Reliability and validity of the COSMED K5 portable metabolic system during walking

Purpose: Portable methods for assessing energy expenditure outside the laboratory and clinical environments are becoming more widely used. As such, it is important to understand the accuracy of such devices. Therefore, the purpose was to determine the reliability and validity of the COSMED K5 portable metabolic system. Methods: Reliability and validity were assessed in 27 adults (age: 27 ± 5 years; n = 15 women) using a walking protocol. The protocol consisted of a 5-min walk/2-min rest cycle starting at 1.5 mph and increasing in 0.5-mph increments to 4.0 mph. During visit one, participants wore the K5 to assess oxygen consumption (V ˙ O₂), carbon dioxide production (V ˙ CO₂), and other metabolic variables. Two to seven days later, the protocol was repeated twice with the COSMED K5 and K4b² systems in a randomized, counterbalanced order. Results: Intraclass correlation coefficients (ICC) revealed that the K5 reliably measured V ˙ O₂ (ICC 0.64–0.85) and V ˙ CO₂ across all walking speeds (ICC 0.50–0.80), with stronger reliability at faster walking speeds compared with slower speeds. Moderate-to-strong relationships were observed for measured gases between the K5 and K4b². Specifically, V ˙ O₂ exhibited a moderately high-to-high relationship between devices (r = 0.72–0.82), and a similarly moderately high-to-high relationship was observed for V ˙ CO₂ (r = 0.68–0.82). While there were no differences in V ˙ O₂ measured between devices (p ≥ 0.10), the K5 provided lower V ˙ CO₂ readings than the K4b² during the 3.0, 3.5, and 4.0 mph walking speeds (p ≤ 0.02). Conclusions: The K5 provided reliable and valid measures of metabolic variables, with greater reliability and validity at faster walking speeds.