Dietary Restraint

Originally conceptualized by C. Peter Herman and Deborah Mack, dietary restraint is defined as the conscious restriction of food intake to control body weight. Controlling food intake is important for the prevention and treatment of obesity. Several investigators have shown that higher dietary restraint is associated with a healthier body weight and greater success during weight loss. Yet, others show no relationship or negative associations between dietary restraint and body weight or weight-loss success. Recent findings suggest that further refinement of the construct using subscales to define flexible and rigid styles of dietary restraint may reconcile previously conflicting evidence and fuel new advances in the development behavioral approaches for the treatment and prevention of obesity.

Three self-rating questionnaires have been developed to assess dietary restraint. The Three Factor Eating Questionnaire (TFEQ), also known as the Eating Inventory, is the most extensively studied and most commonly used scale.

Showing restraint by controlling food intake is one of the most significant ways to reduce and maintain weight.

It is comprised of three factors—dietary restraint, disinhibition, and hunger—and was developed to improve upon the validity of the original Restraint Scale, introduced by C. Peter Herman and Deborah Mack. Another scale, the Dutch Eating Behavior Questionnaire (DEBQ), assesses dietary restraint as well as emotional and external eating behavior. Findings on the relationship between measures of dietary restraint and obesity are equivocal, with some showing no relationship, while others find significant positive or negative correlations between dietary restraint scores and body weight or measures of adiposity. The later refinement of the restraint construct into two distinct subscales helps to reconcile the seemingly contradictory research findings.

The subscales distinguish between flexible restraint, defined as a graduated, adaptive approach to dieting, [Page 184]and rigid restraint, an all-or-nothing approach that is often maladaptive. Results from a series of studies conducted in Germany showed negative correlations between body mass index and flexible restraint and positive correlations between body mass index and rigid restraint.

Thus, men and women with higher scores on the flexible restraint scale were found to have lower body mass indices, whereas men and women with higher rigid restraint scores had higher indices. Furthermore, increases in flexible restraint during weight loss were associated with greater success whereas increases in rigid restraint were not associated with weight-loss success. Items on the DEBQ reflect a flexible restraint style; thus, it is not unexpected that increases in scores on this scale have been associated with more successful weight loss, whereas in studies where restraint scores from the TFEQ were used, no relationships were found.

These findings illustrate that flexible but not rigid restraint is associated with weight-loss success. Further research is needed to determine if dietary restraint style can be modified. Specifically, the findings suggest that intervention designed to promote the substitution of flexible for rigid restraint behaviors may enhance the efficacy of current behavioral approaches to the treatment and prevention of obesity.