• 1 July 1984
    • journal article
    • Vol. 38  (4) , 299-306
Abstract
Measurements of body composition were made on 104 women aged 14-60 years whose fat content varied from 6 per cent to 60 per cent of body weight. Estimates of fat content were systematically lower when based on a measurement of body density than when based on body water, and were higher still when based on total body potassium. When body weight and body fat were corrected for stature by dividing by height squared there was a coefficient of correlation between these two variables of 0.936, 0.921 and 0.938 for estimates of fat based on density, water and potassium respectively. When the mean of all three estimates of fat was used the correlation was 0.955. Neither the slope of the regression line nor the strength of the correlation was affected if gross body weight was correlated with total body fat without correction for stature. It is concluded that differences in weight between women of similar height is attributable to tissue which is 70-78 per cent fat and 22-30 per cent lean, and that in the treatment of obese patients it is desirable that no more than 22 per cent of the weight loss should be lean tissue.

This publication has 0 references indexed in Scilit: