me assert my firm belief that the only thing we have to fear is
Franklin D. Roosevelt
saturated fat bad for your heart? Thatís the conventional wisdom,
but I shall describe what led to this belief and show that the
available evidence does not support it. Then I shall explain how
the nocebo effect might override this evidence-based conclusion.
to make a determination of the effect of a nutrient on one of
the little rodents kept prisoner in a laboratory. First you must
obtain a series of identical subjects by inbreeding or cloning.
Then you must compare two individuals (or, better, two groups
of individuals) who differ only in whether or not they have received
the nutrient in question. You must know where to look for differences
and how to measure them (blood tests, growth rates, etc.). And
you can be confident of your final result only with respect to
these particular subjects and their particular living conditions,
which, in the case of the little prisoners, usually include an
unstimulating environment and a lack of exercise.
is even more difficult for investigators who want to make such
a determination on people. People have relatively long life spans,
on the order of magnitude of the investigatorsí. Groups of identical
subjects are hard to come by and their living conditions are not
readily subject to control. And because different groups of free-living
people (occupants of different countries, followers of different
diets, members of different ethnic groups) differ in so many characteristics,
one must be very cautious in imputing an observed difference in
an outcome to an observed difference in behavior. Thus, it comes
as no surprise that different investigators have come to different
conclusions about the effect on people of eating saturated fat,
the nutrient which is the subject of this essay.
such questions of only academic interest? After all, some time
ago Dr. John Yudkin, a British physician, pointed out that our
appetites should lead us to suitable diets. Our remote ancestors
chose their diets on the basis of palatability (I use the term
to include not only taste and smell, but also other attributes
such as appearance). Their diets were good enough to enable them
to survive and procreate and they passed on to us the tastes which
led them to choose suitable diets. You canít have ancestors whose
tastes led them to diets which resulted in early and childless
death. And tastes can be shown to be hereditary. For instance,
thiourea, a substance found in plants of the cabbage family is
considered by some people (including me) to be extremely bitter
but is tasteless to others. The trait of experiencing it as bitter
is passed along in accordance with the Mendelian rules of inheritance.
thrive by choosing their dietary constituents according to their
palatability and availability, but it doesnít work as well for
us because of a number of ways in which we have separated the
palatability which signals the nutrient content of foods from
the nutrients whose presence they normally signify. An extreme
example is Kool-Aid which has some of the palatability of fruit,
but none of its important nutrients. Artificial flavors and colors
defeat the evolved system for choosing a proper diet. Other factors
include religious and social taboos, storing or shipping food
long distances during which nutrients are lost, preserving foods
by heating, canning or freezing them, and politeness Ė have you
ever ignored what your inherited sense of palatability was telling
you and eaten something that failed to appeal to your senses so
as not to hurt the feelings of the person who offered it to you?
for foods containing saturated fat, (e.g., fatty meats, butter,
coconut oil) suggests that these may be essential components of
our diet, yet the conventional wisdom now seems to be that eating
saturated fat is a risk factor for heart disease. How did this
belief come about?
a clear-cut villain here. Ancel Keys, a researcher from Minnesota
started the boule de suif rolling by demonstrating that
a plot of fat intake versus coronary mortality, using data from
six countries, showed a smooth correlation between the two. Others
soon pointed out that if data from all twenty-two countries for
which such data were then available were plotted, the result was
a formless random array, that the data from countries which Keys
had left out suggested that level of physical activity was the
most accurate predictor of cardiac risk and that sugar was the
dietary constituent most closely correlated with heart disease.
This didnít stop Keys from going on to present a massive collection
of data from seven carefully chosen countries which bolstered
his claim of a correlation of saturated-fat intake with heart
disease. No matter, that as Dr. Malcolm Kendrick has shown, the
data would have shown an inverse correlation if the countries
chosen had been Finland, France, Germany, Israel, Netherlands,
Sweden and Switzerland rather than Finland, Greece, Italy, Japan,
Netherlands, Yugoslavia and the U.S.A. I note also that the lack
of value of such country comparisons is illustrated by the fact
that if all these countries are compared to see if there is a
correlation between primary language spoken and the incidence
of heart disease, the correlation between speaking English or
Finnish and developing heart disease is quite impressive.
hypothesis became the conventional wisdom and was quickly adopted
and promulgated by many groups in the field of public health.
