Objective To compare the relative risk of waist circumference (WC) and/or BMI on cardiovascular risk factors. Methods A cross-sectional data of 41 087 adults (19 567 male and 21 520 female) from the 2002 China National Nutrition and Health Survey were examined. According to the obesity definition of WGOC (BMI, 24 kg/m2 and 28 kg/m2; WC, male 85 cm and 95 cm for male, 80 cm and 90 cm for female), the study population were divided into 9 groups. The prevalence and odds ratio (ORs) of cardiovascular disease (CVD) risk factors (hypertension, high fasting plasma glucose and dyslipidemia) were compared among these 9 groups. Stepwise linear regression analyses were used to compare the likelihood of BMI and/or WC on CVD risk factors. Results Both the indexes levels and the odds ratios of CVD risk factors were significantly increased (decreased for HDL-C levels) along with the increase of WC and/or BMI, even when the effect of age, sex, income, education, sedentary activity and dietary factors were adjusted. The variances (R2) in CVD risk factors explained by WC only and BMI only were quite similar, but a little bit larger when WC and BMI were combined. The standard β was higher of BMI when predicting systolic BP and was higher of WC when predicting TG, TC and HDL. Conclusions BMI and WC had independent effects on CVD risk factors and combination of BMI and WC would be more predictive. Findings from the present study provided substantive evidence for the WGOC recommendation of a combined use of BMI and WC classifications. Objective To compare the relative risk of waist circumference (WC) and / or BMI on cardiovascular risk factors. Methods A cross-sectional data of 41 087 adults (19 567 male and 21 520 female) from the 2002 China National Nutrition and Health Survey were examined. According to the obesity definition of WGOC (BMI, 24 kg / m2 and 28 kg / m2; WC, male 85 cm and 95 cm for male, 80 cm and 90 cm for female), the study population were divided into 9 groups The prevalence and odds ratio (ORs) of cardiovascular disease (CVD) risk factors (hypertension, high fasting plasma glucose and dyslipidemia) were among these 9 groups. Stepwise linear regression analyzes were used to compare the likelihood of BMI and / or WC on both risk and risk factors. Results Both the indexes levels and the odds ratios of CVD risk factors were significantly increased (decreased for HDL-C levels) along with the increase of WC and / or BMI, even when the effect of age, sex, income , education, sedentary activity and dietary factors were adjusted. The variances (R2) in CVD risk factors explained by WC only and BMI only were quite similar, but a little bit larger when WC and BMI were combined. The standard β was higher of BMI when predicting systolic BP and was higher of WC when predicting TG, TC and HDL. Conclusions BMI and WC had independent effects on CVD risk factors and combination of BMI and WC would be more predictive. Findings from the present study provided substantive evidence for the WGOC recommendation of a combined use of BMI and WC WC classifications.