Lesser degrees of excess weight don't increase mortality

Miriam E Tucker

January 02, 2013

Hyattsville, MD - Severe obesity was associated with an increased risk of death from all causes, but lesser amounts of excess weight either did not increase the risk or appeared protective, according to the results of a systematic review and meta-analysis of 97 studies [1].

The findings were in the January 2, 2013 Journal of the American Medical Association by Dr Katherine Flegal (National Center for Health Statistics, Hyattsville, MD) and colleagues.

In an accompanying editorial, however, Drs Steven Heymsfield and William Cefalu (Pennington Biomedical Research Center, New Orleans, LA) caution against relying on weight alone to stratify risk [2].

The analysis used body-mass index (BMI) categories drawn up by the National Heart , Lung , and Blood Institute (NHLBI) and combined data from 97 studies from around the world.

In all, the studies included more than 2.88 million subjects and over 270 000 deaths. All studies investigated the relationship of BMI and all-cause mortality and provided hazard ratios (HRs) for standard BMI categories (although some studies used slightly different ranges for the lowest BMI categories).

Compared with the normal-weight group, the HR for the overweight group was 0.94 (95% CI 0.91-0.96). For all grades of obesity together, the HR was 1.18 (95% CI 1.12-1.25). But when the obesity categories were broken down separately, grade 1 obesity (BMI 30 to <35 kg/m2 was not associated with increased all-cause mortality (HR 0.95, 95% CI 0.88-1.01).

Obesity grades 2 and 3 (BMI >35 kg/m2), on the other hand, were associated with greater mortality risk, with an HR of 1.29 (95% CI 1.18-1.41).

"Relative to normal weight, obesity (all grades) and grades 2 and 3 obesity were both associated with significantly higher all-cause mortality. Grade 1 obesity was not associated with higher mortality, suggesting that the excess mortality in obesity may predominantly be due to elevated mortality at higher BMI levels. Overweight was associated with significantly lower all-cause mortality," the authors write.

In a subsequent analysis that excluded 34 studies that were considered possibly overadjusted (ie, adjusted for factors such as hypertension that are considered to be part of the causal pathway between obesity and mortality) and 10 studies that were considered possibly underadjusted (ie, neglected to adjust for factors such as age, sex, or smoking), results for the remaining 53 adequately adjusted studies did not significantly alter the results.

Moreover, an analyses of contributors to heterogeneity, including study adjustment levels, whether BMI data were measured or self-reported, age group, and slight differences in BMI categorization, did not reveal a significant impact of heterogeneity on the overall conclusions of the meta-analysis.

Fits in with previous studies

Flegal and colleagues note that their findings are consistent with previous studies, which have also shown lower mortality among overweight and moderately obese individuals. Possible explanations have included earlier presentation of heavier patients for medical care and increased likelihood of receiving aggressive risk-factor treatment, cardioprotective metabolic effects of increased body fat, and beneficial effects of higher metabolic reserves.

In the editorial, Heymsfield and Cefalu comment that relying on weight alone is not enough, as individuals with the same BMI can differ widely from one another in terms of factors affecting health and mortality.

"Sole use of BMI as a health-risk phenotype aggregates people with substantial differences in nutritional status, disability, disease, and mortality risk together into similar BMI categories," they point out.

Recognizing that, the NHLBI also recommends using the additional marker of waist circumference to help quantify risk, the editorialists note.

Also, the NHLBI's classification of normal weight as a BMI 18.5 to 25 kg/m2 obscures the fact that people with a BMI between 18.5 and 22 kg/m2 have been found to have higher mortality than those with a higher BMI of 22-25 kg/m2. Lumping them together raises the mortality rate for the normal-weight group, which could explain why their observed mortality is similar to those with grade 1 obesity.

However, the editorialists point out that there appears to be a protective effect of the overweight or low-obesity BMI categories in the presence of chronic conditions (eg, heart disease or diabetes) or older age, the so-called "obesity paradox."

"Even in the absence of chronic disease, small excess amounts of adipose tissue may provide needed energy reserves during acute catabolic illnesses, have beneficial mechanical effects with some types of traumatic injuries, and convey other salutary effects that need to be investigated," they add.

Clinically, this means that "not all patients classified as being overweight or having grade 1 obesity, particularly those with chronic diseases, can be assumed to require weight-loss treatment. Establishing BMI is only the first step toward a more comprehensive risk evaluation," they conclude.

The study was funded by the C enters for D isease C ontrol and Prevention and National Cancer Institute . The authors reported no conflicts of interest. Heymsfield reported serving as a consultant to EISIA and Merck; serving on an advisory board for Tanita Medical; having travel expenses paid for by the Korean Society for the Study of Obesity; and receiving retirement payments from Merck. Cefalu reported serving on advisory boards or as a consultant to Halozyme, Lexicon, Intarcia, AstraZeneca, S anofi, and Johnson & Johnson; and receiving grants from Johnson & Johnson, AstraZeneca, Bristol-Myers Squibb, GlaxoSmithKline, Mannkind, Merck, Lilly, Amylin, and Intarcia.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.