The book critiques observational studies
versus randomized experiments (p 21):
"Observational studies are less reliable than
randomized experiments." The book then
goes on to use the example of observing two groups,
smokers and non-smokers, and then looking at the
rate of heart attacks in those two groups.
The book is correct in saying that this
methodology has a confounding
effect because there is no reason to believe that
the rate of smoking among men is the same as among women
or, for that matter, that the rate of heart
attacks is the same for the two groups.
The book is guilty, however, in claiming
that (p. 21): "It is known that men have
a higher risk of heart attack than women."
This "known" fact is not a fact, it is a societal prejudice.
Heart attacks in women are vastly under-reported and
far too often misdiagnosed as everything from
anxiety to fatigue to hysteria.
This prejudice is reinforced by the Pygmalian effect,
that is, doctors expect to
see heart attacks in men and therefore are
quick to interpret symptoms as a heart attack,
whereas doctors do not expect to see
heart attacks in women and therefore are less likely
to interpret symptoms as
a heart attack. This disparity is magnified because
the symptoms may be significantly different
between genders.