Relationship between body mass index and mortality among Europeans

Background/Objectives: To investigate the relationship between body mass index (BMI) and mortality from various causes. Subjects/Methods: Data of 72 947 European men and 62 798 women aged 24-99 years at baseline were collaboratively analyzed. Both absolute and relative mortality risks were estimated within each BMI categories. The hazard ratio was estimated using Cox regression analysis adjusting for age, cohort and smoking status. Results: Over a median follow-up of 16.8 years, 29 071 participants died, 13 502 from cardiovascular disease (CVD) and 8748 from cancers of all types. All-cause and cancer mortality showed a U-shaped relationship: decreased first, leveled off, and then increased with increasing BMI with the lowest mortality risk approximately between 23.0 and 28.0 kg/m(2) of BMI in men and 21.0 and 28.0 kg/m(2) in women. The U-shaped relationship held for all-cause mortality but disappeared for cancer mortality among non-smokers. The CVD mortality was constant until a BMI of approximately 28.0 kg/m(2) and then increased gradually in both men and women, which was independent of age, cohort and smoking status. Conclusions: A U-shaped relationship of BMI with all-cause mortality but a graded relationship with CVD mortality at BMI >28.0 kg/m(2) was detected. The relationship between cancer mortality and BMI largely depended on smoking status, and need to be further investigated with site-specific cancers. European Journal of Clinical Nutrition (2012) 66, 156-165; doi:10.1038/ejcn.2011.145; published online 10 August 2011