Skip to main content
ARS Home » Research » Publications at this Location » Publication #166236
Title: SELF-REPORTED VARIANCE FROM USUAL AMOUNT OF FOOD IN 24-HOUR RECALL TELEPHONE SURVEY

Author
item McCabe Sellers, Beverly
item SIMPSON, P - DELTA NIRI
item GOSSETT, J - DELTA NIRI
item Bogle, Margaret

Submitted to: Journal of Nutrition Education
Publication Type: Abstract Only
Publication Acceptance Date: 4/1/2004
Publication Date: 7/20/2004
Citation: McCabe-Sellers, B.J., Simpson, P.M., Gossett, J., Bogle, M.L. 2004. Self-reported variance from usual amount of food in 24-hour recall telephone survey [abstract]. Journal of Nutrition Education and Behavior. 36(Supplement 1):S31.

Interpretive Summary:

Technical Abstract: In designing community nutrition intervention research and in assessing cross sectional nutritional surveys, 24-hour recalls are frequently used to assess dietary intake. While 24-hour recalls are considered appropriate for obtaining group data, the question must be considered as to how representative a given day is to usual food intake in assessing dietary adequacy. In a representative sampling of 1732 rural Southerners in the Lower Mississippi Delta region, the question was asked if the reported intake was the same, more than, or less than usual intake. Pearson Chi-square tests were used to determine if the usual amount of food varied by age, race, gender, education, income and if energy intake varied by obesity status, gender, or by day of the week. ANOVA was used to determine if total energy differed by usual intake and if total energy intake differed by day of intake. Significant differences were found in usual intake by age (p< 0.001) and gender (p<0.012) but not by race, education, and income. Usual intake was reported by 62%, less intake by 27.5%, and more intake by 10.5%. Obese persons were less likely to describe intake as usual (p<0.001) but did not report greater total energy intake than nonobese (p<0.17). Subjects who reported more than usual intake had higher total energy intakes. Day of intake did not appear to affect total energy intake. These data can be used to estimate potential variance in usual intake in research design.