A comparison of the validity of three scales for the assessment of dietary restraint.

The construct validity of Herman and Polivy's Restraint Scale (RS), the restraint factor of the Three Factor Eating Questionnaire (TFEQ-R), and the restraint scale of the Dutch Eating Behavior Questionnaire (DEBQ-R) were investigated by relating these scales to self-reported mean caloric intake per day and to other measures associated with disordered eating and figure consciousness. A factor analysis showed that the three restraint scales measure different components of the restraint construct. A high score on the RS was closely related to consequences of mostly unsuccessful dieting, such as disinhibited eating and weight fluctuations, but not to successful overall caloric restriction in everyday life. High scores on the TFEQ-R and the DEBQ-R represented the more successful dieting behavior component of restraint. The three scales have in common a motivational component of restrained eating, including concerns about shape and weight, and desire for thinness.

Behavioral and biological correlates of dietary restraint in normal life.

Based on laboratory results, restrained eating has been linked to the development of binge eating and eating disorder syndromes such as bulimia nervosa. This study was designed to extend the scope of investigation of the concept of restrained eating beyond the laboratory. Eating behavior and biochemical indices of nutritional state were investigated in 60 young women, who were divided in restrained and unrestrained eaters by questionnaire. Seven-day records of food intake showed that the high-restraint group ate around 400 kcal a day fewer than the low-restraint group. Group differences in actual macronutrient intake and long-term food preferences pointed to a qualitatively altered eating pattern in restrained eaters. Actual protein portion was higher in restrained eaters. They tried to avoid calorie dense food items of high carbohydrate and fat content. Instead, they preferred food regarded as low-caloric and healthy. Plasma levels of triiodothyronine and glucose, which could be taken as indices of long-term adaptation to starvation, were not decreased in the high-restraint group. However, significantly higher levels of triglycerides in restrained eaters may reflect a biological state due to short-term starvation. The results indicate that the concept of dietary restraint predicts eating behavior not only under experimental conditions, but also in normal life. As a consequence of altered eating patterns, psychological and physiological deprivation can be hypothesized in restrained eaters, making them prone to the occurrence of overeating.