Longitudinal accuracy of web-based self-reported weights: Results from the hopkins power trial

Websites and phone apps are increasingly used to track weights during weight loss interventions,yet the longitudinal accuracy of these self-reported weights is uncertain. objective: our goal was to compare the longitudinal accuracy of self-reported weights entered online during the course of a randomized weight loss trial to measurements taken in the clinic. we aimed to determine if accuracy of self-reported weight is associated with weight loss and to determine the extent of misclassification in achieving 5% weight loss when using self-reported compared to clinic weights. methods: this study examined the accuracy of self-reported weights recorded online among intervention participants in the hopkins practice-based opportunities for weight reduction (power) trial,a randomized trial examining the effectiveness of two lifestyle-based weight loss interventions compared to a control group among obese adult patients with at least one cardiovascular risk factor. one treatment group was offered telephonic coaching and the other group was offered in-person individual coaching and group sessions. all intervention participants (n=277) received a digital scale and were asked to track their weight weekly on a study website. research staff used a standard protocol to measure weight in the clinic. differences (self-reported weight- clinic weight) indicate if self-report under (-) or over (+) estimated clinic weight using the self-reported weight that was closest in time to the clinic weight and was within a window ranging from the day of the clinic visit to 7 days before the 6-month (n=225) and 24-month (n=191) clinic visits. the absolute value of the differences (absolute difference) describes the overall accuracy. results: underestimation of self-reported weights increased significantly from 6 months (mean-0.5kg,sd 1.0kg) to 24 months (mean-1.1kg,sd 2.0kg; p=.002). the average absolute difference also increased from 6 months (mean 0.7kg,sd 0.8kg) to 24 months (mean 1.3,sd 1.8kg; p<.001). participants who achieved the study weight loss goal at 24 months (based on clinic weights) had lower absolute differences (p=.01) compared to those who did not meet this goal. at 24 months,there was 9% misclassification of weight loss goal success when using self-reported weight compared to clinic weight as an outcome. at 24 months,those with self-reported weights (n=191) had three times the weight loss compared to those (n=73) without self-reported weights (p<.001). conclusions: underestimation of weight increased over time and was associated with less weight loss. in addition to intervention adherence,weight loss programs should emphasize accuracy in self-reporting. trial registration: clinicaltrials.gov: nct00783315; http://clinicaltrials.gov/show/nct00783315 (archived by webcite at http://www.webcitation.org/6r4gdak5k).