Evaluation of a food frequency questionnaire used to assess nutrient intakes in pregnant women

Objective: To compare nutrient intakes assessed by food frequency questionnaire (FFQ) with those determined from food diaries. Design: A 100-item FFQ was administered to women at 15 weeks of pregnancy. Food diaries were kept for a 4-day period at 16 weeks of pregnancy. Setting: Community-based study of a general population sample of pregnant women booked for delivery at the Princess Anne Maternity Hospital, Southampton, UK. Subjects: 603 women were recruited. Complete dietary data were provided by 569 women. Results: Nutrient intakes determined by FFQ were greater than those from food diaries. Spearman rank correlation coefficients for macronutrients ranged from 0.27 (protein and starch) to 0.37 (fat). Stronger correlations for energy, fat and carbohydrate were seen in women who did not experience nausea, suggesting that the level of agreement observed between the FFQ and food diary in the whole group may be an underestimate of the true agreement. The percentage of individuals classified to the same quarter of the distribution of nutrient intake by the FFQ and diaries ranged from 30% (starch) to 41% (calcium), with between 4% (riboflavin) and 8% (energy, protein and vitamin E) classified to the opposite quarters. Using serum vitamin C as an independent biomarker of intake, the percentage of individuals classified to the correct quarter of intake was similar for the FFQ and diary (34% and 37%), with 8% (FFQ) and 6% (diary) misclassified to the opposite quarter. Conclusion: The FFQ appears to give meaningful estimates of nutrient intake in early pregnancy which can be used to rank individuals within the distribution.