Abstract
This secondary analysis sought to understand how engagement with individual intervention components relates to outcomes. Young adults were randomly assigned to (1) targeted, (2) tailored, or (3) contact control groups in which content was delivered via Facebook, text messaging, and feedback reports. Participants were given one point for each component with which they engaged per week, for a maximum of 78 possible points over 6 months. Tertiles of engagement were defined as follows: low engagement (0-51% engaged), moderate engagement (52-66% engaged), and high engagement (67-100% engaged). Of the 354 participants with 6-month weight measurement, overall engagement decreased 33.6% from weeks 1 to 26 and was related to weight loss (p < 0.01) with high-engaged participants losing 1.8 kg more than low-engaged participants (p < 0.01). For every 10% increase in month 1 Facebook engagement, participants lost 0.34 kg (p < 0.01), which may have been driven by the targeted group (0.41 kg, p = 0.03). For every 10% engagement increase in the feedback report over 6 months, participants lost 0.27 kg (p = 0.02), which may have been driven by the tailored group (0.45 kg, p < 0.01). Weight loss interventions should consider increasing media use and personalized feedback for greater efficacy. TRIAL REGISTRATION: ClinicalTrial.gov Identifier: NCT02342912.