Raised BMI is an established risk factor for death from vascular causes, cancer, and several other diseases, but uncertainties persist about whether reported associations with specific diseases have been affected by weight loss resulting from preexisting disease or by inadequate control of confounding variables. This observational study, a Prospective Studies Collaboration report, examined the association of BMI with overall and cause-specific mortality in a population of 894,576 individuals from 57 prospective studies, conducted mainly in Western Europe and North America. Data on individual patients were obtained from medical records. The study population was 61% (n = 541,452) male. The mean recruitment was at age 46 and mean BMI was 24.8 kg/m2. Enrolment of patients occurred during the 1970s or 1980s. Associations were adjusted by Cox regression for study, sex, age at risk, and smoking. To limit effects of preexisting disease on baseline BMI, the first 5 years of follow-up were excluded. During a mean of 8 ± 6 years of further follow-up, 66,552 deaths of known cause were identified. The mean age of death was 67 years. Specific causes of death were vascular, diabetic, renal, hepatic, liver, neoplastic, and respiratory. Overall mortality in both sexes was lowest at about 22.5 to 25 kg/m2. Above this minimum range, positive associations were found for several specific causes and there were no inverse association. Throughout the range 15 to 35 kg/m2, the absolute excess risks for higher BMI and smoking were roughly additive; there were minimal data for the range 35 to 50 kg/m2 for this association. Above the minimum range 22.5 to 25 kg/m2, each 5 kg/m2 rise in BMI was associated on average with 30% higher all-cause mortality; a 40% higher vascular mortality; a 60% to 120% increase in diabetic, renal, and hepatic mortality; 10% increase in neoplastic mortality; and a 20% increase in respiratory and all other mortality. No specific cause of death was inversely associated with BMI in the lower range (15–25 kg/m2). There was an overall inverse association with BMI predominantly due to strong inverse associations for smoking-related respiratory disease (including lung and chronic obstructive pulmonary disease). These inverse associations were much steeper in smokers than in nonsmokers; interestingly, cigarettes smoked per day had little effect on BMI. These findings show that BMI is a strong predictor of overall mortality both above and below the apparent optimum of about 22.5 to 25 kg/m2. Above this range, the absolute excess mortality is mainly due to vascular disease whereas below this range, the primary cause is smoking-related diseases. Median survival in both sexes is reduced by 2 to 4 years at BMI 30 to 35 kg/m2 and by about 8 to 10 years at BMI 40 to 45 kg/m2.
No takes yet. Share an insight, caveat, or question.
Prospective Studies Collaboration (2009) studied this question.