The predictive equations of resting metabolic rate for Tibetan adolescents aged 13–18 in Tibet, China

Abstract

Objectives The resting metabolic rate (RMR) predictive equations suitable for Tibetan adolescents in Tibet, China, were developed to provide a reference for their reasonable energy intake. Methods We measured RMR by indirect calorimetry and body composition by bioelectrical impedance analysis in 325 Tibetan adolescents aged 13–18 years in Tibet, China. Stepwise regression analysis was used to develop the predictive equations. Pearson correlation analysis, paired sample t test, bias rate, concordance correlation coefficient (CCC) and Bland–Altman were used to verify the validity of the predictive equations. Results The R2 of Model 8 (0.642) was larger than Model 2 (boys, 0.642; girls, 0.533) and Model 7 (0.540), and Model 10 (0.534) was larger than Model 4 (boys, 0.531; girls, 0.443) and Model 9 (0.477).Compared with the existing predictive equations, the correlation (0.68–0.84) between the predicted values and the measured values, CCC (0.74–0.81) and consistency (Similar proportions within the upper and lower limits but lower differences) were higher and the bias rate (−1.0% to −2.5%) and root mean square error (207.4–263.7 kcal/day) were lower in this study. By comprehensive comparison, Model 8 and Model 10 were more valid. Conclusions The existing predictive equations cannot accurately predict the RMR of Tibetan adolescents in Tibet, China. In this study, the age segmentation predictive equations with age, sex, and fat free mass (FFM) as independent variables were more valid. The predictive equations were as follows: RMR (kcal/day) = 50.1 × FFM (kg) − 202.8 × Sex (F: 0; M: 1) − 72.1 × Age + 930.3, 13–15 years; RMR (kcal/day) = 58.4 × FFM (kg) − 441.1 × Sex (F: 0; M: 1) − 702.2, 16–18 years.