Validity of a food frequency questionnaire varied by age and body mass index

BACKGROUND AND OBJECTIVE: The validity of food frequency questionnaires (FFQs) in measuring food consumption and nutrient intake has to be assessed. The objective of this study was to assess the validity of a 128-item FFQ in specific subgroups of Finnish adults. METHODS: The study included 294 subjects (137 men and 157 women). A 3-day food record was used as the reference method. RESULTS: The mean intake of all nutrients except alcohol was higher measured with the FFQ than with the food records. In general, the Pearson correlations for energy adjusted nutrients between the FFQ and the food records were higher in women than in men. The correlations ranged from 0.14 (retinol) to 0.66 (fiber and alcohol) in men, and from 0.20 (long-chain n-3 fatty acids) to 0.70 (alcohol) in women. The results in subgroups showed that measuring nutrient intakes is more difficult among younger (30-50 years) women and overweight men and women than among others. CONCLUSIONS: The study showed that the FFQ is a useful tool in epidemiologic studies in measuring the diet of Finnish adults given that the problems among specific subgroups are taken into account in interpretation.