Health beyond the headlines

Databyte: Long Live the Overweight?

A closer look at the results of a recent meta-analysis in JAMA

Recently, we featured an analysis of the controversy shrouding a systematic review and meta-analysis published in JAMA on New Year’s Day. The review rigorously examined the link between BMI-based weight categories and risk of death from all causes. The most incendiary finding of the study was that overweight people seem to live about six percent longer than normal weight people.

In the article, the study results were strikingly visually arrayed as curvy spines, with each of the constituent studies as vertebrae. This adaptation of a figure (left) presents data comparing risk of death in those who are overweight to those who are normal weight, using BMI measures that were collected by the researchers themselves. The blue vertebrae represent studies that found being overweight was linked to longer life spans (hazard ratio less than 1). The two pink vertebrae represent studies that found no difference in length of life (hazard ratio equals 1), and the red vertebrae represent studies that found being overweight was linked to shorter life spans (hazard ratio greater than 1). The dashed lines represent those studies that were not statistically significant (the confidence intervals were in the longer life area and the shorter life area). While the scoliotic map of scientific evidence showed a wide range of individual study results, far more studies found that being overweight was associated with a longer life. The reverse was true for the obesity-normal BMI comparison (not pictured). For the overweight, but not obese, it appears this study has their backs.

One Response to “Databyte: Long Live the Overweight?”

  1. February 01, 2013 at 1:07 pm, Erin Kent said:

    Did the study adequately control for the effects of wasting, i.e. pronounced weight loss that often occurs with advanced chronic illness, like cancer, and often precipitates death? If not, it would give the appearance that being underweight is associated with mortality, and the researchers might have interepreted this as causal.


Elevate the conversation

The views and opinions expressed on this website are solely those of the authors and do not represent those of the Department of Epidemiology, the Mailman School of Public Health, or Columbia University.