Effect of participant baseline characteristics on adherence to self-monitoring of diet, physical activity and weight in a 12-month weight loss mHealth trial

Abstract Background Self-monitoring (SM) of lifestyle behaviors is the centerpiece of standard behavioral treatment (SBT) for weight loss as SM increases one’s awareness of their behaviors. Extensive evidence demonstrates a significant relationship between adherence to SM and successful weight loss outcomes. Purpose Examine baseline characteristics of participants that are associated with higher adherence to SM of diet, physical activity (PA), and weight in a 12-month mHealth weight loss trial. Methods SMARTER was a single-site, 2-arm randomized controlled trial. We randomized 502 adults with a body mass index (BMI) 27-43 kg/m2 to diet, weight, and PA SM plus tailored feedback messages (SM+FB, n=251) or SM alone (n=251). We provided the SM+FB group feedback messages via the SMARTER app on diet daily, PA every other day and weight weekly. The study provided a wearable PA tracker and smart scale for weight and utilized a smartphone app for dietary SM. Adherence of dietary SM was defined as the % of days with ≥50% of calorie goal recorded and PA SM was % of days with ≥500 steps. Weight SM was the % of days with weight based on the scale record. Linear mixed modeling was applied to assess the monthly percent adherence to SM of diet, PA, and weight over a 12-month period. Our analysis incorporated time effects in months and polynomials of time, treatment group assignment (SM+FB vs. SM), baseline variables of interest (e.g., BMI, age, sex, race, education years, income), and the interactions between treatment and time effects with other variables. For parsimony, non-significant effects were removed. Results The sample (N=502) was 82% white (n=414), 79% female (n=399), and had a mean age of 45.0±14.4 years and BMI of 33.7±4.0 kg/m2. Over 12 months, percent adherence to SM of diet, PA, and weight declined non-linearly. Treatment groups did not demonstrate significant differences. The impact of treatment on the monthly decline in adherence to diet SM was curvilinear and varied significantly over 12 months. Older age was associated with greater adherence to the SM of diet (b=0.72, 95% CI: 0.55-0.88, P < .0001), PA (b=0.50, 95% CI: 0.35-0.66, P < .0001), and weight (b=0.71, 95% CI: 0.57-0.85, P < .0001). Conversely, higher baseline BMI was associated with lower adherence to SM of diet (b=-1.14, 95% CI: -1.73 to -0.55, P < .0001) and PA (b=-0.79, 95% CI: -1.35 to -0.22, P= .006). Other baseline variables were not significantly related to adherence. Conclusion The findings suggest that individuals with a higher BMI at baseline may need additional enhancement strategies to improve their engagement with the digital tools for SM lifestyle which could lead to improved weight outcomes.