Deciphering the constrained total energy expenditure model in humans by associating accelerometer‑measured physical activity from wrist and hip Fernández‑Verdejo, Rodrigo Alcántara Alcántara, Juan Manuel Galgani, José E. Acosta Manzano, Francisco Miguel Migueles Hidalgo, Jairo Amaro Gahete, Francisco José Labayen, Idoia Ortega Porcel, Francisco Bartolomé Ruiz Ruiz, Jonatan Thanks to Dr. Herman Pontzer (Duke University) for his valuable feedback. We also thank the following agencies for their funding: Fondo Nacional de Desarrollo Científico y Tecnológico (11180361 to R.F.-V.); Spanish Ministry of Education, Culture and Sport (FPU15/04059 to J.M.A.A.; FPU15/02645 to J.H.M.; FPU14/04172 to F.J.A.-G.); University of Granada (Plan Propio de Investigación 2019 [Programa Contratos-Puente] to F.J.A.-G.; Plan Propio de Investigación 2016 [Excellence actions: Unit of Excellence on Exercise and Health UCEES]); Spanish Ministry of Economy and Competitiveness (ACTIBATE study; ACTIVEBRAINS study); Fondo de Investigación Sanitaria del Instituto de Salud Carlos III (PI13/01393 to ACTIBATE study); Junta de Andalucía, Consejería de Conocimiento, Investigación y Universidades and European Regional Development Funds (FEDER: ref. SOMM17/6107/UGR to ACTIBATE study); Redes Temáticas de Investigación Cooperativa RETIC (Red SAMID RD16/0022 to ACTIBATE study); EXERNET Research Network on Exercise and Health in Special Populations (DEP2005-00046/ACTI); Fundación Iberoamericana de Nutrición (ACTIBATE study); AstraZeneca HealthCare Foundation (ACTIBATE study); PTA 12264-I to FIT-AGEING study. Supplementary Information The online version contains supplementary material available at https://doi. org/10.1038/s41598-021-91750-x. The constrained total energy expenditure (TEE) model posits that progressive increases in physical activity (PA) lead to increases in TEE; but after certain PA threshold, TEE plateaus. Then, a compensatory reduction in the expenditure of non-essential activities constrains the TEE. We hypothesized that high PA levels as locomotion associate with a compensatory attenuation in arm movements. We included 209 adults (64% females, mean [SD] age 32.1 [15.0] years) and 105 children (40% females, age 10.0 [1.1] years). Subjects wore, simultaneously, one accelerometer in the non-dominant wrist and another in the hip for ≥ 4 days. We analyzed the association between wrist-measured (arm movements plus locomotion) and hip-measured PA (locomotion). We also analyzed how the capacity to dissociate arm movements from locomotion influences total PA. In adults, the association between wrist-measured and hip-measured PA was better described by a quadratic than a linear model (Quadratic-R2 = 0.54 vs. Linear-R2 = 0.52; P = 0.003). Above the 80th percentile of hip-measured PA, wrist-measured PA plateaued. In children, there was no evidence that a quadratic model fitted the association between wrist-measured and hip-measured PA better than a linear model (R2 = 0.58 in both models, P = 0.25). In adults and children, those with the highest capacity to dissociate arm movements from locomotion—i.e. higher arm movements for a given locomotion—reached the highest total PA. We conclude that, in adults, elevated locomotion associates with a compensatory reduction in arm movements (probably non-essential fidgeting) that partially explains the constrained TEE model. Subjects with the lowest arm compensation reach the highest total PA. 2021-06-18T09:42:14Z 2021-06-18T09:42:14Z 2021-06-10 journal article Fernández-Verdejo, R., Alcantara, J.M.A., Galgani, J.E. et al. Deciphering the constrained total energy expenditure model in humans by associating accelerometer-measured physical activity from wrist and hip. Sci Rep 11, 12302 (2021). [https://doi.org/10.1038/s41598-021-91750-x] http://hdl.handle.net/10481/69272 10.1038/s41598-021-91750-x eng http://creativecommons.org/licenses/by/3.0/es/ open access Atribución 3.0 España Springer Nature