Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies

Abstract

Background: The main associations of body-mass index (BMI) with overall and cause-specific mortality can best be assessed by long-term prospective follow-up of large numbers of people. The Prospective Studies Collaboration aimed to investigate these associations by sharing data from many studies.

Methods: Collaborative analyses were undertaken of baseline BMI versus mortality in 57 prospective studies with 894 576 participants, mostly in western Europe and North America (61% [n=541 452] male, mean recruitment age 46 [SD 11] years, median recruitment year 1979 [IQR 1975-85], mean BMI 25 [SD 4] kg/m(2)). The analyses were adjusted for age, sex, smoking status, and study. To limit reverse causality, the first 5 years of follow-up were excluded, leaving 66 552 deaths of known cause during a mean of 8 (SD 6) further years of follow-up (mean age at death 67 [SD 10] years): 30 416 vascular; 2070 diabetic, renal or hepatic; 22 592 neoplastic; 3770 respiratory; 7704 other.

Findings: In both sexes, mortality was lowest at about 22.5-25 kg/m(2). Above this range, positive associations were recorded for several specific causes and inverse associations for none, the absolute excess risks for higher BMI and smoking were roughly additive, and each 5 kg/m(2) higher BMI was on average associated with about 30% higher overall mortality (hazard ratio per 5 kg/m(2) [HR] 1.29 [95% CI 1.27-1.32]): 40% for vascular mortality (HR 1.41 [1.37-1.45]); 60-120% for diabetic, renal, and hepatic mortality (HRs 2.16 [1.89-2.46], 1.59 [1.27-1.99], and 1.82 [1.59-2.09], respectively); 10% for neoplastic mortality (HR 1.10 [1.06-1.15]); and 20% for respiratory and for all other mortality (HRs 1.20 [1.07-1.34] and 1.20 [1.16-1.25], respectively). Below the range 22.5-25 kg/m(2), BMI was associated inversely with overall mortality, mainly because of strong inverse associations with respiratory disease and lung cancer. These inverse associations were much stronger for smokers than for non-smokers, despite cigarette consumption per smoker varying little with BMI.

Interpretation: Although other anthropometric measures (eg, waist circumference, waist-to-hip ratio) could well add extra information to BMI, and BMI to them, BMI is in itself a strong predictor of overall mortality both above and below the apparent optimum of about 22.5-25 kg/m(2). The progressive excess mortality above this range is due mainly to vascular disease and is probably largely causal. At 30-35 kg/m(2), median survival is reduced by 2-4 years; at 40-45 kg/m(2), it is reduced by 8-10 years (which is comparable with the effects of smoking). The definite excess mortality below 22.5 kg/m(2) is due mainly to smoking-related diseases, and is not fully explained.

Figures

Figure 1

Vascular risk factors versus BMI at baseline in the range 15–50 kg/m2 Adjusted for baseline age, baseline smoking status (apart from the smoking findings), and study. Numerical values are shown for 20–22·5 kg/m2, for 30–32·5 kg/m2, and for the extreme BMI groups. Boundaries of BMI groups are indicated by tick marks. 95% CIs are not shown, but most are narrower than the heights of the plotted symbols. (A) Blood pressure (in 533 242 males and 348 790 females). (B) Blood cholesterol fractions (in 62 364 males and 52 575 females with total and HDL cholesterol both measured); dashed line indicates the ratio of mean non-HDL cholesterol to mean HDL cholesterol (mean of the individual ratios would be about 8–12% greater). (C) Prevalences in males for alcohol drinking (168 283), cigarette smoking (334 496), and diabetes (378 854). (D) Prevalences in females for alcohol drinking (129 301), cigarette smoking (226 307), and diabetes (319 401).

Figure 2

All-cause mortality versus BMI for each sex in the range 15–50 kg/m2 (excluding the first 5 years of follow-up) Relative risks at ages 35–89 years, adjusted for age at risk, smoking, and study, were multiplied by a common factor (ie, floated) to make the weighted average match the PSC mortality rate at ages 35–79 years. Floated mortality rates shown above each square and numbers of deaths below. Area of square is inversely proportional to the variance of the log risk. Boundaries of BMI groups are indicated by tick marks. 95% CIs for floated rates reflect uncertainty in the log risk for each single rate. Dotted vertical line indicates 25 kg/m2 (boundary between upper and lower BMI ranges in this report).

