Page Banner

United States Department of Agriculture

Agricultural Research Service

Title: Highly Motivated Subjects: Comparative Accuracy of Dietary Intake Methodologies

Authors
item Blanton, Cynthia
item Moshfegh, Alanna
item Baer, David
item Ingwersen, Linda
item Omolewa-Tomobi, Grace
item Staples, B - USDA, ARS, BELTSVILLE, MD
item Kretsch, Mary

Submitted to: American Dietetic Association Annual Meeting
Publication Type: Abstract Only
Publication Acceptance Date: May 3, 2005
Publication Date: October 1, 2005
Citation: Blanton, C.A., Moshfegh, A.J., Baer, D.J., Ingwersen, L.A., Omolewa Tomobi, G., Staples, B.S., Kretsch, M.J. Highly motivated subjects: comparative accuracy of dietary intake methodologies. American Dietetic Association Annual Meeting. 2005. 105(8) Supplement:A-17.

Technical Abstract: The subjective nature of dietary self-report introduces error & bias into nutrition assessment. This study sought to compare dietary intake methodologies in a best-case context using subjects selected for high motivation. Energy intake (EI) assessed by the USDA Automated Multiple-Pass Method (AMPM), a relatively new computer-administered dietary recall, & three widely used dietary assessment methods [food records (FR), the Block food-frequency questionnaire, & NCI’s Diet History Questionnaire (DHQ)] was evaluated against the gold standard for estimating energy expenditure (EE), doubly-labeled water (DLW). Twenty highly motivated, normal-weight women [age 30.0 ± 3.6 y (mean ± SD); BMI 22.1 ± 1.9] participated in a free-living study during which they completed two unannounced AMPM recalls (d 1 & 16), 14 days of food records (d 2-15), & the Block & DHQ (d 16). AMPM & FR were analyzed using USDA’s Food & Nutrient Database for Dietary Studies; the Block & DHQ were analyzed by Block DDS & DietCalc software, respectively. Results: AMPM & FR measures of EI did not differ significantly from DLW EE [AMPM: 2145 ± 627 kcal; FR: 2010 ± 529; DLW: 2127 ± 449; (GLM Mixed Model with multiple comparisons (SAS v.8.2): main effect of method, p<0.0001; AMPM vs. DLW p=0.89; FR vs. DLW p=0.29)]. Conversely, EI was underestimated by ~30% by the questionnaires (Block: 1520 ± 523; DHQ: 1484 ± 472; p<0.0001 vs. DLW). Conclusions: In a best-case scenario, AMPM measured EI as well as the labor-intensive 14-d FR but was significantly more accurate than the Block & DHQ.

Last Modified: 4/17/2014