May 26, 2005 -- A CDC study underestimated U.S. obesity-related deaths and the death risk from being overweight, a Harvard panel says.
Just last month, CDC researchers Katherine M. Flegal, PhD, and colleagues suggested that the number of U.S. deaths directly linked to obesity is 112,000 per year. That's a lot lower than previous estimates of up to 400,000 annual obesity-related deaths.
Even more provocatively, the study showed no increased risk of death -- and even a slight protective effect -- for people who were merely overweight. And it showed that, perhaps because of medical advances, obesity is less likely to kill than in previous years.
Fiddlesticks, Harvard researchers said today at a highly publicized panel discussion held in Boston and webcast to the public. JoAnn Manson, MD, DrPH, professor of epidemiology at the Harvard School of Public Health, and colleagues went back over the same data used by the CDC researchers.
"It is biologically implausible that being overweight reduces mortality," says JoAnn Manson, MD, DrPH, professor of epidemiology at the Harvard School of Public Health. "We saw a very strong positive association with mortality -- a relative risk of [twofold] -- among those who were obese, and a clear increase in mortality among those who were overweight."
Also weighing in on the issue was Michael Thun, MD, vice president for epidemiology and surveillance research at the American Cancer Society in Atlanta.
"Overweight people do have higher all-cause death rates," Thun says. "This increase in mortality includes major cancers as well as cardiovascular disease. In my opinion, the CDC study seriously underestimated mortality from obesity and overweight."
The Method Is the Message
What's going on? Harvard and the American Cancer Society took a radically different approach to the data. Because smoking has a dramatic effect on health, they looked only at people who never smoked. And because undetected illness can make a person thin, they did not count deaths in the short term.
"Smoking masks the impact of obesity," says Frank Hu, MD, PhD, MPH, associate professor of nutrition and epidemiology at Harvard School of Public Health. "In our study there was no evidence of a reduced impact of obesity on mortality over time."
Researchers trying to determine the impact of obesity on death, Hu says, must exclude smokers and people who may be ill when the study begins. And, he says, it is very important for these studies to include very large numbers of people and to follow them for at least 20 years.
What does the CDC say? Flegal and colleagues did not attend the meeting. In their place was Donna Stroup, PhD, MSc, director of the coordinating center for health promotion at the CDC.
"I would not disagree with any of my colleagues on the program," Stroup says. "We are all trying to get to the same side of this issue -- that obesity is a very important problem. There are interventions that work, and we need to get about doing that. I do not disagree with their analysis. My point is that the differences [in estimates of obesity-related deaths] are less important than getting on with interventions that we know work."
These scientific arguments don't mean much to patients, says Scott Grundy, MD, PhD, director of the center for human nutrition at the University of Texas Southwestern Medical Center at Dallas. What patients need to know from their doctors is how their weight affects their health.
"Just because it is difficult to define the role of obesity and overweight as a cause of death, it doesn't mean there is no consequence for individuals to be overweight," Grundy says. "Morbidity is extremely important, but we are more concerned about quality of life than about what age we die. A lot of people are old and sick even though they are still alive."