Body fat distribution has been reported as a significant risk factor for the development of type 2 diabetes mellitus and other forms of Glucose Intolerance. Anthropometric Indices including BMI, WC, and WHR are used as surrogate measures of body fat distribution. Definitive values of BMI, WC, and WHR are used to characterize individuals into general or central obesity. Few studies have evaluated the relationship between anthropometric indices and Glucose Intolerance in Nigerian population. This study was carried out to compare the discriminatory ability of these indices in correctly identifying individuals with glucose intolerance among Urban adults in Jos. OBJECTIVES To evaluate the anthropometric Indices of BMI, WC, WHR as correlates of Glucose Intolerance and to determine the prevalence of IFG, IGT, and DM, in the study population and also the prevalence of obesity in the population. SUBJECTS, MATERIALS AND METHODS A three-stage random sampling was used to enroll 800 subjects ≥20 years of age residing in Jos. Ill and pregnant subjects were excluded. Out of the 800 subjects, 709 responded and participated in the study giving a response rate of 88.6%. Demographic and clinical data were obtained using a modified WHO STEPS Questionnaire. Anthropometric indices including weight, height, waist circumference and hip circumference were measured. BMI and WHR were calculated. Fasting plasma glucose (FPG) was estimated after 8-12 hours overnight fast. Oral Glucose tolerance test (OGTT) with 75g glucose was also carried out for all subjects. Fasting plasma Insulin was assayed in a subset of 119 subjects to determine Insulin Resistance using Homeostasis model assessment ( HOMA-IR). Data was analyzed with Epi info 3.6.1.and SPSS 16.0. Receiver-Operator characteristic curve was used to test the prediction of glucose intolerance by the anthropometric indices. All p-values less than 0.05 was considered significant. RESULTS The subjects were made up of 43.4% of males and 56.6% of females. The mean (SD) age of subjects was 43.21 (14.74) years with a range of 20-89. There was no significant difference in mean age of male and females. Mean (SD) WC and BMI were higher in females than males [WC: 87.5(12.5)cm vs 85.2(10.8)cm, p=0.009, BMI: 27.4 (5.1)kg/m2 vs 25.3(3.7)kg/m2 , p=0.001] while WHR was higher in males than females [0.88(0.07) vs 0.86(0.07), p=0.02]. The differences were statistically significant. The overall prevalence of IFG was 2.7% (3.2% in men, 2.2% in women, p=0.41) while that of IGT was 7.5% (5.5% in men, 8.5% in women, p=0.13). The prevalence of DM in the study population was 4.1% (4.5% men, 3.7% in women, p=0.59). The prevalence of IFG and DM were higher in men while that of IGT was higher in women, however none was significant. The prevalence of Overweight and Obesity using BMI was 35% and 21% (10.7% in men, 28.9% in women, p=0.0001) respectively. The prevalence of Central Obesity using WHR was 52.6% (39.3% in men, 63.1% in women, p=0.0001) while it was 30.7% (8.8% in men, 47.6% in women, p=0.0001) using WC (NCEP 111) and 47.8% (21.8% in men, 67.8% in women, p=0.0001) using WC (IDF). Obesity was significantly commoner in women than men. Receiver – Operator Characteristic Curve showed that all the studied indices had significantly higher AUC than the Null hypothesis true area of 0.5, but WHR had the highest Area under the Curve (AUC) for IFG, IGT and DM in both men and women. It performed better than WC and BMI in that order. CONCLUSION The results from this study showed that the prevalence of IGT was higher than that of IFG (7.5% vs 2.7%). There was also higher prevalence of IGT than DM (7.5% vs 4.1%) with implications for the future prevalence of DM. Prevalence of Central Obesity using WHR was higher than using WC and also higher than generalized obesity with BMI (52.6% vs 30.7% vs 21%). Out of the three tested anthropometric indices (BMI, WC, WHR), WHR had the best ability to identify Individuals with Glucose Intolerance (GI) in the studied population. WHR, a measure of central obesity is better correlated to Glucose Intolerance than other anthropometric indices