Objective: Assess the association between reporting bias of dietary energy intake and the behavioral and psychological profiles in women.

Design: At baseline a series of questionnaires were administered to 37 women,(the Marlowe-Crowne Social Desirability Scale, Weinberger Adjustment Inventory (WAI), the Eating Disorder Inventory (EDI), the Restraint Scale and Sorensen-Stunkard’s silhouettes). Subjects received training on how to record dietary records. Subjects recorded three days of dietary records to measure energy intake (EI) during a study to determine total energy expenditure (TEE) using doubly labeled water. Reporting accuracy (RA= EI/TEE× 100) was determined for each subject. Statistical analysis of the data used a mixed effects model accounting for within subject variability to determine if the psychological scores were associated with reporting accuracy.

Setting and subject: Women were recruited with local advertisements in Tucson, Arizona. The women had a mean (±1 sd) age of 43.6±9.3 yrs, body mass index (BMI) of 28.7±8.5 kg/m 2 and total body fat (% TBF) of 31.9±7.3%.

Results: Age and% TBF were significantly and inversely associated with RA. Furthermore, Social Desirability was negatively associated with RA. Body dissatisfaction and associating a smaller body size than one’s own as being more healthy were also associated with a lower RA.

Conclusions: These results suggest that Social Desirability and self image of body shape are associated with RA. Modifications in subject training may reduce the effect of these factors on RA.