Improving the accuracy of prediction of an individual's maximum oxygen intake

Abstract

Maximum oxygen intakes, heights, weights and skinfold thicknesses were measured on 106 Bantu male recruits to the gold mines. Three different tribes from different parts of Africa were represented in the sample. Regression equations were calculated for the relationships between maximum oxygen intakes and 1. gross body weights, 2. gross body weights and heights, and 3. gross body weights, heights and per cent body fat. Residual mean squares were also calculated. From these results it can be estimated that 27 per cent of the variations between individuals in maximum oxygen intakes can be accounted for by their differences in gross body weight and a further 3 per cent by differences in height (weight is positively correlated and height negatively correlated with maximum oxygen intake). Excluding the effect of differences in per cent body fat does not alter the residual mean square so that none of the variations between individuals in maximum oxygen intake can be attributed to this factor. Treating the tribal groups separately also did not reduce the residual mean squares of any of the three tribes. Of the factors measured in this experiment only differences in weight and height in the Bantu contribute to the variations between individuals in maximum oxygen intake and 70 per cent of the variations is left unaccounted for.

These results differ from those ofBuskirk andTaylor on U.S. Army recruits where 40 per cent of the variation in maximum oxygen intake was due to differences in gross body weight and where, when per cent body fat was eliminated, 70 per cent of the variations could be accounted for. A possible reason for this difference is that the Bantu subjects wereall uniformly thin. The mean per cent of body fat in the Bantu was 4.19 per cent compared with 13.7 in the Caucasians and the variations between different Bantu were also very small. Having very little body fat, one might have expected the percentage of the variation in maximum oxygen intake, which is accounted for by differences in gross body weight, to be nearer the Caucasian figure of 70 per cent (for lean body mass) than the actual figure of 27 per cent. However, there are probably large differences between these men in their nutrition, health and levels of activity on recruitment and all affect the maximum oxygen intake.