Responsivity to Problem-Solving Skills Training in Mothers of Children With Cancer
Michael J. Dolgin, Katie A. Devine, Dana Tzur-Bitan, Martha A. Askins, Diane L. Fairclough, Ernest R. Katz, Robert B. Noll, Sean Phipps, Olle Jane Z. Sahler
Abstract
OBJECTIVE: Bright IDEAS (BI) is a problem-solving skills training (PSST) program that has been demonstrated in earlier randomized controlled trials (RCTs) to be an effective and specific intervention for improving problem-solving skills and reducing negative affect in caregivers of children with cancer. The objectives of this study were to (a) offer an approach to defining meaningful treatment response and to determine the rates of responsivity to PSST; and (b) identify characteristics of PSST responders and nonresponders. METHODS: Data from 154 mothers receiving the BI intervention were analyzed. Drawing on the literature on minimal clinically important differences, two criteria for determining responsivity were calculated for the primary outcome of problem-solving skills: (a) The reliable change index (RCI) based on group data, and; (b) The effect size (ES) of each participant's pre/postintervention change score as a function of the group's baseline SD. RESULTS: Thirty-three percent of the sample met both responsivity criteria immediately posttreatment (39% at follow-up) and 38% (39% at follow-up) met neither. An additional 29% demonstrated a small or greater ES (≥ 0.2) but did not meet the RCI criteria, suggesting possible benefit. The single consistent predictor of responsivity was participants' pretreatment problem-solving skills, with lower skills at baseline predicting greater improvement (p < .001). CONCLUSIONS: These findings highlight the need to go beyond group data in interpreting RCTs and to incorporate measures of meaningful treatment response. Our ability to predict and screen for meaningful treatment response is critical to more precise targeting, enhanced outcomes, and better resource allocation.