Reducing the variability of oxygen consumption measurements.

The oxygen consumption (O(2)) of 10 able-bodied adults each walking at a variety of cadences and hence speeds was measured. The effect on variability of subtracting the resting oxygen rate from gross measurements and of normalising walking speed to the subject's height was investigated. Both of these were indeed found to reduce variability. A total reduction of more than 40% in the variability of O(2) cost measurements was observed. Using these techniques it was found that, at walking speeds below the average, nett O(2) cost per height was very nearly independent of walking speed.