Figure 3

Ischaemic heart disease and stroke mortality versus BMI in the range 15–50 kg/m2 (excluding the first 5 years of follow-up) Relative risks at ages 35–89 years, adjusted for age at risk, sex, smoking, and study, were multiplied by a common factor (ie, floated) to make the weighted average match the PSC mortality rate at ages 35–79 years. Floated mortality rates shown above each square and numbers of deaths below. Area of square is inversely proportional to the variance of the log risk. Boundaries of BMI groups are indicated by tick marks. 95% CIs for floated rates reflect uncertainty in the log risk for each single rate.

Figure 4

Ischaemic heart disease (A) and stroke mortality (B) versus BMI in the upper BMI range (25–50 kg/m2) only (excluding the first 5 years of follow-up, except as indicated) Hazard ratios are per 5 kg/m2—eg, 30 kg/m2 versus 25 kg/m2—and are, when appropriate, adjusted for age at risk, sex, smoking, and study. Mean ages at death are given, but the dependence of the HR on mean age at death is not corrected for in analyses of factors other than age. The area of each square is inversely proportional to the variance of the log hazard ratio. White squares include the first 5 years of follow-up; black squares and white diamonds do not. Subarachnoid=subarachnoid haemorrhage (not included in haemorrhagic stroke).

Figure 5

Mortality rates at ages 35–79 years for main disease categories versus BMI in the range 15–50 kg/m2 (excluding the first 5 years of follow-up) Relative risks at ages 35–79 years, adjusted for age at risk, smoking, and study, were multiplied by a common factor (ie, floated) to make the weighted average match the age-standardised European Union (15 countries) mortality rate at ages 35–79 years in 2000. Neoplastic mortality is split into the types most strongly associated with smoking (cancers of the lung and upper aerodigestive tract) and all other specified types. By contrast with figures 2–4, risk is indicated on an additive rather than multiplicative scale, with floated mortality rates shown above or below each symbol. The estimates for 35–50 kg/m2 are based on limited data, so lines connecting to those estimates are dashed. Boundaries of BMI groups are indicated by tick marks. 95% CIs for floated rates reflect uncertainty in the log risk for each single rate.

Figure 6

All-cause mortality at ages 35–79 years versus BMI in the range 15–50 kg/m2, by smoking status (excluding the first 5 years of follow-up) Relative risks at ages 35–79 years, adjusted for age at risk, sex, and study, were multiplied by a common factor (ie, floated) so that the mean for all participants (including ex-smokers and anyone with missing smoking data) matches the European rate at ages 35–79 years in 2000. Results for ex-smokers and those with missing smoking data not shown (but are, taken together, only slightly above those for never smokers). Note that many smokers were at only limited risk, since they had not smoked many cigarettes during early adult life, or had stopped shortly after the baseline survey. Risk is indicated on an additive rather than multiplicative scale. The estimates for 35–50 kg/m2 are based on limited data, so lines connecting to those estimates are dashed. Floated mortality rates shown above each square and numbers of deaths below. Area of square is inversely proportional to the variance of the log risk. Boundaries of BMI groups are indicated by tick marks. 95% CIs for floated rates reflect uncertainty in the log risk for each single rate.

Figure 7

BMI versus lifespan in western Europe, year 2000 Estimated effects of the BMI that would be reached by about 60 years of age on survival from age 35 years, identifying European Union (EU) mortality rates in 2000 with those for BMI 25–30 kg/m2 and combining the disease-specific EU mortality rates with disease-specific relative risks (for details, see webappendix pp 18–20). The absolute differences in median survival (but probably not in survival to age 70 years) should be robust to changes in mortality rates, and therefore generalisable decades hence. (A) 3 main BMI categories. (B) 3 main and 2 higher BMI categories. (The 2 higher BMI categories account for just 2% of PSC participants, and so are indicated by dashed lines.)