of work: In collaboration with scientists from the National Institute for Longevity Sciences in Nagoya, comprehensive analyses of the relation of anthropometric and metabolic variables have been performed. Body Mass Index and Percent Body Fat (from DEXA) as well as estimates of the pattern of fat distribution are being related to ten well-established coronary heart disease risk factors (systolic and diastolic blood pressures, fasting and two-hour post-glucose plasma glucose concentration, fasting plasma insulin and insulin resistance (HOMA), and four plasma lipid moieties) in younger and older men and women. Waist circumference has emerged as the dominant measure of fat distribution. The NHLBI """"""""Evidence Report"""""""" on obesity was issued in July 1998 and provided single health cutpoints for the waist circumference measurements in men (102 cm) and in women (88 cm); thus two zones for the body fat distribution were established. No """"""""grey zone"""""""" for waist circumference was defined, although the BMI range was stratified into six diagnostic categories. A 1998 WHO report however recommended cutpoints of 94 and 102 cm for men and 80 and 88 cm for women, thus creating normal, moderately high risk, and very high risk zones. We have tested the applicability of these WHO standards to younger and older men and women. Results of the analyses in BLSA participants show clearly that both the BMI and waist circumference individually remain very significant determinants of the traditional coronary heart disease risk factors in old age. The waist circumference (WC) however, is highly correlated with the body mass index. Thus, a major question is whether the assessment of WC adds significantly to the predictive power of the BMI. This issue is of importance in deciding whether WC is simply a surrogate for BMI (or vice versa), that is, whether measurement of WC is of any practical importance in risk assessment. We therefore used the statistical technique of partial regressional logistic regression to examine this question. The results are clear: (1) in both younger men and younger women, WC improves the prediction for multiple risk factors; (2) in the older age groups (65 + yr), WC offers no improvement in predicting risk in either men or women.
