Category: Obesity / Overweight
Introduction: Food cravings have been implicated in eating and weight-related pathologies such as binge eating disorder, bulimia nervosa, and obesity. Acceptance-Based Treatments (ABT) target experiential avoidance and have been shown to be effective in reducing craving frequency, including in the context of weight loss trials. The mechanisms driving the success of ABT in reducing the impact of cravings on consumption remain uncertain. This study sought to examine the role of dietary restraint and thought suppression as mediators in the relationship between experiential avoidance and craving. We hypothesized that acceptance of food-related thoughts is inversely related to dietary restraint and thought suppression, which in turn are positively related to food cravings.
Methods: Participants (n = 298, 51.5% female) completed the Food Acceptance and Awareness Questionnaire (FAAQ), the “restrained eating” subscale of the Dutch Eating Behavior Questionnaire (DEBQ-R), the White Bear Suppression Inventory (WBSI), and the Food Craving Questionnaire- Trait, Reduced (FCQ-T-red). The hypothesized parallel multiple mediation model was tested using Hayes’ (2013) PROCESS macro for assessing indirect pathways.
Results: Scores on the FAAQ were inversely associated with scores on the FCQ-T-red (r = -.36, p < .001), DEBQ-R (r = -.28, p < .001), and WBSI (r = -.23, p < .001). FCQ-T-red scores were positively correlated with scores on the DEBQ-R (r = .38, p < .001) and WBSI (r = .31, p < .001). The regression of FAAQ scores on the outcome variable of interest, ratings on the FCQ-T-red, was significant (β = -.36, SE = .04, p < .001, 95% C.I.: -.45:-.28). Scores on the FAAQ were also a significant predictor of the hypothesized mediators, scores on the DEBQ-R (β = -.27, SE = .05, p < .001, 95% C.I.: -.36:-.18) and the WBSI (β = -.21, SE = .05, p < .001, 95% C.I.: -.31, - .12). Controlling for FAAQ scores, scores on the DEBQ-R (β = .27, SE = .04, p < .001, 95% C.I.: .18:.36) and the WBSI (β = .18, SE = .04, p < .001, 95% C.I.: .09:.27) were both significant predictors of ratings on the FCQ-T-red. The total indirect effect of acceptance on craving through dietary restraint and thought suppression was β = -.11 and significantly different from zero (95% C.I.: -.16:-.07), thereby providing support for our hypothesis. Controlling for the two hypothesized mediators, DEBQ-R and WBSI, FAAQ scores remained a significant predictor of FCQ-T-red scores (β = -.25, SE = .04, p < .001, 95% C.I.: -.34:-.17), suggesting partial parallel mediation.
Discussion: Findings provide initial evidence that dietary restraint and thought suppression partially mediate the relationship between food-specific experiential avoidance and food cravings. Given that the direct effect of FAAQ scores on craving ratings was more than twice the size of the total indirect effect through the two mediator variables, the indirect effects likely play a relatively minor role in explaining the variance in craving scores. Nevertheless, findings add to our understanding of proximal antecedents of food craving episodes and have implications for the development of more effective interventions targeting the adverse impact of craving on caloric intake and body weight.
University at Albany, State University of New York
Albany, New York