The effectiveness and safety of a mobile application-based self-regulation intervention to support weight loss among adults living with obesity: a large-scale pragmatic randomised controlled trial

Background: Obesity is a leading risk factor for avoidable ill health. Effective interventions that can be delivered at scale are needed to support people living with excess weight to lose weight. Methods: We conducted a two-arm randomised controlled trial among adults aged ≥ 18 years living with obesity in the United Kingdom. Participants were recruited via online social media advertisements and allocated 1:1 to receive six months of unlimited access to a purpose-built mobile application based on self-regulation theory (i.e. daily self-weighing, action planning, and weekly reflections), or advice to lose weight. Co-primary outcomes were change in weight and the proportion of participants achieving ≥ 5% weight loss at six months. We also assessed whether the intervention had adverse effects on symptoms of disordered eating. All participants were included in the intention-to-treat analyses. Per-protocol analyses included participants that successfully completed a minimum of one weigh-in and action on at least four separate weeks and had at least one action in their toolbox. Data were analysed using linear mixed effects and analogous logistic models over the six-month period. The trial was registered on ClinicalTrials.gov: NCT05787652. Results: From 13 April to 15 May 2023, 1607 participants were randomly assigned to control (n = 806), or intervention (n = 801). Weight was reported by 632 (39.3%) participants at six months. Mean difference in weight change between groups at six months in intention-to-treat analyses was - 1.85 kg (95% CI: - 2.53, - 1.17, p < 0.001), and the odds of losing ≥ 5% were 2.11 (95% CI: 1.48, 3.03, p < 0.001). Per-protocol analyses showed participants using the app lost an additional - 2.18 kg compared to control, and the odds of losing ≥ 5% were 2.44 (95% CI: 1.67, 3.59, p < 0.001). The proportion of participants scoring above threshold for symptoms of disordered eating declined in intervention relative to control at six months (adjusted OR: 0.51, 95% CI: 0.29, 0.91, p = 0.024). Conclusions: An app with no human contact, designed to foster self-regulatory behaviours, increased weight loss and reduced symptoms of disordered eating in people living with obesity and could be safely deployed at the population level to support effective weight management.