Objective the purpose of this study was to test a new hand-held device, the BodyGem (HealtheTech Inc., Golden, CO), that measures resting metabolic rate (RMR). Subjects/design Sixty-three adults (43 women, 20 men) (mean+/-SD, age 41.3+/-11.2 years and body mass index, 26.5+/-6.6 kg/m(2)) were tested during two separate sessions within a 2-week period, and, in each session, two BodyGem and two Douglas bag RMR measurements were made in a random and counterbalanced order. Main outcome measures Resting oxygen consumption and energy expenditure. Statistical analyses performed A 2 (methods) X 4 (time points) repeated measures ANOVA, Pearson product-moment coefficients, Bland-Altman plots. Results Within session reliability for measurement of 0, consumption was high on both days for the BodyGem (r=0.97). During the 4 single tests, BodyGem and Douglas bag O-2 consumption values were significantly correlated (r=0.81-0.87), with SEEs ranging from 22 to 28 mL . min(-1). Mean 02 consumption and RMR values for all 4 tests were 241+/-46 and 240+/-45 mL . min(-1) (r=0.91, SEE 18.7 mL min(-1)) and 1,657+/-324 and 1,650+/-307 kcals . day(-1) (r=0.91, SEE 134 kcals . day(-1)) for the BodyGem and Douglas bag methods, respectively. Applications/conclusions These data indicate that the BodyGem is an accurate and reliable device for measuring oxygen consumption and calculating RMR during repeated tests within a day, single tests on separate days, or when measurements are averaged.