Social Support and Perceived Dietary Barriers Mediate Weight Loss in the AMPLIFY Web-Based Diet Intervention.

Abstract

Background: Web-based interventions that promote healthy diet and weight loss among cancer survivors are growing, but theoretical constructs driving their success remain unclear.

Objective: This study explores whether Social Cognitive Theory (SCT) constructs (self-efficacy, social support, perceived barriers) mediate the effects of the AMPLIFY-Diet intervention on energy intake, diet quality, and body weight.

Design: This mediation analysis used data from the AMPLIFY-Diet randomized controlled trial (RCT).

Participants/settings: Data collected from March 2020 to March 2024 at baseline, 3 months, and 6 months were analyzed from 346 overweight-or-obese cancer survivors enrolled nationwide.

Intervention: Cancer survivors received a serialized, theoretically-grounded, 24-week, web-based intervention to improve diet quality and weight loss.

Main outcome measures: Energy intake and diet quality were measured using two non-consecutive 24-hour dietary recalls, and body weight was measured using digital scales at baseline and 6 months. SCT mediators were assessed using self-reported instruments at baseline, 3 months, and 6 months.

Statistical analyses: Mediation analyses were conducted to estimate indirect effects.

Results: At 3-months, the effect of the AMPLIFY-Diet intervention on weight was significantly mediated through improvements in social support from baseline (unstandardized indirect effect estimate=-0.42 (SE=0.21); 95% CI [-0.894, -0.084]); whereas, at 6-months, the effect was significantly mediated through reductions in perceived dietary barriers from baseline (unstandardized indirect effect estimate=-0.45 (SE=0.21); 95% CI [-0.928,-0.103]). Self-efficacy, social support, and perceived dietary barriers did not emerge as significant mediators for energy intake or diet quality models.

Conclusions: Findings suggest that the AMPLIFY-Diet intervention achieved weight loss among cancer survivors through time-dependent mechanisms, improving social support earlier in the intervention and reducing perceived dietary barriers later on. However, these constructs did not explain changes in models for dietary intake. Future web-based dietary interventions should measure and examine additional theoretical constructs to better understand the mechanisms underlying dietary change.