Validity and reliability of BodyGem for measuring resting metabolic rate on Taiwanese women.

Resting metabolic rate (RMR) accounts for about two thirds of total energy expenditure. The widely used Harris-Benedict equations systematically overestimate RMR. This study assessed overall reliability and validity of a handheld indirect calorimeter, BodyGem, on a sample of women. Thirty healthy nurses participated in this study with an age of 41.9+/-9.0 years old and a body mass index of 24.0+/- 2.8 kg/m2. The Deltatrac Metabolic Monitor was used as the criterion method to validate BodyGem. Reliability was estimated by repeated measures of BodyGem to test internal consistency and stability. Analysis indicated that measurements of Deltatrac and BodyGem are well correlated (r=0.76, P<0.001). The correlation coefficients of two BodyGem RMR measurements were of large statistical significance (r=0.96, P<0.001, mean difference = 15.8+/-55.8 Kcal/d). A significant difference (F =3.81, P=0.04) in repeated measures ANOVA and post hoc revealed a difference between BodyGem and Deltatrac. There was a systematic difference between both methods (mean difference between BodyGem and Deltatrac =36.4 to 52.2 Kcal/d). After adjustment of additional energy demand by holding BodyGem in position, the difference became non-significant (F =1.62, P=0.22). Bland-Altman plots revealed that there was no significant trend in both methods, and repeated measurements of Bodygem. In conclusion, RMR obtained using the BodyGem has a high degree of reproducibility and an acceptable validity compared to the Deltatrac. Further validity research is needed in Taiwanese women.