Authors: T. Colin Campbell,Thomas M. Campbell
THE CHINA STUDY
188
obvious reasons, this is one of the most contentious issues in nutrition
today.
One of the more remarkable reports on this cow's milk effect was
published over a decade ago, in 1992, in the New England Journal of
Medicine. 12 The researchers, from Finland, obtained blood from Type
1 diabetic children, aged four to twelve years. Then they measured the
levels of antibodies that had formed in the blood against an incomplete-
ly digested protein of cow's milk called bovine serum albumin (BSA).
They did the same process with non-diabetic children and compared
the two groups (remember, an antibody is the mirror image, or "mold,"
of a foreign antigen) . Children who had antibodies to cow's milk protein
must have previously consumed cow's milk. It also means that undi-
gested protein fragments of the cow's milk proteins had to have entered
the infant's circulation in order to cause the formation of antibodies in
the first place.
The researchers discovered something truly remarkable. Of the 142
diabetic children measured, every Single one had antibody levels higher
than 3.55. Of the seventy-nine normal children measured, every single
one had antibody levels less than 3.55.
There is absolutely no overlap between antibodies of healthy and
diabetic children. All of the diabetic children had levels of cow's milk
antibodies that were higher than those of all of the non-diabetic chil-
dren. This implies two things: children with more antibodies consumed
more cow's milk, and second, increased antibodies may cause Type 1
diabetes.
These results sent shock waves through the research community. It
was the complete separation of antibody responses that made this study
so remarkable. This study,12 and others even earlier,1 -17 initiated an ava-
5
lanche of additional studies over the next several years that continue to
this day.13. 18. 19
Several studies have since investigated this effect of cow's milk on
BSA antibody levels. All but one showed that cow's milk increases BSA
antibodies in Type 1 diabetic children,18 although the responses were
quite variable in their magnitude.
Over the past decade, scientists have investigated far more than
just the BSA antibodies, and a more complete picture is coming into
view. Very briefly, it goes something like this 13 , 19: infants or very young
children of a certain genetic background,2o,21 who are weaned from the
breast too early22 onto cow's milk and who, perhaps, become infected
AUTOIMMUNE OISEASES 189
with a virus that may corrupt the gut immune system,l9 are likely to
have a high risk for Type 1 diabetes. A study in Chile23 considered the
first two factors, cow's milk and genes. Genetically susceptible children
weaned too early onto cow's milk-based formula had a risk of Type 1 di-
abetes that was 13.1 times greater than children who did not have these
genes and who were breast-fed for at least three months (thus minimiz-
i n g their exposure to cow's milk). Another study in the U.S. showed that
genetically susceptible children fed cow's milk as infants had a risk of
disease that was 11.3 times greater than children who did not have these
genes and who were breast-fed for at least three months. 24 This eleven
to thirteen times greater risk is incredibly large (1,000-1,200%!); any-
t h i n g over three to four times is usually considered very important. To
put this in perspective, smokers have approximately ten times greater
risk of getting lung cancer (still less than the eleven to thirteen times
risk here) and people with high blood pressure and cholesterol have a
2.5-3.0 times greater risk of heart disease (Chart 9.2) .18
So how much of the eleven to thirteen times increased risk of Type
1 diabetes is due to early exposure to cow's milk, and how much is due
CHART 9.2: RELATIVE RISKS OF VARIOUS FACTORS
ON VARIOUS DISEASE OUTCOMES
1,200%
1,000%
c
0
800%
'p
ru
:::J
Heart Type 1
ru
600%
> Disease Diabetes
L.LJ
-'t::
Vl
a: 400%
200%
High Blood Pressure Cow's Milk
Smoking
+ High-Risk Genes
and Cholesterol
Risk Factors
THE CHINA STUDY
190
to genes? These days, there is a popular opinion that Type 1 diabetes is
due to genetics, an opinion often shared by doctors as well. But genetics
alone cannot account for more than a very small fraction of cases of this
disease. Genes do not act in isolation; they need a trigger for their ef-
fects to be produced. It has also been observed that after one member of
identical twin pairs gets Type 1 diabetes, there is only a 13-33% chance
of the second twin getting the disease, even though both twins have
the same genes. l 3 , 20, 21, 25. 26 If it were all due to genes, closer to lOO% of
the identical twins would get the disease. In addition, it is possible that
the 13-33% risk for the second twin is due to the sharing of a common
environment and diet, factors affecting both twins.
Consider, for example, the observation shown in Chart 9.3, which
highlights the link between one aspect of environment, cow's milk con-
s u m p t i o n , and this disease. Cow's milk consumption by children zero
to fourteen years of age in twelve countries 27 shows an almost perfect
correlation with Type 1 diabetes. 28 The greater the consumption of
cow's milk, the greater the prevalence of Type 1 diabetes. In Finland,
Type 1 diabetes is thirty-six times more common than in Japan. 29 Large
amounts of cow's milk products are consumed in Finland but very little
is consumed in JapanY
CHART 9.3: ASSOCIATION OF COW'S MILK CONSUMPTION AND
INCIDENCE OF TYPE 1 DIABETES IN DIFFERENT COUNTRIES
40
~ 30
(J)
~
o
o
8" 20
~
.s DENMARK
(J)
u
: NEW ZEALAND
c
~ 10 • NETHERLANDS
'u
CANADA
c
JAPAN • • ISRAEL
o~.-_·~~~FR~A~N~CE~.-~~~~r-~.-~~--r
o 200 300
100
Cow Milk Consumption (liters/person/year)
AUTOIMMUNE DISEASES 191
As we have seen with other diseases of affluence, when people migrate
from areas of the world where disease incidence is low to areas of the
world where disease incidence is high, they quickly adopt the high inci-
dence rates as they change their diet and lifestyle. 3O-32This shows that even
though individuals may have the necessary gene(s) , the disease will occur
only in response to certain dietary and/or environmental circumstances.
