Results of CDC Study on Weight vs. Mortality Questioned by Experts
Posted Mar 01 2013 12:00am
Everyone "knows" that overweight people have an increased mortality, right? When an article was published by a researcher from the CDC, the obesity researchers were aghast. It concluded that slightly overweight people have a lower all-cause mortality than normal weight and underweight people. These researchers convened a panel at Harvard to look at the meta-analysis methodology used in the controversial article. The result of all of this is described in an article on the web site of the Harvard Medical School (see: Weight and Mortality ), Below is an excerpt from it. Read the whole article for more details.
In January, when the Journal of the American Medical Association published a meta-analysis of 100 studies that probed the relationship between body mass index and mortality—studies that found slightly overweight people have lower all-cause mortality than normal weight and underweight people—media around the globe trumpeted the news....Many [experts] were perplexed by the findings gathered by Centers for Disease Control epidemiologist Katharine Flegal, which contradict a preponderance of research indicating that there is a direct correlation between the risk of mortality and being overweight once factors such as lower weight from cigarette smoking, chronic disease and wasting from frailty in the elderly are taken into account. To clear up the confusion, [a panel of experts was convened] to discuss the findings ....Flegal, a senior scientist at the CDC’s National Center for Health Statistics, was invited to the event, but did not attend....The panelists evaluated Flegal’s findings and pointed out a number of methodological errors in the study that they said resulted in the artificial appearance of a protective benefit in being overweight or mildly obese....The selection criteria that Flegal used for her meta-analysis ruled out high-quality studies of 6 million people....These studies, in aggregate, show that the highest survival rates are in normal weight people, not the overweight....The studies that Flegal did use included many samples of people who were chronically ill, current smokers and elderly....These factors are associated with weight loss and increased mortality. In other words, people are not dying because they are slim....They are slim because they are dying—of cancer or old age, for example. By doing a meta-analysis of studies that did not properly control for this bias, Flegal amplified the error in the original studies.
In statistics, a meta-analysis refers to methods focused on contrasting and combining results from different studies, in the hope of identifying patterns among study results, sources of disagreement among those results, or other interesting relationships that may come to light in the context of multiple studies. In its simplest form, this is normally by identification of a common measure of effect size, of which a weighted average might be the output of a meta-analysis. The weighting might be related to sample sizes within the individual studies. More generally there are other differences between the studies that need to be allowed for, but the general aim of a meta-analysis is to more powerfully estimate the true effect size as opposed to a less precise effect size derived in a single study under a given single set of assumptions and conditions.
One could glean a number of lessons from this article. The first lesson is that scientists from the prestigious CDC are not infallible. The second is, logically, that some people in a large patient population may be slim because they are chronically ill or dying. The third is that bias can creep into any article including those based on a meta-analysis of multiple other published studies. Lastly, be very careful when you publish results the fly in the face of ideas that are widely held. The contradicted experts may take a very hard look at the data and methodology used in such an article.