Accuracy of Self-Reported Weight Collected Through a Web-Based Platform in a Weight Loss Trial: Validation Study of the POEmaS Clinical Trial.

Self-reported anthropometric data in web-based weight loss interventions may be inaccurate. We studied the agreement between online self-reported and measured weight in the course of the POEmaS randomized controlled trial. Measured weight was not different from reported one (-0.4 kg; 95%CI -0.93 to 0.12). 95.6% of the cases were within the limits of agreement (Bland-Altman method). Self-reported weight collected online was accurate, which suggests that interventions and outcomes assessment can rely on these data.

Using the Behaviour Change Wheel for Designing an Online Platform for Healthy Weight Loss - "POEmaS".

INTRODUCTION: Behaviour change is a key point in weight management. Digital health interventions are attractive tools to deliver behaviour interventions for weight loss, due to the potential to reach a large number of people. We aimed to report how the Behaviour Change Wheel (BCW) was used to develop and implement a web platform to promote weight loss in Brazilian adults with overweight and obesity. Moreover, we aimed to describe the first 12 weeks of usage of the platform in a randomized controlled trial. METHODS: The BCW framework was used to define intake of fruit/vegetables, ultra-processed products and sweetened beverages, leisure physical activity and sitting time as target behaviours. The BCW components of behaviour-capability, opportunity and motivation were used to make a behaviour diagnosis of the population and BCW second layer oriented the selection of information, goal setting, self-monitoring, feedback, social support and incentives as behaviour techniques. Using these behaviour techniques, a 24-week behaviour intervention delivered by seven different platform functionalities was developed. The platform was tested in a three-arm parallel (basic platform versus enhanced platform versus minimal intervention control group) randomized controlled trial from September 2017 to April 2018. In the present analysis, we classified the platform functionalities according to the BCW behaviour component (capability, opportunity and behaviour) and used descriptive statistics and Spearman correlations to report functionalities usage according to the BCW behaviour component over the first 12 weeks of the trial. The study was approved by the Ethics Committee of the Federal University of Minas Gerais, Brazil and was registered under NCT 03435445. RESULTS: Over the first 12 weeks of the RCT, the 809 participants (619, 76.5% women; mean age 33.7 years, SD 10.3; mean BMI 29.9 kg/m2, SD 4.3) were enrolled for use of the basic and enhance versions of the platform. Capability-driven functionalities were accessed by 455 (56.2%) users with median access of 1 (IQR 1-6) times, whereas opportunity-driven platform functionalities were accessed by 592 participants with 8 (IQR 1-27) median access times and motivation-driven functionalities were accessed by 560 (69.2%) participants with 13 (IQR 1-30) median times of access. Spearman correlations between the use of capability and opportunity functionalities, capability and motivation functionalities and opportunity and motivation functionalities were 0.74 (95% CI 0.70-0.77), 0.74 (95% CI 0.70-0.78), 0.89 (95% CI 0.87-0.91), respectively. DISCUSSION: BCW provided a systematic approach to planning, designing and implementing a complex weight loss intervention based on behaviour change. Moreover, it promoted a clear understanding of the relation between platform functionalities and behaviour determinants. The low use of the capability-driven functionalities might have been related to lack of accuracy in the behaviour diagnosis, as well as to implementation issues. The high correlation between the functionalities use suggests that the BCW approach did not determine the platform usage profile. CONCLUSION: The BCW provided a framework for an evidence-based intervention on weight loss delivered by a web platform. Using the framework led to a clear understanding of the behaviour determinants and their relation to the platform features.