Cicchetti, D., Rogosch, F. A., & Toth, S. L. (2006). Fostering secure attachment in infants in maltreating families through preventive interventions. Development and Psychopathology, 18, 623-649.
The deflection of maladaptative development due to maltreatment in childhood onto more adaptive pathways can be achieved with the implementation of preventive interventions targeting the mother-infant attachment. Mother-infant dyads with, recruited through child protection services, were randomly assigned to one of three conditions: psycho-educational parenting intervention (PPI; parenting skills teaching), infant-parent psychotherapy (IPP; improving family attachment and interactions), and a community standard (CS; typical child protection social services); their trajectory was compared to a non-maltreated comparison (NC) group. Families were largely minority and characterized by poverty, little education, large family size, and unstable marriages. Participant of both the PPI and the IPP treatments demonstrated remarkable improvements in attachment, with 55% and 57% going from an insecure to secure status respectively, and no secure to insecure transitions. In comparison, there were negligible improvements in the CS group (2%), and an equal percentage of the NC group (18%) improved and deteriorated with respect to attachment. Changes in maternal sensitivity/representations, stress, and social support did not mediate the pre- to post-intervention changes in attachment. Future research is needed to understand why both treatment groups improved equally, and what is mediating the improvement.
The study is a gold standard with respect to intervention assessment. In addition to the randomization and having two comparison groups, they used a high-risk and non-voluntary sample, and different measures of attachment (self-report, observation of mother-infant play, and the Strange Situation test). This allowed them to determine that all groups had lower than average attachment ratings, but the maltreated groups had the lowest scores. They also took efforts to ensure that attrition (of those recruited for the study, 40% to 50% of each group refused to participate and 21% to 42% were lost during the intervention) did not confound the results. However, the final group sizes were low: 28 for the IPP and 22 for the PPI. Incidentally, neglect was the most common form of maltreatment (85%) reported; 9% were physically abused. According to the authors, reasons why their intervention was so successful compared to prior studies include: extensive training and experience of staff, intervention models were manualized, close supervision, and low case loads (i.e., high staff to participant ratio).