Four-day multimedia diet records underestimate energy needs in middle-aged and elderly women as determined by doubly-labeled water

Systematic problems exist in the quantification of food intake in populations using traditional self-reported measures. The objective of this study was to determine the effectiveness of an innovative multimedia diet record (MMDR) for dietary energy intake assessment. Dietary intake was estimated by combining the use of a microcassette tape recorder and 35-mm camera in 53 women whose ages ranged from 50 to 93 y (64.9 +/- 11.3 y), with body weights of 62.4 +/- 12.2 kg and body mass indexes (BMI) of 24.4 +/- 4.0 kg/m(2). Using household measures, subjects voice-recorded and photographed all food and beverages consumed for four consecutive days. A two-point doubly-labeled water (DLW) method was used over 13 d to calculate carbon dioxide production, total body water, and subsequently, total energy expenditure (TEE) through the use of a food quotient, Mean body weight did not change between d 1 and 14, TEE and reported energy intake were compared using MMDR, Mean reported energy intakes 7.5 +/- 1.9 MJ/d (1774 +/- 476 kcal/d) were lower (P < 0.01) than TEE by 10.4 +/- 3.1 MJ/d (2477 +/- 736 kcal/d), indicating underreporting of food intake. Reporting accuracy (reported energy intake/TEE' 100%) was 76.0 +/- 22.9%. Mean energy expenditure (MJ/d), as determined by doubly-labeled water, was higher (P < 0.01) in each stratified age range when compared to reported energy intake by MMDR, There were no significant differences in reporting accuracy among the stratified age groups. Using the MMDR method, this population of weight-stable women underreported their food intakes compared to their determined energy expenditure estimated by DLW.