Dire Warnings About Obesity Rely
on a Slippery Statistic

by Laura Johannes and Steve Stecklow
Copyright 1998 Dow Jones & Co., Inc.
The Wall Street Journal (February 9, 1998)

When the U.S. Food and Drug Administration approved the new diet drug Meridia in November, it called obesity a "widespread, chronic disease."

Roche Laboratories, which is seeking FDA approval for another diet drug, agrees. It states in a brochure: "By any standard, obesity is a disease."

But is it?

In fact, medical science is split. The confident assertions of the FDA and Roche trace back along a murky trail to some far less categorical scientific data.

More than semantics is at stake. In the case of obesity, the debate has heightened in the wake of major diet-pill recalls last year. Many new diet nostrums are in various stages of testing, and the FDA's bar for approving new drugs is lower for disease treatments than for other problems, such as baldness or skin wrinkles. The agency is less likely to approve a drug for a nondisease condition when it is shown to have serious side effects -- such as those that diet drugs produce. Meridia, for instance, raises blood pressure when taken in high doses; that could lead to strokes.

In part, the FDA appears to accept that obesity is a disease because of oft-repeated testimony, from drug companies and others, that fatness results in a measurable number of deaths. In its Meridia approval, the agency repeated that obesity "contributes to the deaths of an estimated 300,000 Americans each year."

That figure first surfaced three years ago when former Surgeon General C. Everett Koop launched Shape Up America!, a physical-fitness campaign coordinated by a nonprofit coalition of industry and health-related organizations. Standing on the South Lawn of the White House, Dr. Koop called obesity "a major public health threat" and said it had become the nation's No. 2 cause of death, behind smoking, "resulting in about 300,000 lives lost each year."

Dr. Koop says he derived the statistic from a study called "Actual Causes of Death in the United States" that appeared in November 1993 in the Journal of the American Medical Association. But the study -- a review of death certificates filed in 1990 -- never said obesity killed that many people. It said, rather, that "dietary factors and activity patterns that are too sedentary are together accountable for at least 300,000 deaths each year." It also said what many already knew -- that obesity is "associated with increased risk for cardiovascular disease, certain cancers and diabetes" but stopped short of calling obesity a disease by itself.

Some researchers believe obesity may be more of a symptom than a disease. In another JAMA study last year, the Cooper Institute for Aerobics Research in Dallas found that skinny people with poor physical fitness were more likely to die prematurely than overweight people who exercised regularly. "We tend to focus on body size and body weight. In fact, our habits have more to do with our mortality and what diseases we get than our body size does," says Cooper Medical Director Jody Wilkinson.

Some highly publicized findings have spotlighted the genetic antecedents of obesity in mice. But no such link has been established in humans, except in very rare cases. Even if some people are genetically predetermined to be fat, others "just like to eat -- and in that case it's no more of a disease than bank robbery is a disease," says Marcia Angell, author of a recent editorial in the New England Journal of Medicine that questioned the obesity-as-a-disease model.

The debate echoes those involving other medical conditions, caught in their own quagmires in which the roles of genetics and self-control are unclear. Alcoholism, for example, is widely regarded as a disease, which troubles some specialists. "How do you say this is a disease if people can voluntary reverse it?" asks Joseph Gerstein, past president of the Smart Recovery self-help network for alcoholics.

Barbara J. Moore, who became executive director of Shape Up America! in June 1995, says she warned Dr. Koop that he was misrepresenting the study's findings by making so direct a link between obesity and death. But Dr. Koop defends his interpretation and predicts that "it will eventually be shown" that obesity does kill as large a number as 300,000 -- just as preliminary findings on smoking led to more definitive research later.

Dr. Koop's declaration at the White House soon found its way into numerous news articles and other publications. It has often been quoted by Interneuron Pharmaceuticals Inc., the developer of Redux, a diet pill the FDA approved in 1996 but which was withdrawn at its request in 1997 amid concerns that it damages heart valves. Interneuron Chief Executive Glenn Cooper and Redux supporters cited the figure at least seven times at FDA advisory panel hearings on the drug.

