We sought to determine which measures of adiposity can predict cardiovascular disease (CVD) risk and to evaluate the extent to which overall and abdominal adiposity are associated with cardiometabolic risk factors among working adults in Ethiopia. This was a cross-sectional study of 1,853 individuals (1,125 men, 728 women) in Addis Ababa, Ethiopia. The World Health Organization STEPwise approach was used to collect sociodemographic data, anthropometric measurements, and blood samples among study subjects. Fasting blood glucose (FBG) and lipid concentrations were measured using standard approaches. Spearman's rank correlation, receiver operating characteristic (ROC) curves, and logistic regression were employed to determine the association and predictive ability (with respect to CVD risk factors) of four measures of adiposity: BMI, waist circumference (WC), waist-to-hip ratio (WHR), and waist-to-height ratio (WHtR). Overall, FBG is best associated with WHtR in men and WC in women. Systolic blood pressure (SBP) is most strongly associated with BMI in men, but with WC in women. Compared to those with low BMI and low WC, the risk of having CVD is the highest for those with high BMI and high WC and those with high WC and low BMI. Review of ROC curves indicated that WC is the best predictor of CVD risk among study subjects. Findings from our study underscore the feasibility and face validity of using simple measures of central and overall adiposity in identifying CVD risk in resource-poor settings.