Poster #134 - Change Trajectories for Parent Skill Acquisition during Parent-Child Interaction Therapy for Child Welfare-Involved Families

• In Event: 2-163 - Poster Session 08
  In Poster Session: PS 08 Section - Parenting & Parent-Child Relationships

Integrative Statement

Child Maltreatment (CM) is a prevalent and costly social problem within the U.S, with the lifetime economic burden estimated at $210,012 per nonfatal victim (Fang et al., 2012). The majority of CM incidents are parent-perpetrated (Sedlack et al., 2010) and most child victims of CM remain in the home after a substantiated report (USDHHS, 2015). Parenting has therefore emerged as the most proximal and targetable area for intervention in cases of CM.
The current study seeks to understand the mechanisms of change and skill growth among CM parents in an evidence-based parenting intervention for families involved in child welfare, Parent-Child Interaction Therapy (PCIT; Eyberg & Boggs, 1998). PCIT is identified as an Evidence-Based Practice for CM families (California Evidence-Based Clearing House (http://www.cebc4cw.org), as several RCTs demonstrated significant effects on lowering observed harsh parenting, and reducing child abuse recidivism (Chaffin et al., 2004; Chaffin et al., 2011). Understanding parenting therapeutic change reflects a critical next-step direction in evidence-based research. The aims of this study are to map session-by-session trajectories of parenting ‘do’ skills acquisition and ‘don’t” skills declines through the first phase of PCIT, child directed interaction (CDI).
Participants (Mothers with their children, age 3 to 7; N = 20) were drawn from an ongoing randomized clinical trialof PCIT for child-welfare involved families. Multilevel growth curve modeling was used to analyze trajectories in total positive parenting skills taught, i.e., “Do Skills,” (specifically labelled praise, behavior description, and reflection), and total behaviors to avoid, i.e., “Don’t Skills” via session-by-session (7-14 time points) observational coding of interactions (using the Dyadic Parent-Child Interaction Coding System-IV; Eyberg et al., 2005) to determine the nature of change through the first phase of PCIT. Linear and nonlinear (i.e., quadratic, and cubic models) were explored to determine the most appropriate model of fit for positive parenting skills and for behaviors to avoid skills.
Results indicate positive linear changes (β_1j = 1.32, t(17) = 3.69, p < .01) in “Do Skills,” and then quadratic deceleration ( β_2j = -.28, t(17) = -3.45, p < .01) across sessions, while controlling for SES and parent age. Results also show negative linear change (β_1j = -.59, t(74) = -3.53, p < .01) and quadratic deceleration (β_2j = .21, t(17) = 3.28, p < .01) in “Don’t Skills.” For both models, there were significant random linear and quadratic effects across participants.
Parents demonstrated significant growth in skill acquisition of positive parenting skills, with some slowing of gains across the CDI phase of treatment. Parents also exhibited decreases in negative parenting skills across CDI, again with some slowing of reductions in the Don’t Skills. Significant random effects indicate that these changes varied across participants. Understanding change trajectories in PCIT is an important step in clarifying pathways through which PCIT leads to positive changes in family functioning for CM families. Exploring session-by-session trajectories of change in parenting skills has important implications in refining treatment and in providing guidance for policy makers to implement cost-effective interventions that can reduce and prevent CM.

Authors

• Kathleen Anne Johnson, Univeristy of Oregon
  Presenting Author

• Maria Lynn Schweer-Collins, University of Oregon
  Non-Presenting Author

• Emma Lyons, University of Oregon
  Non-Presenting Author

• Elizabeth A. Skowron, University of Oregon
  Non-Presenting Author