Submitted to: Journal Of The American Dietetic Association
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/19/2008
Publication Date: 7/1/2009
Citation: Carithers, T., Talegawkar, S., Rowser, M., Henry, O., Dubbert, P.M., Bogle, M.L., Taylor, H.A., Tucker, K.L. 2009. Validity and calibration of Food Frequency Questionnaires used with African American adults in the Jackson Heart Study. Journal of the American Dietetic Association. 109(7):1184-1193. Interpretive Summary: Food Frequency Questionnaires (FFQ) are instruments designed to collect food intake information from a large group of people within a relatively short time. The FFQ must be specific to a given population and include all the usual foods that people in that region usually eat. Development of the instrument needs to be followed by a validation study to make certain the foods included are the right ones for the intended group of people. No such instrument has been available for the Lower Mississippi Delta region. This study describes how the Delta FFQ was validated by use in the Jackson Heart Study (JHS) Diet and Physical Activity substudy of 499 men and women in Jackson, Mississippi. The FFQ, both a long and short form, was compared to 24-hour diet recalls obtained from the JHS. Both the long and short forms of the FFQ appear to be a valid way of assessing dietary intake of African Americans in the Delta. The Delta FFQ can be used in future research studies in the Delta.
Technical Abstract: This paper details the calibration process and validity of two Food Frequency Questionnaires (FFQs) developed for use in investigating diet disease relationships existing within the African American population. A calibration design was used which included comparison of mean nutrient intakes from four 24-hour diet recalls with a short FFQ and a long FFQ developed by the Delta Nutrition Intervention Research Initiative (Delta NIRI). The setting was the Jackson Heart Study in Jackson, Mississippi. A subset of participants (n=499) from the Jackson Heart Study (JHS) cohort (n=5302) was selected as representative of participants in the Diet and Physical Activity Substudy (DPASS) in Jackson, Mississippi. For both men and women, both nutrient intake estimates tended to be higher on the long and lower on the short FFQ compared to the recalls. Energy adjusted correlations ranged from 0.15 for sodium to 0.60 for carbohydrate intake for the short FFQ and from 0.19 for poly-unsaturated fat to 0.64 for magnesium intake for the long FFQ. Both the short and long Delta NIRI FFQs appear to be valid for assessment of dietary intake in African Americans. The long FFQ exhibited higher correlations for most nutrients analyzed.