[Intra- and inter-individual variations in diets of the middle-aged and the elderly].

This study was conducted to examine intra- and inter-individual variations in diets of the middle-aged and the elderly (40 years or older, 46 men and 42 women). The coefficients of variations for intakes of nutrients and food groups were computed from four 4-day weighed dietary records performed at 3-month intervals from June 1996. The results were as follows: a) The highest intra-individual variation (%) for nutrient intake was observed in retinol (men 293.5, women 283.8) and the lowest in carbohydrate (men 17.7, women 22.1). b) The highest inter-individual variation (%) was found in retinol (58.2) in men, and in carotene (56.7) in women. The lowest inter-individual variation (%) was observed in magnesium (17.0) in men, and in carbohydrate (14.4) in women. c) Nuts and seeds showed the highest intra-individual variation (%) for food group intake (men 291.5, women 391.8), while rice presented the lowest (men 30.5, women 38.9). d) The highest inter-individual variation for food group intake (%) was seen in milk and dairy products (111.7) in men and in alcoholic beverages (162.3) in women. The lowest inter-individual variation was observed in potatoes and starches (20.7) in men and in pulses (26.0) in women. e) The number of days necessary to estimate true average nutrient intake was much longer for such vitamins as retinol and carotene (over 50 days) than for macronutrients (3-5 days) except for fat. More than one year was required to estimate intake of nuts and seeds in both sexes and alcoholic beverages or seaweeds in women, whereas only 9-15 days for intake of rice. In conclusion, energy, protein and carbohydrate can be estimated by short-period dietary recalls or records, since their intra-individual variations were relatively small. On the other hand, many days, were found to be required to estimate usual dietary intake of such vitamins as retinol or carotene and that of each food group except for rice. It would therefore be very difficult to estimate usual intake of these nutrients and food groups by short-period dietary recalls or records.