| In early January of 2005,
headlines reading “Vitamin E May Kill You”
or similar phrases were broadcast across the world,
shocking and scaring many people. What is the
evidence behind these headlines?
The information in question comes from research
reported at the American Heart Association’s
Scientific Sessions, and released on the web site
of the journal Annals
of Internal Medicine. The researchers concluded
that daily Vitamin E doses of 400 international
units (IU) or more can increase the risk of death
and should be avoided.
This conclusion was based on a meta-analysis
of 19 Vitamin E studies that looked at 136,000
patients with heart disease, cancer or Alzheimer’s
disease. “It was clear that as the Vitamin
E dose increased, so does the all-cause mortality,”
stated Johns Hopkins University researcher and
associate professor of medicine Edgar R. Miller,
MD, PhD.
But was this broad statement justified? Only
one of the 19 studies showed any statistical significance
with regards to Vitamin E and mortality –
the other 18 showed no increase in mortality.
It appears that the data was pooled to arrive
at a desired conclusion that is based on a statistical
artifact. The conclusion was not based on a true
scientific analysis of the 19 studies. Rather,
it was based on an artificial statistic that was
made to look comprehensive and convincing by referring
to the 19 studies.
The one study that did show supposed statistical
significance in fact showed only a “relative”
10% increase in mortality. A “relative increase”
is a statistical term for a manipulation of data
that makes small differences appear much larger.
For example, if 10 of 1000 people taking no vitamin
E died, and 11 of 1000 taking vitamin E died,
that would be a 10 per cent relative increase
in mortality in those taking Vitamin E. But the
actual increase was only 1 person in 1000 –
0.1 percent. This kind of trick is often performed
to make drugs look good and vitamins, herbs and
food supplements look bad.
Furthermore, Miller and the other authors of
the meta-analysis did not specify what forms of
vitamin E were used in the 19 studies. Many studies
use the synthetic form of alpha-tocopherol, which
has a distinctly different chemistry and less
beneficial effect than the natural form. And such
studies typically do not use Vitamin E containing
the important beta-, delta- and gamma-tocopherols.
These are reasonable explanations why the studies
did not show a decrease in mortality in the subjects
taking Vitamin E.
|
