Obesity Study Overstated Effects

A widely reported government study that said obesity is about to overtake smoking as the No. 1 cause of death in the United States contained statistical errors and may have overstated the problem, health officials acknowledged Tuesday.

The government is working on a rare correction to the study.

The Centers for Disease Control and Prevention said in March in a study co-authored by its director, Dr. Julie Gerberding, that poor diet and physical inactivity were responsible for 400,000 deaths in 2000, a 33 percent jump from 1990.

However, the CDC admitted Tuesday that it made an error in calculating how many people died from obesity in the last decade.

Although CDC officials declined to specify the corrected number of deaths, saying they are still determining the correct number, The Wall Street Journal reported that the agency may have overstated the figure by 80,000, representing an increase of less than 10 percent from 1990 to 2000. The errors were first reported by the Journal on Tuesday.

The CDC has corrected articles in major journals before, but "unfortunately it was a paper that received a lot of attention and had our director's name on it. To my knowledge that confluence of events really hasn't occurred at the same time," said Dr. Dixie Snider, the CDC's chief of science, who was appointed to lead the agency's investigation of the error.

The mistakes consisted of simple mathematical errors, such as including total deaths from the wrong year, the newspaper reported.

"Eighty-thousand is an estimate from one of our scientists — that's not a number we're going with," said Snider, who declined to speculate on the correct number of obesity-related deaths. "We regret any confusion that we may have caused about obesity as a public health issue and we regret the inadvertent computational error."

The CDC plans to submit a correction to the Journal of the American Medical Association, which published the study in March. The correction will explain how the error was made, said CDC spokesman Tom Skinner.

The agency also has asked the Institute of Medicine, a federal scientific advisory organization, to hold a two-day workshop next month in Washington to reach a consensus on the proper way to calculate the health effects of obesity. That is because the study also caused disagreement in scientific circles over how deaths can be labeled obesity-related.

In addition, the agency is reviewing how to prevent "miscommunication" among scientists when subjecting studies to expert review before they are published.

The errors apparently were discovered soon after the CDC study was published, as scientists inside and outside the agency began to dispute its findings, publishing letters in JAMA and the journal Science.

That prompted the CDC to begin an internal review. Rep. Henry Waxman, D-Calif., also asked the General Accounting Office, the auditing arm of Congress, to investigate.

The agency originally announced that more Americans could soon be dying of obesity instead of smoking if current trends persisted. It put the number of obesity deaths at 400,000, compared with 435,000 from tobacco.

But even when the errors are corrected, Skinner said, "it's not going to change the fact that obesity is the second leading cause of preventable death."