Validation of generalized anthropometric equations for body density estimate in military women

Abstract

The validation of the equations for predicting body density allows verifying the accuracy of estimated body density on a population different from that used in the original equation. In this way, the purpose of this study was to verify the validity of six generalized prediction equations of body density in military women. The sample was composed by 107 military women, with ages varying from 18 to 45 years (30.48 ± 6.40 years), total body mass of 58.60 ± 6.99kg and stature of 164.78 ± 5.81cm. Six skinfolds and one circumference were measured according to the selected equations. To estimate body density, the indirect method of hydrostatic densitometry (validation criterion) and the doubly indirect methods of prediction equations by Jackson et al. (1980, 3 skinfolds), and by Petroski (1995) were used. Mean body density measured by hydrostatic densitometry was 1.046058 ± 0.011001 g/ml. To validate the equations Pearson´s linear correlations were used and the results were considered acceptable (r = 0.71 to 0.76). When comparing predicted to measured density, it was found that only the quadratic equation using 5 skinfolds by Petroski (1995) did not show any statistical difference (t = 0.573 to p value = 0.568). The correlation coefficient for this equation was r = 0.74, which is also considered acceptable. Total error was ET =0.0075 g/ml and the standard error of the estimate was EPE = 0.0066 g/ml, both onsidered very good values for the validation process. These results permit the conclusion that the generalized quadratic equation by Petroski (1995), using 5 skinfolds (subscapular, triceps, suprailiac, abdominal and medial calf) is concurrently valid for estimating body density among military women.