Reductions in sugar-sweetened beverage intake and weight are maintained following a digital health intervention: findings from a randomized-controlled trial.

Abstract

Background: Reducing sugar-sweetened beverage (SSB) intake is a global public health priority, but little is known about the long-term maintenance of SSB behaviors after an intervention.

Objective: The objective of this study was to examine 18-month post-intervention maintenance effects on SSB and weight outcomes and to explore the mediation of SSB changes on weight changes.

Design: This study used a two-group randomized controlled trial (RCT) [iSIPsmarter versus patient education (PE)] with assessments conducted at 4-time points (baseline, 9-weeks, and 6-months and 18-months post-intervention) during 2021-2024.

Participants/setting: Of 249 enrolled adult participants, 211 (85%) were retained at 18-months post-intervention. Recruited from Appalachia and surrounding counties, participants were 89% White, 83% female, 50% with incomes <$55,000/year, 86% from more rural counties, and 82% with overweight or obesity.

Interventions: Guided by the Theory of Planned Behavior and health literacy concepts, iSIPsmarter is a highly interactive, structured, digital health intervention comprised of six Cores, behavioral tracking, and personalized action planning. The PE condition was a static self-guided website without personalized information.

Main outcome measures: Self-reported SSB intake and objective weight were measured with the validated Beverage Intake Questionnaire and a cellular-enabled scale, respectively.

Statistical analyses performed: Statistical analyses included generalized linear models and mediation models.

Results: At 18-months post-intervention iSIPsmarter participants (n=102) showed significantly greater reductions in SSB [-30.1 ounces/day (95% CI = -35.3, -24.9; p<0.001)] than PE participants (n=109) [-20.9 ounces/day (95% CI= -25.7, -16.1; p<0.001)](effect size=0.34; p=0.011). Similarly, iSIPsmarter participants experienced a significantly greater percentage weight change [-3.0% (95% CI = -1.5, -4.8; p<0.001)] than PE participants [<-0.01% (95% CI = 0.00, 0.00; p=0.570)](effect size=0.58; p<0.001). The total iSIPsmarter treatment effect on 18-month post-intervention percent weight change was -3.02% (p=0.028), comprising a concurrent indirect effect through SSB intake (-0.94%; p=0.050) and a direct effect (-2.08%; p=0.119).

Conclusions: A digital behavioral intervention, iSIPsmarter, supported long-term improvements in SSB intake and modest, meaningful weight loss. Its digital format suggests scalability to other high-risk populations.