Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis

Abstract

Importance: Estimates of the relative mortality risks associated with normal weight, overweight, and obesity may help to inform decision making in the clinical setting.

Objective: To perform a systematic review of reported hazard ratios (HRs) of all-cause mortality for overweight and obesity relative to normal weight in the general population.

Data sources: PubMed and EMBASE electronic databases were searched through September 30, 2012, without language restrictions.

Study selection: Articles that reported HRs for all-cause mortality using standard body mass index (BMI) categories from prospective studies of general populations of adults were selected by consensus among multiple reviewers. Studies were excluded that used nonstandard categories or that were limited to adolescents or to those with specific medical conditions or to those undergoing specific procedures. PubMed searches yielded 7034 articles, of which 141 (2.0%) were eligible. An EMBASE search yielded 2 additional articles. After eliminating overlap, 97 studies were retained for analysis, providing a combined sample size of more than 2.88 million individuals and more than 270,000 deaths.

Data extraction: Data were extracted by 1 reviewer and then reviewed by 3 independent reviewers. We selected the most complex model available for the full sample and used a variety of sensitivity analyses to address issues of possible overadjustment (adjusted for factors in causal pathway) or underadjustment (not adjusted for at least age, sex, and smoking).

Results: Random-effects summary all-cause mortality HRs for overweight (BMI of 25-<30), obesity (BMI of ≥30), grade 1 obesity (BMI of 30-<35), and grades 2 and 3 obesity (BMI of ≥35) were calculated relative to normal weight (BMI of 18.5-<25). The summary HRs were 0.94 (95% CI, 0.91-0.96) for overweight, 1.18 (95% CI, 1.12-1.25) for obesity (all grades combined), 0.95 (95% CI, 0.88-1.01) for grade 1 obesity, and 1.29 (95% CI, 1.18-1.41) for grades 2 and 3 obesity. These findings persisted when limited to studies with measured weight and height that were considered to be adequately adjusted. The HRs tended to be higher when weight and height were self-reported rather than measured.

Conclusions and relevance: Relative to normal weight, both obesity (all grades) and grades 2 and 3 obesity were associated with significantly higher all-cause mortality. Grade 1 obesity overall was not associated with higher mortality, and overweight was associated with significantly lower all-cause mortality. The use of predefined standard BMI groupings can facilitate between-study comparisons.

Figure 1

Hazard Ratios for All-Cause Mortality Relative to Normal Weight in Studies That Used Measured Data for Participants With a Body Mass Index of 25 to Less Than 30 Data markers indicate hazard ratios and error bars indicate 95% confidence intervals. SAHS indicates San Antonio Heart Study; and SALSA, Sacramento Area Latino Study on Aging.

Figure 2

Hazard Ratios for All-Cause Mortality Relative to Normal Weight in Studies That Used Self-reported Data for Participants With a Body Mass Index of 25 to Less Than 30 Data markers indicate hazard ratios and error bars indicate 95% confidence intervals.

Figure 3

Hazard Ratios for All-Cause Mortality Relative to Normal Weight in Studies That Used Measured Data for Participants With a Body Mass Index of 30 or Greater Data markers indicate hazard ratios and error bars indicate 95% confidence intervals.

Figure 4

Hazard Ratios for All-Cause Mortality Relative to Normal Weight in Studies That Used Self-reported Data for Participants With a Body Mass Index of 30 or Greater Data markers indicate hazard ratios and error bars indicate 95% confidence intervals.

Figure 5

Hazard Ratios for All-Cause Mortality Relative to Normal Weight in Studies That Used Measured Data for Participants With a Body Mass Index of 30 to Less Than 35 Data markers indicate hazard ratios and error bars indicate 95% confidence intervals. SAHS indicates San Antonio Heart Study; and SALSA, Sacramento Area Latino Study on Aging.

Figure 6

Hazard Ratios for All-Cause Mortality Relative to Normal Weight in Studies That Used Self-reported Data for Participants With a Body Mass Index of 30 to Less Than 35 Data markers indicate hazard ratios and error bars indicate 95% confidence intervals.

Figure 7

Hazard Ratios for All-Cause Mortality Relative to Normal Weight in Studies That Used Measured Data for Participants With a Body Mass Index of 35 or Greater Data markers indicate hazard ratios and error bars indicate 95% confidence intervals. SAHS indicates San Antonio Heart Study; and SALSA, Sacramento Area Latino Study on Aging.

Figure 8

Hazard Ratios for All-Cause Mortality Relative to Normal Weight in Studies That Used Self-reported Data for Participants With a Body Mass Index of 35 or Greater Data markers indicate hazard ratios and error bars indicate 95% confidence intervals.