Objective: To determine the relative validity of the pre-coded food diary applied in the Danish National Survey of Dietary Habits and Physical Activity. Design: A cross-over study among seventy-two adults (aged 20 to 69 years) recording diet by means of a pre-coded food diary over 4 d and a 4 d weighed food record. Intakes of foods and drinks were estimated, and nutrient intakes were calculated. Means and medians of intake were compared, and crossclassification of individuals according to intake was performed. To assess agreement between the two methods, Pearson and Spearman's correlation coefficients and weighted kappa coefficients were calculated. Setting: Validation study of the pre-coded food diary against a 4 d weighed food record. Subjects: Seventy-two volunteer, healthy free-living adults (thirty-five males, thirty-seven females). Results: Intakes of cereals and vegetables were higher, and intakes of fruit, coffee and tea were lower, in the weighed food record compared with the food diary. Intakes of nutrients were grossly the same in the two methods, except for protein, where a higher intake was recorded in the weighed record. In general, moderate agreement between the two methods was found. Conclusions: Participants were classified moderately correct according to food and nutrient intakes assessed in the pre-coded food diary; however values of absolute food intakes should be used and interpreted with caution. Improvement of the methods to estimate portion size may increase the accuracy of the dietary intake estimates.
Keywords
Food diary Weighed food recordRelative validity Cross-classificationThe Danish National Survey of Dietary Habits and Physical Activity is a nation-wide, representative, crosssectional survey collecting data on diet and physical activity with the aim of estimating intakes of foods and nutrients and monitoring trends and variations in habitual dietary patterns and physical activity in Denmark (1)(2)(3)(4) . Until now data have been collected for 8551 individuals (4041 males and 4510 females) aged 4-75 years, all randomly drawn from the centralised Civil Registration System to create representative samples of the Danish population. Data were collected from 2000-2002, 2003-2004 and 2005-2008 by means of a pre-coded food diary and a physical activity diary, for seven consecutive days. Furthermore, anthropometric and demographic variables, as well as lifestyle information and information on attitudes towards dietary habits, were collected through interviews prior to the food recording (5) . The pre-coded food diary applied in the Danish National Survey of Dietary Habits and Physical Activity to collect information on dietary intake contains pre-coded answers for the most commonly eaten foods and drinks in Denmark, and it is organised according to a typical Danish diet comprising breakfast, lunch, dinner and three snack meals. The participants were asked to fill in the diary prospectively for 7 d, thus not relying on remembering food intake in the past. Ongoing recruitment took...