Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BHNRC) » Beltsville Human Nutrition Research Center » Food Components and Health Laboratory » Research » Publications at this Location » Publication #131465


item Conway, Joan
item Ingwersen, Linda
item Moshfegh, Alanna

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 6/1/2002
Publication Date: 8/5/2002
Citation: Conway, J.M., Ingwersen, L.A., Moshfegh, A.J. 2002. Restrained eaters report food intake more accurately than unrestrained eaters. International Association for the Study of Obesity, Sao Paolo, Brazil.

Interpretive Summary: none

Technical Abstract: To determine the relationships between psychological measures of eating behavior and the accuracy of reporting food intake we studied 49 women between 21-65y who selected all meals and snacks, during one day, from a wide variety of foods. The USDA 5-Step Multiple-Pass Method for 24-hour recall was administered by telephone the next day. The 3-Factor Eating Inventory was used to determine restraint, disinhibition, and tendency towards hunger. Basal Metabolic Rate (BMR) was estimated by equations and factors that influence food selection were measured by questionnaire. The women were divided by under- or overreporting of energy intake (EI) within and greater than 10% (Groups I greater than 10% underreporters, II less than 10% underreporters, III less than 10% overreporters, IV greater than10% overreporters). The most accurate reporters of EI (Group II) had the highest restrained eating score (14.9 +/- 4.3) and under ate (EI/BMR = 1.31 +/- 0.41) on the day of observation. These women had the highest food selection scores for "attempting to lose weight". There was no difference in disinhibition and hunger score among the 4 groups. No pattern of underreporting of CHO or FAT intake was found. CHO intake was misreported to a greater extent than FAT intake in the overreporters (Group IV). These data suggest that the use of the EI/BMR ratio alone may misclassify women as underreporters who in fact under eat during national surveys. Larger validation studies of dietary assessment methods are necessary in populations diverse in sex, ethnicity, and age.