|
About the only
thing they are eating these days at the US Centers for Disease Control is crow.
That’s because the Center has had to admit that its much hyped figure of 400,000
deaths per year from obesity is dead wrong. According to a story by Betsy McKay
in the Wall Street Journal the Center has determined that “mathematical errors
may have inflated the study’s death toll by about 80,000 fatalities, or 20% of
the total deaths.”
Critics of the
study argue that even that number is conservative, claiming that the true figure
of obesity caused deaths is actually closer to 200,000. According to the Journal
the study, which showed deaths from obesity overtaking those from tobacco, was
criticized by insiders at the CDC for its assumptions, methodology and data
sources even before publication. But in the rush to justify the growing war on
fat these concerns were apparently pushed aside.
While the
Center was quick to claim that the mistaken estimate of obesity related deaths
was merely a “statistical error” and not an attempt to falsify the data, the
more one looks at the way the report was compiled the more it raises substantial
scientific and ethical worries about the entire war on fat. Much of that worry
comes from the fact that even after being forced to admit that its obesity death
toll was widely inaccurate, the CDC was still unwilling to admit that obesity
was at best a phantom killer. As the co-author of the study and Director of CDC
told the Journal “The bottom line is that obesity is a leading cause of death.”
But is it?
The difficulty
with the idea that obesity is a cause of death is that both the logic and the
statistics of death and disease simply don’t support it. For instance, obesity
itself is not a disease which means that if it has a role in mortality it can
only do so through some real disease like heart disease, cancer or diabetes. But
the problem here is that both heart disease and cancer are diseases with
multiple risk factors, indeed hundreds of them. So it is impossible to know
which risk factor led to each death and thus to know what role obesity might
have played in any single death.
But there are
other difficulties. If obesity is causing an epidemic of deaths than you should
expect to find a sharp increase from deaths from things like heart disease,
cancer and diabetes. But data shows that deaths from cardiovascular disease,
coronary heart disease and many cancers have actually been declining not
increasing over the last decade, even while deaths from obesity have allegedly
been soaring. Even the question of diabetes mortality is complicated since it is
not at all clear that blood sugar levels are in fact rising and the definition
of diabetes was changed from a fasting blood sugar level of 140 to 126, which
overnight created millions of new diabetics. Even the CDC’s own diabetes
figures- if we can trust them - show that the incidence of Type 2 ( which
accounts for 90% of diabetes) increased by only .04% over the last decade in
which obesity rates have allegedly increased from 45-60%.
Then there is
the little matter of how the elderly figure into the death by obesity equation.
We know, for instance, that in the US there are about two million deaths each
year, of which roughly 70% are of people over age 65. Since obesity has little
effect on the elderly, this means that there are only about 600,000 available
deaths from the rest of the population. If some 400,000 deaths are caused by
obesity this means that about 65-70% of all people under 65 who die in the US
every year would have to die from obesity- an absurd figure supported by
absolutely no data.
And speaking of
data, just where did the CDC turn for its numbers? According to the Center for
Consumer Freedom it relied on the methodology of a David Allison from the
University
of South Alabama. Not only did Allison assume, wrongly, that all obese people
die from their obesity, but he used estimates on the likelihood of death from
diseases linked to obesity that were on average 20 years and in some instances
50 years out of date and which took no account of modern treatments that reduced
mortality rates.
The biggest
problem with the CDC’s numbers of course is that they contradict a vast
literature on obesity, overweight and mortality that shows that being overweight
or obese does NOT translate into an increased risk for mortality. For example,
in an analysis of over 500,00 men and women done by the US National Center for
Health Statistics, the lowest mortality was for those individuals with BMI’s
between 23-29, in fact most of the “overweight” range. Nor was this simply a
freak outcome. In the famous 1999 study in JAMA that claimed to find 300,000 US
obesity-related deaths, the data showed that those with BMI’s of 20-ideal- run
the same mortality risk as those with BMI’s of 30- the obese.
All of this
suggests that this is not merely some storm in the statistical teacup but a
major credibility issue for the entire war on fat. The only basis for the
massive interventions into the nation’s stomachs being proposed by the
government and the public health community is that we are all getting
excessively fat and our fat is killing us. But between the doubts about the CDC
numbers, the evidence that most of us gain about a pound a year (most of it
during the holidays according to the New England Journal of Medicine) and the
huge literature that fails to find a link between overweight and premature
mortality, the rationale for the war on fact looks like it is dissolving under a
weight of junk science.
John Luik, a health policy analyst, is co-author with Mike Waterson, of
Advertising and Markets |
