Ryan M. Smith is a 2010 graduate of the PsyD Program in Clinical Psychology at Antioch University, Santa Barbara.


early childhood, infant mental health, attachment, Devereux Early Childhood Assessment, Working Alliance Inventory, therapeutic relationship, parenting, home visitation, quantitative, bootstrapping, mediation

Document Type


Publication Date



An infant’s attachment relationship with primary caregivers has been demonstrated to have a long-term relationship to an individual’s social and emotional functioning throughout the lifespan. Recognizing the critical importance of this period, interventions to facilitate secure attachment are now being evaluated for treatment efficacy. Evaluation of these treatments has typically focused on the components of treatment, examining changes in maternal sensitivity, parental attachment representations, and concrete support to address basic needs, housing, or other contextual factors, and evidence has been found to support the inclusion of these factors. However, little is known regarding what elements of treatment impact the effectiveness of dyadic parent-child interventions; the research that has been completed has focused primarily on aspects of the intervention. There continues to be considerable debate in the psychotherapy literature regarding whether the specific components of an intervention, or the common factors present in all interventions, are responsible for therapeutic change. The purpose of the present study was to examine the role of one common factor, therapeutic alliance, in facilitating attachment-based protective factors in the child. It was hypothesized that therapeutic alliance, as rated by the parent, would predict improvements in attachment-related protective factors as rated by the treating interventionist following 6 months of dyadic intervention. The results of the present study found that specific subscales of the therapeutic alliance (the goal, task, and total alliance scales) predicted changes in children’s initiative behavior, but not their attachment-related engagement behavior. Therapeutic alliance also predicted treatment participation, and it was not possible to rule out treatment exposure as a mediating variable between therapeutic alliance and change in initiative behavior. Implications for future research and practice are discussed.