LITERACY AND BODY FATNESS ARE ASSOCIATED WITH UNDERREPORTING OF ENERGY-INTAKE IN US LOW-INCOME WOMEN USING THE MULTIPLE-PASS 24-HOUR RECALL- A DOUBLY LABELED WATER STUDY

Objective The accuracy of the multiple-pass 24-hour recall method for estimating energy intake in low-income women in the United States was ascertained by comparing the method with measurements of total energy expenditure. The multiple-pass 24-hour recall is designed to provide r espondents with multiple cues and opportunities to report their food i ntake. It consists of 3 distinct passes: the quick list, detailed desc ription, and review. Predictors of energy intake misreporting (energy intake - total energy expenditure) in the sample were determined. Desi gn Four multiple-pass 24-hour recalls (2 in person, 2 by telephone) we re obtained over a 14-day period to estimate energy intake. Total ener gy expenditure was measured over the same 14-day period using the doub ly labeled water method. Body composition was measured using dual ener gy x-rag absorptiometry: and Literacy was measured by the Wide Range A chievement Test (WRAT) for reading and spelling. Subjects/settings Thi rty-five low-income women between the ages of 19 and 46 years mere tes ted at the General Clinical Research Center at the University of Vermo nt, Burlington. Low income was defined as a household income at or bel ow 130% of the federal poverty level. Statistical analysis Pearson pro duct moment correlation coefficients, t tests, paired t tests, and ste pwise multiple regression analysis were used to test the relationships among study variables. Results Mean energy intake was significantly l ower than mean total energy expenditure (2,197+/-607 vs 2,644+/-503 kc al, P=.001) and the correlation between the 2 measures was poor (r=.22 , P=.20). Percentage body fat and the combined age-adjusted reading an d spelling WRAT scores were the best predictors of misreporting of ene rgy intake (R=.52, P=.006). Conclusions The multiple-pass 24-hour reca ll did not generate a group measure of energy intake that was accurate or unbiased in this sample. Underreporting was strongly associated wi th increased body fatness. The ai;ability to read and spell as measure d by the WRAT improved the accuracy of the women's recall of their foo d intake. Applications Dietetics professionals should take into consid eration the problem of underreporting whenever conclusions are made ab out associations between diet and health and/or when evaluating the im pact of food assistance programs on dietary intake.