[Dietary interview. 2. Comparison of the results obtained with different numbers of daily recalls]

From energy, nutrient and food intake data obtained with 14 24-hours recalls from 43 male and 53 female employees, aged from 30 to 40 years, the authors calculated and compared the single and group means for 1, 3, 5, 7, 10 and 14 days, respectively. It is tried to deduce from them the lowest number of interviews necessary for reliable statements. As the intra-individual variances were greater than the variances within the groups, the diet of such a group cannot be characterized on the basis of a single interview. Statements regarding the mean intakes of energy, nutrients and high-aggregated food groups could be made for groups of subjects on the basis of 3-5 interviews, with a tolerance of +/- 5%. The stable estimation of the consumption of single foods (with a tolerance of +/- 10%) required at least 7 interviews. 10 interviews were necessary for estimating (with a tolerance of +/- 10%) the mean intakes of energy, nutrients and great food groups of single persons. These results will hold true only when the workdays and weekends are in due proportion.