Effect of adherence to a web-based behavioural modification intervention on health behaviours and reduction in overweight/obesity among adolescents

PURPOSE: The purpose of this study was to: 1) investigate whether adherence to components of a web-based weight management intervention have an effect on health behaviours (moderate-to-vigorous physical activity (MVPA), sedentary behaviour, and dietary quality) and change in body mass index (BMI) z-score of adolescents; and 2) to examine how health behaviours mediate the relationship between adherence to components of the intervention and change in BMI-z-score. METHOD: A total of 158 overweight/obese adolescents and their parents participated in an 8-month web-based weight management intervention. Path analysis was used to examine the effect of adherence to intervention components on health behaviours associated with obesity (MVPA, sedentary behaviour, and dietary quality) and change in BMI z-score at 4 and 8 months and to test for mediation effects. Adherence assessed the percentage of content viewed and number of weeks of self-monitoring of physical activity (steps), sedentary time, and dietary intake. MVPA and sedentary behaviour were assessed with accelerometers and self-reported questionnaires. Three self-administered 24-hour dietary recalls were used to compute a dietary quality index. RESULTS: Adolescents viewed 49% and 39% of the web content at 4 and 8 months, respectively. They self-monitored their steps, sedentary time, and dietary intake for 7.5, 2.0, and 3.9 weeks during the first four months and 10.9, 2.7, and 5.6 weeks for the duration of the intervention period (8 months), respectively. The amount of content adolescents’ viewed had a significant direct effect on their dietary quality at both 4 and 8 months (Standardized Coefficient (SC)=0.19, p=0.09 and SC=0.24, p=0.01, respectively) and a significant direct effect on change in BMI z-score at 8 months (SC=0.26, p=0.01). None of the health behaviours mediated the effect of adherence to intervention on change in BMI z-score. CONCLUSION: The study highlights that viewing more content was associated with improved dietary quality and greater reduction in BMI z-score but use of the self-monitoring tools did not explain these changes. Health behaviours could not explain the underlying process of the intervention. Finding ways to maintain adolescents engaged with the intervention is necessary given its effects on health outcomes.Medicine, Faculty ofPopulation and Public Health (SPPH), School ofGraduat