COMPARISON OF BODY-SIZE MEASUREMENTS AS PREDICTORS OF NIDDM IN PIMA-INDIANS

OBJECTIVE-To determine and compare the abilities of various anthropome tric measurements to predict the development of non-insulin-dependent diabetes mellitus (NIDDM) in Pima Indian men and women. RESEARCH DESIG N AND METHODS-A total of 290 male and 443 female Pima Indians were fol lowed for up to 6 years for the development of NIDDM. A proportional h azards analysis was used to assess the ability of anthropometric measu rements evaluated at baseline to predict NIDDM. Receiver operating cha racteristic (ROC) curves were used to compare individual variables in predicting NIDDM. RESULTS-In separate models controlled for age and se x, body mass index (BMI), waist circumference, thigh circumference, wa ist-to-thigh ratio (WTR), weight, and percentage body fat (PBF) estima ted by bioelectric resistance each predicted NIDDM, which developed in 30 men and 52 women. The highest incidence rate ratios (IRRs; for 1 S D of a variable) were for WTR in men and for PBF in women, although th e confidence interval (CI) for PBF was wide. In stepwise analyses, WTR was the most significant predictor in men (IRR for 1 SD = 1.58, 95% C I = 1.20-2.07), and BMI was the most significant predictor in women (I RR for 1 SD = 1.65, 95% CI. = 1.29-2.11). However, by ROC analyses, th igh circumference was the only variable significantly worse than WTR i n men or BMI in women at predicting NIDDM. CONCLUSIONS-Measurements su ch as waist circumference, WTR, weight, and BMI may be as useful as mo re complicated measurements, such as PBF by bioelectric resistance, fo r identifying groups of individuals whose body habitus places them at high risk of developing NIDDM.