ASSESSMENT OF A MODIFIED HOUSEHOLD FOOD WEIGHING METHOD IN A STUDY OFBONE HEALTH IN CHINA

Objective: A 'modified' household dietary method to estimate individua l intakes from total household intakes was evaluated in comparison to the individual food weighing method, i.e. direct weighing of foods con sumed by individuals. Design: Foods consumed by the household were wei ghed and recorded over a 3 day period and were proportionally allocate d to a specific person in the household by using energy conversion fac tors based on age, gender and physical activity. The individual dietar y intakes were standardized by body weight (ratio of individual body w eight to reference body weight from Chinese Recommended Daily Allowanc es (Acta Nutr. Sin. 12, 1-9), and were compared with the estimates fro m the individual food weighing method for both accuracy and associatio ns with bone density. Setting: The subjects were randomly selected fro m five rural countries in China, characterized by different dietary pa tterns. Subjects: A total of 712 women aged 35-75 years participated i n the dietary and bone measurements. Results: Nutrient intakes estimat ed by the modified household method (adjusting for body weight were ve ry similar to those obtained by the individual method (r=0.53-0.78, P< 0.001), except for sodium which had the largest deviation from the ind ividual methods (r=0.23, P<0.001). Calcium intakes were 405 +/- 271 an d 409 +/- 323 mg/day, respectively, as estimated by the individual and modified household methods. A very similar degree of correlation with radial bone density also was obtained for both calcium estimates. Con clusion: The results indicate that the modified household method is ap propriate for assessing food intakes of individuals in large nutrition al studies.