Treadmill validation of an over-ground walking test to predict peak oxygen consumption

The purpose of this study was to determine whether a test developed to predict maximal oxygen consumption (VO2max) during over-ground walking, was similarly valid as a predictor of peak oxygen consumption (VO2) when administered during a 1-mile (1.61 km) treadmill walk. Treadmill walk time, mean heart rate over the last 2 full min of the walk test, age, and body mass were entered into both generalized (GEN Eq.) and gender-specific (GSP Eq.) prediction equations. Overall results indicated a highly significant linear relationship between observed peak VO2 and GEN Eq. predicted values (r = 0.91), a total error (TE) of 5.26 ml.kg-1.min-1 and no significant difference between observed and predicted peak VO2 mean values. The peak VO2 for women (n = 75) was predicted accurately by GSP Eq. (r = 0.85; TE = 4.5 ml.kg-1.min-1), but was slightly overpredicted by GEN Eq. (overall mean difference = 1.4 ml.kg-1.min-1; r = 0.86; TE = 4.56 ml.kg-1.min-1). No significant differences between observed peak VO2 and either GEN Eq. (r = 0.85; TE = 4.3 ml.kg-1.min-1) or GSP Eq. (r = 0.85; TE = 4.8 ml.kg-1.min-1) predicted values were noted for men (n = 48) with peak VO2 values less than or equal to 55 ml.kg-1.min-1. However, both equations significantly underpredicted peak VO2 for the remaining high peak VO2 men (n = 22). In conclusion, the over-ground walking test, when administered on a treadmill, is a valid method of predicting peak VO2 but underpredicts peak VO2 of subjects with observed high peak VO2 values.