Skip to main content

03-01-2013 | Internal medicine | Article

BMI mortality link questioned


Full free text

medwireNews: Patients classified as overweight according to body mass index (BMI) have a lower risk for mortality than normal-weight patients, show the results of a systematic review.

The findings call into question the use of BMI as a useful predictor for mortality risk.

The review, which appears in JAMA, includes 97 studies, totaling 2.88 million individuals and over 270,000 deaths.

Patients who were classified as overweight (BMI 25-30kg/m2) had a 6% lower risk for death compared with normal-weight patients (BMI 18.5-25 kg/m2). By contrast, patients classified as obese (BMI ≥ 30 kg/m2) had an 18% increased risk for death compared with normal-weight patients.

However, among obese patients, those with a BMI below 35 kg/m2 (grade 1 obesity) showed no significant difference in mortality with normal-weight patients, while those with a BMI of 35 kg/m2 or greater (grades 2 and 3 obesity combined) had a 29% increased risk for death. This suggests that the observed mortality risk among obese patients is largely contributed to by the effect of the highest BMIs, say the authors.

"This study presents comprehensive estimates of the association of all-cause mortality in adults with current standard BMI categories used in the United States and internationally," state Katherine Flegal (Centers for Disease Control and Prevention, Hyattsville, Maryland, USA) and colleagues.

The findings provide further evidence for the so-called "obesity paradox," where overweight or mild obesity appear to have a protective effect. The authors suggest several reasons for their findings, including that overweight and moderately obese patients may be more likely to receive optimal treatment when they present with disease, increased body fat may have cardioprotective metabolic effects, and there may be benefits to having higher metabolic reserves.

In an accompanying editorial, Steven Heymsfield and William Cefalu of Pennington Biomedical Research Center, Baton Rouge, Louisiana, say the research casts doubts over the use of BMI: "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.

"Identification of at-risk individuals for overweight and grade 1 obesity is best captured by considering traditional risk factors, including blood pressure, blood lipid levels, and fasting blood glucose level, in addition to BMI, waist circumference, or both."

By Kirsty Oswald, medwireNews Reporter

Related topics