Reliability of cycling gross efficiency using the Douglas bag method

Purpose: The aim of this study was to establish the reliability of gross efficiency (GE) measurement (the ratio of mechanical power input to metabolic power output, expressed as a percentage) using the Douglas bag method.

Methods: The experiment was conducted in two parts. Part 1 examined the potential for errors in the Douglas bag method arising from gas concentration analysis, bag residual volume, and bag leakage or gas diffusion rates. Part 2 of this study examined the within-subject day-to-day variability of GE in 10 trained male cyclists using the Douglas bag method. Participants completed three measurements of GE on separate days at work rates of 150, 180, 210, 240, 270, and 300 W.

Results: The results demonstrate that the reliability of gas sampling is high with a coefficient of variation (CV) <0.5% for both O2 and CO2. The bag residual volume CV was ∼15%, which amounts to +0.4 L. This could cause the largest error, but this can be minimized by collecting large gas sample volumes. For part 2, a mean CV of 1.5% with limits of agreement of +0.6% in GE units, around a mean GE of 20.0%, was found.

Conclusions: The Douglas bag method of measuring expired gases and GE was found to have very high reliability and could be considered the gold-standard approach for evaluating changes in GE. Collecting larger expired gas samples minimizes potential sources of error.