Bugental, D. & Schwartz, A. (2009). A Cognitive Approach to Child Mistreatment Prevention Among Medically At-Risk Infants. Developmental Psychology, 45(1), 284-88.
Interventions that use cognitive reframing have been found to be effective at reducing child maltreatment. This study compared the outcomes of mothers enrolled in a standard home visitation program (HV) with mothers enrolled in a program that combined the standard home visitation program with an added cognitive reframing intervention (HV+). The main difference between the HV and HV+ interventions were that the HV+ intervention was focused on the facilitation of the mother’s own problem solving, whereas the HV group provided ideas on how to solve problems. A randomized clinical trial design was used. All measures of abuse and neglect were taken at a home visit one year after the intake visit and all questions were asked retrospectively about events that happened anytime during the study. Families were referred to the program by social workers, physicians and public health nurses due to the presence of a medical risk factor (preterm status, respiratory or cardiac problems, etc); 45 and 57 families completed the HV+ and HV programs, respectively. Home visitors met with families 17 times before the child’s first year of life.
Physical abuse was rare in both groups (4 and 5%, respectively, in the HV+ and HV groups). The prevalence of corporal punishment was significantly reduced in the HV+ group versus the HV group (21% vs. 35%); the intervention had a small effect since it accounted for 5% of the variance in corporal punishment. The intervention also had a small, but significant effect reducing self-reported neglect. The home visitation plus cognitive component may be promising for families referred to the program for a wide range of reasons, and may also be modestly effective at reducing corporal punishment and giving mothers alternative problem-solving and discipline techniques.
Although the study provides promising evidence for cognitively based interventions in reducing child abuse and neglect, the study had several limitations worth mentioning. The sample was deemed a low risk for maltreatment; therefore, it is unclear what effect, if any, the added cognitive component would have with a sample at higher risk of committing child maltreatment. All home visitors implemented both the HV+ and HV programs. Although program fidelity was sought through supervision and clinical notes, the possibility of “leakage” is present (accidental sharing of experimental program interventions with the control group). All measures were based on self report by the mothers, which may reflect how the mother’s perceive themselves, rather than presenting a true measure of their actions. In total, 147 families were referred to the program; 69 were assigned to the experimental condition (HV +) and 78 to the control condition (HV). Failure to complete the program was due to reaction to lengthy questionnaires or a move out of the area. The authors state that they could not conduct “intent-to-treat” analysis, which is usually required in RCTs with attrition (drop-out). Therefore the results should be treated with caution prior to further replication. The ethno-cultural backgrounds of the home visitors were matched to the families. The sample was ethnically biased because Latino families were over-represented (50% of births in the region are Latino but 87% of the sample was Latino). The study should therefore be replicated with samples of more diverse mothers. The authors conclude that further research is needed to measure the long term benefits of these interventions.