In evaluating dietary data with reference to guidelines for a healthy diet, some potential pitfalls (i.e., method of food consumption assessment and calculation to include or exclude energy derived from alcohol) were investigated. The percentage of energy intake (en%) derived from total fat, saturated fatty acids (SFA), mono- and disaccharides (MD) and intake of cholesterol and dietary fiber per megajoule were calculated using 2-day records obtained in the Dutch National Food Consumption Survey of 1987-1988. Subjects (aged 4-85, n = 5595) were classified into age-sex groups. Mean values and intake distributions were calculated with and without adjustment for within-person variation. Except for the intake of cholesterol and MD, mean intake levels were not in accord with guidelines. About 20% of the women and 23% of the men met the goal for fat intake, 20% of the men and 27% of the women for dietary fiber, and about 60 and 70% for MD and cholesterol. Only 3% of subjects had SFA intake < or = 10 en%, whereas < 1% had a diet in accord with all criteria studied. After adjustment for within-person variation, percentages meeting the guidelines were generally lower for fat, SFA and dietary fiber and slightly higher for cholesterol and MD. Among elderly, unadjusted results were more in agreement with the prevalence estimates based on habitual dietary intake data than with adjusted ones. Fat intake (en%) was inversely related with intake of added MD and alcohol. Our data indicate that guidelines should state explicitly whether energy-related recommendations include energy derived from alcohol, and that the prevalence of a high-fat intake is more affected by the calculation method than by food consumption assessment.