Development, Challenges, and Evolution of the Log2Lose Intervention for Weight Management: Randomized Controlled...

Long-term adherence to weight loss behaviors is challenging, as most individuals who achieve significant weight loss regain 1-2 kg per year. Financial incentives can reinforce weight-loss initiation and maintenance behaviors, but optimal strategies remain unclear.

This paper describes the design, technical architecture, and operational workflow of Log2Lose, a 5-year, multisite randomized controlled trial testing different financial incentive strategies to promote weight loss and maintenance. We detail the platform's integration with external devices, automated data collection, and adaptations to maintain intervention fidelity in the context of evolving technology and regulatory requirements.

The Log2Lose platform collects daily weight and dietary data from cellular scales and fitness tracking applications, calculates weekly incentive eligibility, and sends automated feedback and motivational text messages. We summarize the technical adaptations, message delivery performance, data completeness, and the balance between automation and manual support required to ensure data integrity.

By the end of the study, 706 participants recorded 181,285 weights and 114,144 daily calorie entries. The platform sent 126,283 text messages and calculated 35,187 incentive payments, with 99.4% (34,976/35,187) processed automatically. Adaptations addressed device integration changes, application programming interface discontinuations, and new text messaging regulations. Despite automation, ongoing technical support was essential for resolving delivery errors, device issues, and data anomalies.

Log2Lose demonstrated that large-scale, fully remote weight loss interventions can be implemented using consumer technology paired with a robust, adaptable platform. Success depended on flexible software design, continuous monitoring, and responsive technical support to navigate regulatory and technological changes. Log2Lose offers a practical model for processing remotely collected longitudinal dietary and weight data, providing valuable guidance for researchers, health care providers, and employers developing similar digital health interventions.

ClinicalTrials.gov NCT04770909; https://clinicaltrials.gov/study/NCT04770909.