In clinical practice, in the absence of objective measures, simple methods to predict energy requirement in patients with chronic obstructive pulmonary disease (COPD) needs to be evaluated. The aim of the present study was to evaluate predicted energy requirement in females with COPD using pedometer-determined physical activity level (PAL) multiplied by resting metabolic rate (RMR) equations.
Energy requirement was predicted in 18 women with COPD using pedometer-determined PAL multiplied by six different RMR equations (Harris-Benedict; Schofield; World Health Organization; Moore; Nordic Nutrition Recommendations; Nordenson). Total energy expenditure (TEE) was measured by the criterion method: doubly labeled water. The predicted energy requirement was compared with measured TEE using intraclass correlation coefficient (ICC) and Bland-Altman analyses.
The energy requirement predicted by pedometer-determined PAL multiplied by six different RMR equations was within a reasonable accuracy (±10%) of the measured TEE for all equations except one (Nordenson equation). The ICC values between the criterion method (TEE) and predicted energy requirement were: Harris-Benedict, ICC =0.70, 95% confidence interval (CI) 0.23-0.89; Schofield, ICC =0.71, 95% CI 0.21-0.89; World Health Organization, ICC =0.74, 95% CI 0.33-0.90; Moore, ICC =0.69, 95% CI 0.21-0.88; Nordic Nutrition Recommendations, ICC =0.70, 95% CI 0.17-0.89; and Nordenson, ICC =0.40, 95% CI -0.19 to 0.77. Bland-Altman plots revealed no systematic bias for predicted energy requirement except for Nordenson estimates.
For clinical purposes, in absence of objective methods such as doubly labeled water method and motion sensors, energy requirement can be predicted using pedometer-determined PAL and common RMR equations. However, for assessment of nutritional status and for the purpose of giving nutritional treatment, a clinical judgment is important regarding when to accept a predicted energy requirement both at individual and group levels.