Reproducibility and responsiveness between separate measurements of running energy cost of transport
Objectives: The metabolic cost of transport of human running (CoT) is a variable with a narrow range of changeover but a substantial individual variation. Thus, determining reproducibility and the responsiveness to complement inferential statistic is a consistent goal. This paper aimed to verify the reproducibility and responsiveness between separate measurements of CoT as well as to compare the CoT obtained in each measure between two running tests at different intensities. Design: An observational cross-sectional study. Method: Each volunteer performed two running assessments, on a treadmill, on two different days, according to running speed tested: a setup at 8.5 km∙h-1 and another at 14.2 km∙h-1. Three measurements of CoT in each setup were made. Pulmonary gas exchange was assessed breath by breath using a gas analyzer and CoT, in J∙kg-1∙s-1, determined following previous methodologies. Reproducibility was represented using intraclass correlation coefficient (ICC) and responsiveness using minimum detectable difference (MDD). Comparisons were made using generalized linear models. Results: Participants were five proficient male long-distance runners. The ICCs between CoT measurements were excellent. The CoT at 14.2 km∙h-1 (3.6 J∙kg-1∙s-1) was higher than that at 8.5 km∙h-1 (3.4 J∙kg-1∙s-1). However, the mean difference between CoT obtained (0.2 J∙kg-1∙s-1) in both setups was lower than the MDD (0.4 J∙kg-1∙s-1). Conclusions: Reproducibility for CoT between two different velocities was high, but the responsiveness was not good, indicating necessary changes in CoT, whether increased or decreased, to be around 12% considering that they are not consequences of individual variations or measurement errors.