These included governmental bodies which provide advice on diet
and the American Heart Association (which had been much slower
to warn against the dangers of smoking tobacco, despite the fact
that 60% of the excess deaths from smoking are from heart disease
Ė only 13% are from lung cancer.) And, of course, manufacturers
of oleomargarine and other substitutes for natural foods were
delighted to devote their massive advertising budgets to the cause.
didnít prove a relationship between intake of saturated fat and
heart disease, but arenít there other, more convincing, studies
which indicate such a causal relationship? Kendrick has recently
surveyed the available evidence, and the following paragraphs
show what he found.
- A review
in the British Medical Journal in 2001 concluded that
"Despite decades of effort and many thousands of people randomised,
there is still only limited and inconclusive evidence of
the effects of modification of total, saturated, monounsaturated,
or polyunsaturated fats on cardiovascular morbidity and
- A Swedish
paper published in the Journal of Internal Medicine in
2005 reported on a study covering 30,000 people over six years.
The conclusion was that "Saturated fat showed no relationship
with cardiovascular disease in men. In women, cardiovascular
mortality showed a downward trend with increased saturated
- A study
involving 48,835 women and lasting more than 8 years was reported
in the Journal of the American Medical Association in
2006. Reducing consumption of saturated fat had no effect on
coronary heart disease, stroke, total mortality or cancer.
- A report
published in The Lancet in 2004 gave the results of an
international search (262 centers in 52 countries) for risk
factors for heart disease. Nine risk factors were identified.
Consumption of saturated fat was not among them.
- This last
conclusion was hardly surprising. A study reported in the Journal
of the American Medical Association in 1982 had shown that
a series of changes which included decreasing the saturated
fat consumption of 361,662 men by 28% produced no beneficial
effect on incidence of coronary heart disease or total
fail to show any detrimental effect on the heart (or anything
else) of eating saturated fat. If anything, they suggest that
there might be a beneficial effect in women. But let me explain
how eating saturated fat can be detrimental to your
health Ė it might even result in a heart attack.
familiar with the placebo effect: If you really believe that some
medicine or treatment can cure a disease condition, it might do
so just because of your belief. For instance, as much as 50% to
75% of the therapeutic effects of antidepressant medications have
been attributed to the placebo effect. The effect is real and
can be demonstrated by studies on the brain as well as by observation
of the patient.
And you may
be aware of the converse as well, but not of its name: the nocebo
effect. This effect explains the real dangers of belief in the
powers of witch doctors and of the evil eye. Hereís how researcher
R. A. Hahn defined it in Preventive Medicine in 1997: "The
nocebo hypothesis proposes that expectations of sickness and the
affective states associated with such expectations cause sickness
in the expectant. The nocebo phenomenon is a little-recognized
facet of culture that may be responsible for a substantial variety
of pathology throughout the world." This was recognized by
reporter Brian Reid, who wrote, in an article on the nocebo effect,
in The Washington Post for April 30, 2002: "Ten years
ago, researchers stumbled onto a striking finding: Women who believed
that they were prone to heart disease were nearly four times as
likely to die as women with similar risk factors who didn't hold
such fatalistic views." He was referring to data amassed
during the massive collection of data on the causes of heart disease
in Framingham Massachusetts.
effect is not often referred to as such in the medical literature,
but continues to be illustrated therein, A study made at the University
of Southern California which was reported on last year, found
that men who exhibited symptoms of anxiety (worried about saturated
fat in their diet?) were up to 76% more likely to have heart attacks
than the less anxious. Worrying about purported risk factors for
heart disease may be the way the risk factors endanger some people.
saturated fat is not harmful for most people. It may even be good
for you. But, if you have fallen for the conclusions drawn from
the bad science initiated by Ancel Keys and promulgated by the
American Heart Association, government agencies and many other
advisors on nutrition, it is a fat nocebo and could kill you.