Evaluating the accuracy of using fixed ranges of METs to categorize exertional intensity in a heterogeneous group of healthy individuals: implications for cardiorespiratory fitness and health outcomes

Background: Appropriate quantification of exertional intensity remains elusive. Objective: To compare, in a large and heterogeneous cohort of healthy females and males, the commonly used intensity classification system (i.e., light, moderate, vigorous, near-maximal) based on fixed ranges of metabolic equivalents (METs) to an individualized schema based on the exercise intensity domains (i.e., moderate, heavy, severe). Methods: A heterogenous sample of 565 individuals (females 165; males 400; age range 18-83 years old) were included in the study. Individuals performed a ramp-incremental exercise test from which gas exchange threshold (GET), respiratory compensation point (RCP) and maximum oxygen uptake (VO2max) were determined to build the exercise intensity domain schema (moderate = METs ≤ GET; heavy = METs > GET but ≤ RCP; severe = METs > RCP) for each individual. Pearson's chi-square tests over contingency tables were used to evaluate frequency distribution within intensity domains at each MET value. A multi-level regression model was performed to identify predictors of the amplitude of the exercise intensity domains. Results: A critical discrepancy existed between the confines of the exercise intensity domains and the commonly used fixed MET classification system. Overall, the upper limit of the moderate-intensity domain ranged between 2 and 13 METs and of the heavy-intensity domain between 3 and 18 METs, whereas the severe-intensity domain included METs from 4 onward. Conclusions: Findings show that the common practice of assigning fixed values of METs to relative categories of intensity risks misclassifications of the physiological stress imposed by exercise and physical activity. These misclassifications can lead to erroneous interpretations of the dose-response relationship of exercise and physical activity.