The aim of this study was to determine predictors of energy intake (EI) misreporting in a sample of 35 low-income, weight stable U.S. women (mean age 30, range 19-46). Total energy expenditure (TEE) was measured by the doubly labeled water method (DLW) over 14 days; EI by four, multiple-pass 24-h diet recalls (2 in-person, 2 by telephone) using 2-dimensional food models to estimate portion sizes and analyzed by the Food Intake Analysis System (FIAS) of the Univ of Texas; literacy by the Wide Range Achievement Test (WRAT); and body composition by dual energy X-ray absorptiometry (DXA). Mean EI was significantly lower than mean TEE (2197 + 607 vs. 2644 + 503 kcal, p=.001). Misreporting was defined as EI-TEE (mean -447 + 697 kcal) and was associated with WRAT scores (r=+0.46**), percent body fat (r=-0.45**), and the number of household members (r=-0.35*). WRAT scores and percent body fat were negatively associated (r=-0.55**). Using multiple regression analysis, WRAT scores explained a significant amount of the variation in misreporting over and above the effect of percent body fat (R=0.52**). Under (n=12 ), valid (n=22) and overreporters (n=1) of EI were defined as EI:TEE <0.71, 0.71-1.29, >1.29 respectively. The underreporters had significantly lower WRAT scores in comparison with the valid reporters of EI (1027 + 35 vs. 1045 + 21, p=.05). In this sample of U.S. low-income women, the ability to read and spell as measured by the WRAT, improved the accuracy of their recall of their food intake. Future plans include surveying the sample by questionnaire in an attempt to explain the psychological and behavioral determinants of energy intake underreporting.
Showing the most recent 10 out of 94 publications
