Crossvalidation of two heart rate-based equations for predicting VO2max in white and black men

The purpose of this investigation was to crossvalidate 2 equations that use the ratio of maximal heart rate (HRmax) to resting HR (HRrest) for predicting maximal oxygen consumption (VO2max) in white and black men. One hundred and nine white (n = 51) and black (n = 58) men completed a maximal exercise test on a treadmill to determine VO2max. The HRrest and HRmax were used to predict VO2max via the HRindex and HRratio equations. Validity statistics were done to compare the criterion versus predicted VO2max values across the entire cohort and within each race separately. For the entire group, VO2max was significantly overestimated with the HRindex equation, but the HRratio equation yielded no significant difference compared with the criterion. In addition, there were no significant differences shown between VO2max and either HR-based prediction equation for the white subgroup. However, both equations significantly overestimated VO2max in the black group. Furthermore, large standard error of estimates (ranging from 6.92 to 7.90 ml·kg(-1)·min(-1)), total errors (ranging from 8.30 to 8.62 ml·kg(-1)·min(-1)), and limits of agreement (ranging from upper limits of 16.65 to lower limits of -18.25 ml·kg(-1)·min(-1)) were revealed when comparing the predicted to criterion VO2max for both the groups. Considering the results of this investigation, the HRratio and HRindex methods appear to crossvalidate and prove useful for estimating the mean VO2max in white men as a group but not for an age-matched group of black men. However, because of inflated values for error, caution should be exercised when using these methods to predict individual VO2max.