National Cancer Institute Home at the National Institutes of Health | www.cancer.gov

Publication Abstract

Authors: Dixon LB, Subar AF, Wideroff L, Thompson FE, Kahle LL, Potischman N

Title: Carotenoid and tocopherol estimates from the NCI diet history questionnaire are valid compared with multiple recalls and serum biomarkers.

Journal: J Nutr 136(12):3054-61

Date: 2006 Dec

Abstract: To improve the measurement of usual dietary intake, the National Cancer Institute developed a cognitively based Diet History Questionnaire (DHQ), which has been validated against four 24-h dietary recalls (4 24-HR) for energy, macronutrients, and several vitamins and minerals. This analysis used data from The Eating at America's Table Study (EATS) to determine the validity of estimates for carotenoids and tocopherols from the DHQ. Over the course of a year, 163 participants provided 1 or 2 blood samples and completed the DHQ and 4 24-HR. For both the DHQ and the 4 24-HR, crude correlations between serum and diet were modest to strong for the provitamin A carotenoids (alpha-carotene, beta-carotene, beta-cryptoxanthin), low to modest for lycopene, and very low for lutein. The individual dietary tocopherols were weakly correlated with the serum tocopherols, but vitamin E from food and dietary supplements was strongly and positively correlated with serum alpha-tocopherol and strongly and inversely correlated with serum gamma-tocopherol for both instruments. Adjustment for energy, BMI, smoking status, serum total cholesterol, and serum triacylglycerol did not appreciably change the correlations. Using the method of triads, validity coefficients for the DHQ were comparable to the 4 24-HR and were especially strong for alpha-carotene, beta-cryptoxanthin, lutein + zeaxanthin, and total vitamin E in men and gamma-tocopherol and total vitamin E in women. In this study, there was no advantage of 2 blood samples over 1, suggesting reasonably stable ranking of individuals for these biomarkers, which is important for large epidemiologic studies that typically obtain only 1 blood sample for biomarker status.

Last Modified: 03 Sep 2013