"The 300,000 figure was the single most important reason the advisory committee approved Redux," says Lynn McAfee, of the Council on Size and Weight Discrimination, an advocacy group, who attended the hearings. "And it's a fake number."

James Bilstad, head of the FDA division that approved Redux, concedes he isn't clear about the origin of that statistic. But he says it wasn't the main evidence supporting the FDA's belief that obesity is a disease that can be helped by drugs. He adds that the FDA hasn't addressed the question of how much extra weight qualifies as a disease.

Interneuron says the statistic was supported by an independent analysis by Theodore VanItallie, a professor emeritus of medicine at Columbia University in New York and a consultant to Laboratoires Servier SA, the French company that licensed Redux to Interneuron. Dr. VanItallie, in a 27-page document labeled "confidential" and distributed only to the FDA and its advisory committee during Redux deliberations, calculated that 20% of Americans, or about 292,000 people a year, die from obesity.

But a copy of that document shows that Dr. VanItallie's figures may be exaggerated. In calculating the figure, he drew on previously published reports on obesity and death, including the extensive Nurses Health Study, which found excess weight at fault in about 20% of the untimely deaths in nonsmoking adults under 70.

But without correcting for age differences or the effects of smoking in these studies, he concluded that obesity is to blame for 20% of all U.S. deaths. The problem is that many deaths occur among heavy smokers, or in people in their 70s and 80s -- an older group than the Nurses Health cohort.

Dr. VanItallie concedes his analysis is "very rough" and says "the reason I have not published it is because I have been agonizing over how to deal with these problems."

The recent New England Journal of Medicine editorial by Dr. Angell and the journal's editor in chief, Jerome P. Kassirer, said the 300,000 number was "derived from weak or incomplete data." In the same issue, researchers from the University of North Carolina and the U.S. government found that the additional risk of death posed by excess weight lessened with age.

The 300,000 figure has also been promulgated by officials from the American Obesity Association, who testified at hearings on Redux and Meridia and who have lobbied Congress to urge greater funding for obesity research. In 1996, the group worked with Shape Up America! to produce a book of recommended treatments for obesity, including drugs, that was distributed to 30,000 doctors and health-care professionals. It was supported by a grant from American Home Products Corp., marketer of Redux and Pondimin.

The association, based in Washington, D.C., was formed in April 1995 by Judith S. Stern, who teaches at the University of California at Davis, and Richard L. Atkinson, a researcher at the University of Wisconsin in Madison. At the advisory panel hearing on Meridia, Dr. Atkinson, the first speaker, described the association as "a lay advocacy group representing the interests of the 70 to 80 million obese American women and children and adults afflicted with the disease of obesity."

The association charges individuals $25 for membership. Dr. Stern says she and Dr. Atkinson belong but she has "absolutely no idea" how many others are members. Asked for a member count, Morgan Downey, the group's executive director, produced a single, completed application and check and said it was from the only paid member he had seen.

Financial information about the group also is hard to come by. The association is required by law to file a nonprofit tax return; but it has never done so, according to its new accountant, Gary C. Pokrant. He adds that a return will be filed shortly.

Dr. Atkinson says the group receives most of its funding -- several hundred thousand dollars in all -- from the pharmaceuticals industry, including Interneuron, American Home Products, Roche Laboratories, Knoll Pharmaceuticals Ltd. and Servier -- all of which market or develop diet pills.

He says drug companies "love me" because of his strong support for diet pills. Although he has received consulting fees from about 20 drug concerns, he says the money doesn't affect his positions.

Material presented on this home page constitutes opinion of the author.
Copyright © 1998 Steven J. Milloy. All rights reserved. Site developed and hosted by WestLake Solutions, Inc.