ARS | Publication request: COMPARING NUTRIENT INTAKES BETWEEN DIETARY SURVEYS

Submitted to: Federation of American Societies for Experimental Biology Conference
Publication Type: Abstract Only
Publication Acceptance Date: December 1, 2004
Publication Date: April 2, 2005
Citation: Ahuja, J., Goldman, J., Perloff, B. 2005. Comparing nutrient intakes between dietary surveys [abstract]. The Federation of American Societies for Experimental Biology Journal. 19(5):A1342.

Technical Abstract: Nutrient database values are updated frequently to account for improvements in analytical and sampling methods and to incorporate data for additional samples. Recently improved analytical data for several key foods such as milk, coffee, and bread were incorporated into the U.S. Department of Agriculture's multi-year Food and Nutrient Database for Dietary Studies. To illustrate the impact of database improvements, two comparisons were made between the intakes from the National Health and Nutrition Examination Survey, 2001-02 and the Continuing Survey of Food Intakes by Individuals (CSFII) 1994-96, 1998. In one comparison, CSFII intakes were coded with the original nutrient database; for the second comparison, CSFII intakes were coded with a database to which data improvements since 1998 had been applied. Differences between the two comparisons in mean intakes and the contribution of selected food groups were estimated. Minor but statistically significant (p=0.001) differences in the comparisons were observed for most nutrients. Greater differences between the surveys were observed for some nutrients such as magnesium, whereas a change in direction of the differences was observed for some nutrients/food components such as vitamin C and caffeine. For example, caffeine intakes appeared to decrease by about 20% between surveys in the first comparison, but were shown to increase slightly when database improvements were applied for the second comparison. Hence, to ensure that comparisons of nutrient intakes between surveys are meaningful, improved nutrient values and food portion weights values may need to be applied to dietary intakes from earlier surveys, especially when there are changes involving key foods.