Submitted to: International Conference of Dietary Assessment Methods
Publication Type: Abstract Only
Publication Acceptance Date: March 24, 2006
Publication Date: April 26, 2006
Citation: Steinfeldt, L.C., Anderson, E.T. 2006. Automated food coding in a national dietary intake survey [abstract]. 6th International Conference on Dietary Assessment Methods, April 27-29, 2006, Copenhagen, Denmark. Program & Abstracts. P04-19. Technical Abstract: Improving the quality of dietary intake data in large scale surveys is a continuing goal for the Food Surveys Research Group in the U. S. Department of Agriculture. In support of this goal, software was developed to automatically code foods and amounts of foods eaten. Based on the Automated Multiple Pass Method (AMPM) instrument used to conduct dietary intake interviews, details about each food are matched against predetermined sets of questions and responses that have been linked to food codes. For the amount eaten, different procedures are used based on how the amount was reported, i.e., food models, food specific portion size descriptions such as a 12 fluid ounce can, and generic descriptions such as one medium. The software determines which procedure to follow by matching the quantity responses against a portion size translation table. For both food and amount coding, the software does simple text matching and data table lookups. Previously coded foods and amounts were used to develop the data tables. Automated coding is being implemented in phases in What We Eat in America, the dietary intake component of the National Health and Nutrition Examination Survey (NHANES) as well as other studies.