Date Presented 4/8/2016
Food refusal in young children is common. Parent-led intervention is needed to support improvement during daily meals. This pilot study supports the use of parent-led intervention and data collection for family-centered intervention optimization.
Primary Author and Speaker: Angela R Caldwell
Contributing Authors: Lauren Terhorst, Ketki Raina, Joan Rogers
RESEARCH QUESTION: Is it feasible to measure the effects of a parent-led, multicomponent feeding intervention in the home using parent-initiated data collection?
RATIONALE: Food refusal is common among young children. When mealtimes become a daily struggle for the family, picky eaters are referred for occupational therapy services. Therapist-led interventions improve food acceptance in the clinic, but there is limited evidence to support generalizability of skills to the home. Parent-led intervention and data collection in the home has the potential to facilitate intervention optimization from a family-centered perspective.
DESIGN: Single-case experimental design with three phases (ABA), each phase consisting of 10 video-recorded meals. No intervention was provided during the A phase. The parent implemented a multicomponent (sensorimotor routine and high-preference food sequence) feeding intervention during the B phase.
PARTICIPANT: We recruited one young child (convenience method).
MEASURES: Ten meals in each phase were video recorded and observed by the researcher at a later time. For each meal, the frequency of accepted bites of preferred and nonpreferred food was documented. Intervention adherence was measured by the parent using a self-administered intervention protocol checklist. The parent also took daily field notes describing experiences and unusual events.
ANALYSIS: Raw feasibility data were converted to means and percentages and compared with benchmarks set before study initiation. Acceptance of nonpreferred food was analyzed using visual analysis, celeration line method, and the c-statistic to determine if there were significant changes between phases in this study. The standard mean difference was calculated to determine the magnitude of the effect size.
RESULTS: All three feasibility benchmarks were met. Data were collected ≤10 days/phase (on average, 9.67 days), video quality was adequate to determine number of bites of nonpreferred food ≥ 8 meals per phase (on average, 9.67 meals per phase), and parental adherence to protocol was ≥90% (100% adherence achieved). The infant demonstrated a significant increase in acceptance of nonpreferred food bites between the baseline (M = 18.70) and withdrawal (M = 25.60) phases, indicating improved acceptance over time. This change was significant (c = 0.78, p < .05), and a moderate effect of time was observed (d = 0.68). During the intervention phase, when preferred food was offered in addition to nonpreferred food, overall acceptance improved (M = 25.4), but intake of nonpreferred foods actually decreased (M = 12.20, c = .64, p < .05).
DISCUSSION: Although causal inferences cannot be made on the basis of this study, it provides preliminary evidence to support the feasibility of implementing a parent-led feeding intervention with data collection in the home. Findings also show that a parent-led intervention with repeated exposure to nonpreferred foods improved acceptance of nonpreferred food over time in a 7-mo-old child.
IMPACT STATEMENT: This proposal supports the use of video recording, self-administered checklists, and field notes to collect data on adherence and the effects of parent-led, home-based feeding interventions. Parent-initiated data collection may support evaluation of intervention effectiveness in the home and the development of family-centered modification strategies.
