Techniques of measurement of body composition. Part I.

The measurement of lean body and fat mass has developed with the increase in sports participation and the prescription of exercise. Quantification of body fat is also related to the treatment of obesity and to assess nutritional status. Different levels of evaluation have been proposed depending on expertise and need. Body density estimation from skinfold measurements has the advantage of simplicity, low cost and reasonable validity with predictions to within 3 to 4% for 70% of the population. The choice of prediction equation will depend on whether a generalised equation or a population-specific equation is appropriate. In all cases tester reliability and standard error of estimation should be established. There is strong evidence that a quadratic equation should be applied and that measures of the lower limb should be included with circumference and diameter measures to strengthen the prediction. Methodological errors need to be reduced by careful training of experimenters. Crossvalidation of regression equation will strengthen their validity, particularly when fat loss is to be quantified. The popularity of skinfold assessment of body fat is enhanced by the use of nomograms to predict body fat, although some accuracy will be lost. Skinfold estimation of body fat will continue to be a useful guide to adiposity for epidemiological studies and for popular usage. The hydrostatic weighing procedure to estimate body density is considered by many to be the criterion method. Under carefully controlled conditions with maximum subject compliance, it is highly reliable, particularly if residual volume can be determined accurately. The conversion of body density to percentage body fat is based on a number of assumptions which need to be considered with respect to the population being studied. Simple methods for adiposity include various weight for height indices, for example, the body mass index of weight (kg) divided by height2 (m) which is often used to define obesity, frame size for the prediction of ideal weight and visual estimation for predicting body fat. Differences in the literature concerning fat cell size and number have their origins in variations of methodology. The choice of site for removal of tissue will influence the size of adipocytes, those obtained from deep sites are generally smaller than subcutaneous fat cells. If total cell number is determined from cell size, there will also be variability of cell number. The correlation between different methods of sizing cells is high for most techniques, but as the validity of the criterion method is not based on statistical evidence, the validity of the other methods cannot be readily accepted.(ABSTRACT TRUNCATED AT 400 WORDS)