Comparison of a self-administered quantitative food amount frequency questionnaire with 4-day estimated food records.

Our objective was to assess the relative validity of a self-administered food amount frequency questionnaire (FAQ), using 4-day estimated food records (FRs) as a reference method, for use in a subsequent clinical study of patients undergoing heart surgery. Thirty healthy subjects (19 males, 11 females), aged 45-75 years, were randomly recruited from patient lists generated by two local GP practices in Oxford. Complete data from the FAQ and FR, administered twice 6 months apart, were available for 25 subjects (16 males, 9 females). For absolute nutrient values, intakes of protein, CHO, total fat, PUFA, thiamin, iron, dietary fibre and alcohol were not significantly different between the FAQ and FR, and Pearson's correlation coefficients ranged from 0.28 for protein to 0.88 for total fat. Estimates from the FAQ were within +/- 10% of the estimates produced by the FR for two-thirds of nutrients. When nutrients were expressed as a percentage of total energy intake, no statistically significant differences were observed for any nutrient between the two methods, and correlations ranged from 0.32 for protein to 0.80 for SFA. In conclusion, the broad dietary patterns obtained by the two methods of assessment were comparable. This simple and inexpensive FAQ can be used to usefully estimate group intakes for a variety of nutrients in the study of patients undergoing heart surgery.