Mechanisms of Action in a Behavioral Weight-Management Program: Latent Growth Curve Analysis

Abstract

Background: A greater understanding of the mechanisms of action of weight-management interventions is needed to inform the design of effective interventions. Purpose: To investigate whether dietary restraint, habit strength, or diet self-regulation mediated the impact of a behavioral weight-management intervention on weight loss and weight loss maintenance. Methods: Latent growth curve analysis (LGCA) was conducted on trial data in which adults (N = 1,267) with a body mass index (BMI) ≥28 kg/m2 were randomized to either a brief intervention (booklet on losing weight), a 12 week weight-management program or the same program for 52 weeks. LGCA estimated the trajectory of the variables over four time points (baseline and 3, 12 and 24 months) to assess whether potential mechanisms of action mediated the impact of the weight-management program on BMI. Results: Participants randomized to the 12 and 52 week programs had a significantly greater decrease in BMI than the brief intervention. This direct effect became nonsignificant when dietary restraint, habit strength, and autonomous diet self-regulation were controlled for. The total indirect effect was significant for both the 12 (estimate =-1.33, standard error [SE] = 0.41, p =. 001) and 52 week (estimate =-2.13, SE = 0.52, p