Are dietary restraint scales valid measures of longer-term dietary restriction? Objective biological and behavioral data suggest not.

Abstract

Observational studies indicate that dieting predicts bulimic symptom onset, but experiments indicate that low-calorie diets reduce bulimic symptoms. One explanation for the inconsistent findings is that dietary restraint scales used in the former studies are not valid measures of dietary restriction. Indeed, we previously found that dietary restraint scales were not correlated with objective measures of acute caloric intake. In this follow-up report, we re-evaluate this finding using two measures of longer-term energy intake: Caloric intake from repeated lunches purchased at worksite cafeterias over a 3-month period and estimates of energy intake derived from two doubly labeled water (DLW) studies. All three studies measured restraint using the Three Factor Eating Questionnaire (TFEQ)—Cognitive Restraint (CR) scale; all three also employed measures of self-reported caloric intake (food diaries and/or 24 h food recalls). CR showed significant inverse correlations with self-reported caloric intake (as it has in past studies), but was not correlated with observationally measured caloric intake in the cafeteria or with DLW-measured caloric intake over 2 weeks in either of the DLW studies. These results are consistent with the hypothesis that restrained eaters may consume less food than they would like to eat but not less than they need to maintain energy balance. The findings could also explain past discrepancies in the literature concerning dieting as a risk factor for bulimic pathology.