CONICITY - A NEW INDEX OF BODY-FAT DISTRIBUTION - WHAT DOES IT TELL US

''Conicity'' (C) is an index of body fat distribution which expresses an individual's waist circumference relative to the circumference of a cylinder generated with that person's weight and height assuming a co nstant for body density (Valdez [1991] J. Clin. Epidemiol, 44:955-956) . The more central a person is in fat distribution, the higher the val ue of C. In a pilot study of cardiovascular reactivity and dimensions of anger and hostility in 60 African-, Anglo-, and Hispanic-American a dolescents, anthropometry and sexual maturation were determined to ass ess their mediating influence on the relation between anger and cardio vascular risk. The concurrent validity of three indices of body fat di stribution was explored: conicity (C), waist/hip ratio (WHR), and cent ral/peripheral skinfold ratio (C/P) by assessing their association wit h cardiovascular variables (CV) and Other anthropometrics. Anthropomet ry included height, weight, fat mass estimated from bioelectrical impe dance, four circumferences, and skinfolds. Maturation variables includ ed menarche in girls, testes size in boys, and pubic hair in both sexe s. Cardiovascular variables included resting diastolic and systolic bl ood pressures and heart rate, and the same three variables after recov ery from a step test. The boys and girls were 15-16 years of age, and there were equal numbers in each sex/ethnic group. Partial correlation s accounting for height (which was affected by ethnicity) and maturati on stratified by gender are reported. C was more strongly related to C V risk in boys and girls than the other indices or the body mass index , Least related to CV risk was C/P, correlated only weakly to central skinfold fat. C and WHR are highly related (0.85) yet differ in import ant respects in both sexes: C is more closely related to body fat (0.6 2-0.66) and fat mass (0.53-0.77) than the WHR (respective correlations : 0.54-0.55 and 0.43-0.66). Thus, C relates not only to body shape but also to body fat. Indices of central fat such as circumference and sk infold ratios, may ''over-correct'' for total fatness and thus miss im portant aspects of risk prediction. Conicity maybe a useful indicator of body fat distribution in studies of adolescents. (C) 1996 Wiley-Lis s, Inc.