A mapping framework to facilitate the merging of existing food frequency questionnaires (FFQs)

The overall aim of this activity was to generate a visual map of previously used FFQs alongside guidelines of how to map across these FFQs. Information on eight pre-existing FFQs was compiled; namely, EPIC (UK based FFQ), Scottish Collaborative Group, Food4Me, eNutri65+ (based on Food4Me), What I eat 60+, Feel4Diabetes, the National Health and Examination survey FFQ (NHANES -American FFQ), CHANCE and the Japanese FFQ.

Firstly, an overview of each FFQ was compiled containing the list and number of food groups, the list of food items contained within each food group, the timeframe response options for frequency of consumption questions, and, the timeframe each FFQ captured (e.g., previous month, previous 3 months, previous year). A summary map was created in excel to visualise the similarities across each of the individual FFQ food groups. Each FFQ was presented in a single column with food group names displayed in red text and individual food items contained within these groups listed underneath. Common food groups across the FFQs were presented in the same row. Food group names which were similar but not exact matches were placed either a row above or below these. Next, each food item was added underneath their respective food group and the same process repeated to align common foods. Food items which were common within each food group across the FFQs were placed in the same row with blanks left to accommodate food items which may not have been included in all FFQs.

The purpose of the developed mapping system for food frequency questionnaires (FFQs) is to facilitate the re-use of previously collected FFQ data by enabling different datasets to be merged together thus creating larger datasets suitable for answering new research questions.