In this third and final column looking at efficacy studies of psychodynamic psychotherapy, I will turn away from the traditional, individually focused psychotherapies to look at psychodynamic psychotherapies focused on a childcaregiver dyad. The work of Toth et al. (2002) at the University of Rochester testing Preschooler-Parent Psychotherapy (PPP) and of Lieberman, Van Horn, and Ippen (2005, 2006) at the University of California-San Francisco evaluating Child-Parent Psychotherapy (CPP) are fine examples of this active area of research. These dyadic therapies bring together the work of Bowlby, Ainsworth, and Main on the quality of infants’ and toddlers’ attachment to their caregivers with the work of Selma Fraiberg on “ghosts in the nursery.” Fraiberg and her colleagues studied the intergenerational transmission of internal representations, aggressive conflicts, and negative selfrepresentations.

Maltreated preschoolers and their mothers are the study population for Toth et al.’s trial of Preschooler-Parent Psychotherapy. The outcome measure of interest to these investigators was the child’s internal representation of self and of mother and of the mother-child relationship as tested on the McArthur Story- Stem Battery (MSSB) and the Attachment Story Completion Task (ASCT). Mothers and their preschoolers receiving PPP were seen for weekly, 60 minute, dyadic sessions with a clinical therapist, usually at the treatment center but with occasional home visits. The therapist was sensitive to distortions enacted within the preschooler-parent interactions and other clues to influences of maternal internal representations on parenting. Therapists strove to provide a corrective emotional experience for the parent by providing empathy, respect, and unfailing positive regard. Within this holding relationship with the therapist, new experiences of self in relationship to others could be internalized by the mother.

Additionally, new representations of the preschooler could be internalized. Positive representations of the therapist, as they evolved in the therapy, were utilized to contrast with the mother’s representations of herself in relation to her own parents and in relation to her child. In this way, PPP addresses the “ghosts in the nursery” so poignantly described by Selma Fraiberg. Within the sessions, the therapist attends to both the interactional and the representational levels manifested in the mother’s and child’s behaviors. By using observations and empathic comments, the therapist assists the mother in recognizing reenactments during her interactions with her preschooler. Although the therapist’s interactions with the child may provide a model of adultchild interaction, no effort is made to be didactic or explicitly instructive in a psychoeducational fashion. The therapist seeks to respond to maternal utterances and interactional patterns, linking current maternal conceptualizations of parentchild relationships to the mothers’ own experience of care in childhood.

Study participants included 87 maltreated preschoolers and their mothers and 35 demographically matched, non-maltreated, normal controls (NC). The 87 dyads in the maltreated group were randomly assigned to Preschooler-Parent Psychotherapy (n=23), Psychoeducational Home Visitation (PHV, n=34) or Community Standard Care (CS, n=30). The PHV treatment emphasized a focus on the present and the provision to mothers of didactic information and parenting skills training as well as cognitive-behavioral techniques designed to change motherchild interactional patterns. Additionally, these children were enrolled in a 10- month, full-day preschool program. The length of treatment for both PPP and PHV was comparable, about 12 months and 32 sessions. Both treatments were manualized. Therapists had comparable levels of professional training in the treatment modality, including mock therapy sessions and review of videotapes.

The results of this study are rich and complex. Change in maladaptive maternal representations over time was dependent on the intervention condition. The Preschooler-Parent Psychotherapy group exhibited a highly significant decrease in maladaptive maternal representations over time, whereas for the children in the psychoeducational home visit condition, only a marginally significant decline occurred. On this measure, the PPP children were significantly distinguished from the CS and NC groups, whereas the PHV children were not. The PPP group was the only group of children to exhibit a significant decrease in negative selfrepresentation. In regard to positive selfrepresentations over time, PPP, CS, and NC child participants exhibited significant increases in positive self-representation over time, whereas for the PHV group, the improvement in positive selfrepresentation was only marginal. Increasingly positive expectations of the mother-child relationship were seen in all four conditions, but were most dramatically positive in the PPP condition. When coupled with the PPP group’s more significant decline in negative maternal representations, this finding indicates that the PPP intervention brought improvement across the board in the child's maternal representations.

By using an outcome measure reflective of internal representations, this study moves in a direction that is meaningful to psychodynamic practitioners. This study begins to develop an evidencebase for the aim of psychodynamic psychotherapy to reach beyond behavioral symptoms, to create psychological health and resilience. This study may also suggest lines for investigation of the sleeper effect hinted at in several outcome studies of psychodynamic psychotherapy: Do positive internal representations continue to yield benefits over time?

Lieberman, Van Horn, and Ippen’s Child-Parent Psychotherapy (CPP) (2005, 2006) also takes the mother-child dyad, rather than an individual child patient, as the treatment unit. As with Preschooler-Parent Psychotherapy, this treatment developed for preschoolers exposed to marital violence draws on attachment theory and the work of Fraiberg. CPP conceptualizes the therapeutic relationship as a key mutative factor and, in this sense, is clearly psychodynamic. CPP consists of weekly parent-child sessions for one year, during which the therapist’s interventions are guided by the unfolding child-mother interactions and by the child’s free play with developmentally appropriate toys selected to encourage social interaction and to elicit trauma play. The therapist seeks to guide the mother and child in creating a joint narrative of the traumatic events they experienced. Additionally, the therapist seeks to change maladaptive behaviors and support developmentally appropriate interactions between mother and child.

Seventy-five preschoolers and their mothers were enrolled in the study and randomly assigned to the CPP treatment group or to a comparison group receiving monthly case management and individual treatment in the community. At baseline, six months, and one year of treatment and then at six months after termination of treatment, the children were assessed with the Child Behavior Checklist and a clinician-administered caregiver interview using a standardized format to systematize the traumatic stress disorder diagnostic criteria of the Diagnostic Classification Manual for Mental Health and Developmental Disorders of Infancy and Early Childhood.

Children assigned to CPP improved significantly more than children receiving the community treatment. Children in the test treatment demonstrated decreased total behavioral problems and decreased traumatic stress disorder. In August 2006, the six month follow-up assessment of the study participants was published in the Journal of the American Academy of Child and Adolescent Psychiatry. For the CPP treatment group there was continued improvement in the children’s behavior problems and in maternal symptoms compared to the control group.

Psychodynamic dyadic psychotherapies have been a mainstay of infant and toddler psychiatry. These studies by Lieberman et al. and Toth et al. suggest that such treatments can be evidence-based in the sense of meeting the test of randomizedcontrol efficacy testing. The psychodynamic psychotherapy community has traditionally been more interested in observational and longitudinal studies than in efficacy and outcome studies. However, in the field of adult psychoanalysis studies looking at outcome goes back at least as far as the Menninger studies begun in the 1950s. These current studies of dyadic treatments of preschoolers and their parents raise the question of whether dyadic psychodynamic treatments later in childhood might be effective and for what conditions. Each of the treatments discussed here uses psychodynamic theories to carefully structure and guide the treatment design and its practice. Such care would certainly be needed if new dyadic psychodynamic psychotherapies were to be developed for school-age children. The treatments reported here were designed to be developmentally appropriate for the children under treatment. Different developmental considerations would need to inform the treatment of older children in dyads with a caregiver. While research clarifies, confirms, or refutes the matter, research also stimulates new questions and expands our horizons.