The effect of cardio-metabolic profile on the relationship of body mass index (BMI) with mortality is unclear. The aim of this study was to explore association between BMI and mortality at all ages, taking account of cardio-metabolic disorders.We followed 377,929 individuals (≥ 20 years), who registered for health checkups in 1996-2007, until 2008 and found 9490 deaths. From multivariable Cox proportional hazards models we estimated mortality hazard ratios (HR) for those in high blood pressure, hyperglycemia, high waist circumference, dyslipidemia, and different BMIs categories (the underweight [< 18.5 kg/m2], low normal weight [18.5-21.9 kg/m2], normal weight [22-23.9 kg/m2, the referent], overweight [24-26.9 kg/m2], obese1 [27-29.9 kg/m2], and obese2 [≥ 30 kg/m2]). Population attributable risk (PAR) provided estimates of the population mortality burden attributable to high blood pressure, hyperglycemia, high waist circumference, dyslipidemia, and deviant BMIs.Higher blood pressure, hyperglycemia, high waist circumference, and dyslipidemia were significantly predictive of higher mortality for nearly all ages. Compared with the referent BMI, underweight (HR = 1.69, 95% confidence interval = 1.51-1.90) and low normal weight (HR = 1.19, 1.11-1.28) were significant mortality risks, while overweight (HR = 0.82, 0.76-0.89) and obese1 (HR = 0.88, 0.79-0.97) were protective against premature death. The mortality impact of obesity was largely attributable to cardio-metabolic profile and attenuated by age. The population mortality burden with high blood pressure (PAR = 7.29%), hyperglycemia (PAR = 5.15%), high waist circumference (PAR = 4.24%), and dyslipidemia (PAR = 5.66%) was similar to that in the underweight (PAR = 5.50%) or low normal weight (PAR = 6.04%) groups. Findings for non-smokers and by gender were similar.The effect of BMI on mortality varies with age and is affected by cardio-metabolic status. Compared to any deviant BMI, abnormal cardio-metabolic status has a similar or even greater health impact at both the individual and population levels.