Energy expenditure measured by doubly labeled water, activity recall, and diet records in the rural elderly

OBJECTIVE: The purpose of this study was to determine whether energy expenditure estimated from physical activity and energy intake were equivalent to total daily energy expenditure in an elderly rural population. METHODS: Twenty-seven elderly male (n = 14) and female (n = 13) subjects (mean age, 74 y) were recruited from a rural Pennsylvania population. Over a 2-wk period, total daily energy expenditure was measured by doubly labeled water (TEE) and estimated from 7-d physical activity recall factors multiplied by weight (PA_WT), estimated basal metabolic rate (PA_BMR) and resting energy expenditure from indirect calorimetry (PA_REE), and energy intake from 3-d self-reported diet records (EI). Analysis of variance was used to determine significant within-subject differences in physical activity, energy intake, and energy expenditure. RESULTS: PA_REE (men: 13.69 ± 3.23 MJ, women: 9.51 ± 2.40 MJ) and PA_BMR (men: 13.69 ± 2.99 MJ, women: 10.15 ± 2.21 MJ) were not significantly different from TEE (men: 12.43 ± 1.63 MJ, women: 9.44 ± 0.90 MJ). EI (men: 8.66 ± 2.34 MJ, women: 7.12 ± 0.93 MJ) was significantly less than TEE, and PA_WT (men: 17.03 ± 4.07 MJ, women: 12.86 ± 3.41 MJ) was significantly greater than TEE. CONCLUSIONS: Whereas 7-d physical activity recall determined with an age- and gender-specific estimate of resting metabolic rate or measured using indirect calorimetry accurately estimated TEE for this group of rural elderly, self-reported diet records consistently underestimated and physical activity recall determined with weight alone consistently overestimated energy expenditure measured by doubly labeled water.