findings suggest that, while PCIT was developed as an outpatient treatment for externalizing problems, it can be implemented using alternative modalities to remediate an array of child symptom profiles. The findings will be discussed considering the importance of community-university partnerships to adapting, implementing and sustaining evidence-based interventions like PCIT in child welfare. The new Family First Prevention Services Act will also be discussed as a mechanism for reallocating child welfare funds to expand community-based mental health services and in-home parent training.