SELECTED NUTRIENT INTAKES OF A COHORT OF CANADIAN POSTMENOPAUSAL WOMEN ESTIMATED FROM 3-DAY DIETARY RECORDS - SOURCES OF VARIANCE IN NUTRIENT INTAKES AND PROBABILITY ESTIMATES OF THEIR ADEQUACY

Three-day estimated dietary records were obtained from 94 non-institutionalized postmenopausal Canadian women (mean age 66.8 years; range 58-89 years), consuming self-selected diets and who were living in Guelph. Intakes of energy and selected nutrients were calculated from food composition data and classified into three age groups (n=38; 55-64 years; n=37; 65-74 years; n=12; 75 years and over). The only significant differences in nutrient intakes (per day; per 1000 kcal) among the three age groups were for vitamin C and vitamin A. Of the subjects, 66% were taking nutrient supplements. Fifty-six per cent, 30% and 27% of the subjects had dietary calcium, vitamin A and D intakes below the Canadian Recommended Nutrient Intakes compared to probability estimates for risk of inadequacies of 25%, 16% and 15%, respectively. Five components of variance were estimated. The inter-subject variance within age groups explained most of the variance for each nutrient. In all cases, except for vitamin A, the inter-subject variance was greater than the intra-subject variance suggesting that the diets within the subjects are rather monotonous, despite large variations among subjects. For vitamin A, both inter- and intra-subject coefficients of variation were high, emphasizing the difficulty of obtaining a precise estimate of usual intake of vitamin A for individuals. © 1990 Pergamon Press plc.