Automated lifestyle interventions and weight loss: a systematic review of randomized controlled trials

Overweight and obesity drive cardiometabolic disease and high-intensity lifestyle interventions are standard. However, many health-systems cannot offer these interventions and many patients cannot participate even when available. Trials have assessed automated digital lifestyle interventions to improve accessibility. This systematic review identified automated digital lifestyle interventions and assessed their impact on weight loss.

The review (CRD42023435700) identified randomized controlled trials of at least 3-months duration assessing automated digital lifestyle interventions' impact on weight loss. Data were managed through Covidence with double-blinded screening of titles/abstracts and double-blinded full-text review to determine study inclusion. Data extraction was completed by one reviewer (NWC) and verified by a second (MM, JS, TE, CE). The Cochrane Collaboration's tool was used to assess bias risk and study quality was rated as High, Moderate, Low, or Very Low.

The search identified 1817 citations. The full-text of 60 reports were assessed and 17 reports of 16 studies were included. The majority (63%) were of moderate quality. No intervention produced 5% weight loss from baseline. Intervention components with the largest impact were text message encouragement and education with a cognitive behavioral approach. No specific form of self-monitoring was most effective, though some form of self-monitoring was included in most trials.

Some components of the automated digital lifestyle interventions showed promise. Research is needed to optimize these components (e.g., tailored messaging, cognitive-behavioral approaches) balanced with human contact. Considering the magnitude of the problem and its disproportionate impact on low socioeconomic and minority patients, interventions optimized for effect and scalability are needed to address overweight and obesity.