VALIDATION OF A SHORT QUALITATIVE FOOD FREQUENCY LIST USED IN SEVERALGERMAN LARGE-SCALE SURVEYS

Our study aimed to test the validity of a short, qualitative food freq uency list (FFL) used in several German large scale surveys. In the su rveys of the MONICA project Augsburg, the FFL was used in randomly sel ected adults. In 1984/85, a dietary survey with 7-day records (DR) was conducted within the subsample of men aged 45 to 64 (response 70 %). The 899 DR were used to validate the FFL. Mean weekly food intake freq uency and mean daily food intake were compared and Spearman rank order correlation coefficients and classification into tertiles with values of the statistic Kappa were calculated. Spearman correlations range b etween 0.15 for the item ''Other sweets (candies, compote)'' and 0.60 for the items ''Curds, yoghurt, sour milk'', ''Milk including butter m ilk'' and ''Mineral water''; values for statistic Kappa vary between 0 .04 (''White bread, brown bread, crispbread'') and 0.41 (''Flaked oats , muesli, cornflakes'' and ''milk including butter milk''). With the e xception of two items, FFL data can be used for analysis on group leve l. Analysis on individual level should be done with caution. It seems, as if some food groups are generally easier to ask for in FFL than ot hers.