ASPO Abstracts
Change in fruit/vegetable intake and physical activity not mediated by patient-reported quality of life among Latina breast cancer survivors in a diet and physical activity intervention
Category: Lifestyles Behavior, Energy Balance & Chemoprevention
Conference Year: 2022
Abstract Body:
Purpose: To evaluate whether patient-reported quality of life (QoL) domains mediated the effect of a diet and physical activity (PA) intervention on fruit/vegetable (F/V) intake, PA, body mass index (BMI), and waist circumference (WC) among Latina breast cancer (BC) survivors.Methods: Mi Vida Saludable was a 2×2 factorial randomized controlled trial testing the effects of an in-person and/or eHealth communication diet and PA intervention among Latina BC survivors (n=167). We previously reported that women receiving in-person sessions improved in F/V intake, minutes/week of moderate PA, BMI, and WC compared to controls, and controls had higher minutes/week of moderate PA minutes than women receiving eHealth. QoL domains were defined using PROMIS-43, which measures the ability to participate in social roles/activities, anxiety, depression, fatigue, pain intensity, pain interference, physical function, and sleep disturbance. Targeted F/V intake, PA, BMI, WC and PROMIS-43 were assessed at baseline and 12 months. Analysis of covariance assessed intervention effects on PROMIS-43 domains and individual items of each domain at 12 months. All domains and items statistically significant at α=0.10 were retained as potential mediators. The bootstrap method was used to assess whether the identified domains and items mediated changes in main outcomes at 12 months.Results: After 12 months, compared to controls, women who attended the in-person sessions had less sleep disturbance and pain intensity (domains), and less pain interference in ability to participate in social activities, overwhelming worries, and feelings of depression (items). Compared to controls, decrease in pain intensity (domain) was identified as a potential mediator for women in the eHealth arm. However, none of these QoL measures mediated the effect of the intervention on the main study outcomes.Conclusions: In this exploratory analysis, changes in QoL did not mediate the effect of in-person sessions or eHealth communications on change in targeted F/V intake, moderate PA, BMI and WC among Latina BC survivors. Mediation analysis in intervention studies can improve our knowledge about mechanisms for behavior change. Future studies to identify other potential mediators are needed.
Keywords: Mediation analysis; Breast cancer survivors; Lifestyle modification