Disease trends over time show the same thing. The worldwide preva-
lence of Type 1 diabetes is increasing at an alarming rate of 3% per year. 33
This increase is occurring for different populations even though there
may be substantial differences in disease rates. This relatively rapid in-
crease cannot be due to genetic susceptibility. The frequency of anyone
gene in a large population is relatively stable over time, unless there are
changing environmental pressures that allow one group to reproduce
more successfully than another group. For example, if all families with
Type 1 diabetic relatives had a dozen babies and all the families without
Type 1 diabetic relatives died off, then the gene or genes that may be
responsible for Type 1 diabetes would become much more common in
the population. This, of course, is not what is happening, and the fact
that Type 1 diabetes is increasing 3% every year is very strong evidence
that genes are not solely responsible for this disease.
It seems to me that we now have impressive evidence shOWing that
cow's milk is likely to be an important cause of Type 1 diabetes. When
the results of all these studies are combined (both genetically suscep-
tible and not susceptible) , we find that children weaned too early and
fed cow's milk have, on average, a 50-60% higher risk of Type 1 diabetes
(1.5-1.6 times increased risk).34
The earlier information on diet and Type 1 diabetes was impressive
enough to cause two Significant developments. The American Academy
of Pediatrics in 1994 "strongly encouraged" that infants in families
where diabetes is more common not be fed cow's milk supplements for
their first two years of life. Second, many researchers 19 have developed
prospective studies-the kind that follow individuals into the future-
to see if a careful monitoring of diet and lifestyle could explain the onset
of Type 1 diabetes.
Two of the better known of these studies have been underway in Fin-
land, one starting in the late 1980S1 and the other in the mid-1990s. 35
5
One has shown that cow's milk consumption increases the risk of Type
1 diabetes five- to sixfold,36 while the second35 tells us that cow's milk
increases the development of at least another three to four antibodies
THE CHINA STUDY
192
in addition to those presented previously (p. 190). In a separate study,
antibodies to beta-casein, another cow's milk protein, were significantly
elevated in bottle-fed infants compared to breast-fed infants; children
with Type 1 diabetes also had higher levels of these antibodies. 37 In
short, of the studies that have reported results, the findings strongly
confirm the danger of cows milk, especially for genetically susceptible
children.
THE CONTROVERSY OF CONTROVERSY
Imagine looking at the front page of the newspaper and finding the
following headline: "Cow's Milk the Likely Cause of Lethal Type 1
Diabetes." Because the reaction would be so strong, and the economic
impact monumental, this headline won't be written anytime soon,
regardless of the scientific evidence. Stifling this headline is accom-
p l i s h e d under the powerful label of "controversy." With so much at
stake, and so much information understood by so few people, it is easy
to generate and sustain controversy. Controversies are a natural part
of science. Too often, however, controversy is not the result of legiti-
m a t e scientific debate, but instead reflects the perceived need to delay
and distort research results. For example, if I say cigarettes are bad for
you and provide a mountain of evidence to support my contention,
the tobacco companies might come along and pick out one unsolved
detail and then claim that the whole idea of cigarettes being unhealthy
is mired in controversy, thereby nullifying all my conclusions. This is
easy to do, because there will always be unsolved details; this is the
nature of science. Some groups use controversy to stifle certain ideas,
impede constructive research, confuse the public and turn public
policy into babble rather than substance. Sustaining controversy as a
means of discrediting findings that cause economic or social discom-
fort is one of the greatest sins in science.
It can be difficult for the layperson to assess the legitimacy of a highly
technical controversy such as that regarding cow's milk and Type 1 dia-
betes. This is true even if the layperson is interested in reading scientific
articles.
Take a recent scientific review38 of the cow's milk-Type 1 diabetes
association. In ten human studies (all case-control) summarized in a
paper published as part of a "controversial topics series,"38 the authors
concluded that five of the ten studies showed a statistically Significant
positive association between cow's milk and Type 1 diabetes and five did
AUTOIMMUNE DISEASES 193
not. Obviously, this at first seems to demonstrate considerable uncer-
tainty, going a long way to discredit the hypothesis.
However, the five studies that were counted as "negative" did not
show that cow's milk decreased Type 1 diabetes. These five studies
showed no statistically significant effect either way. In contrast, there
are a total of five statistically significant studies and all five showed the
same result: early cow's milk consumption is associated with increased
risk of Type 1 diabetes. There is only one chance in sixty-four that this
was a random or chance result.
There are many, many reasons, some seen and some unseen, why an
experiment would find no statistically significant relationship between
two factors, even when a relationship really exists. Perhaps the study
didn't include enough people, and statistical certainty was unattainable.
Perhaps most of the subjects had very similar feeding practices, limiting
detection of the relationship you might otherwise see. Maybe trying to
measure infant feeding practices from years ago was inaccurate enough
that it obscured the relationship that does exist. Perhaps the researchers
were studying the wrong period of time in an infant's life.
The point is, if five of the ten studies did find a statistically significant
relationship, and all five showed that cow's milk consumption is linked
to increasing Type 1 diabetes, and none show that cow's milk consump-
tion is linked to decreasing Type 1 diabetes, I could hardly justify say-
ing, as the authors of this review did, that the hypothesis "has become
quite murky with inconsistencies in the literature. "38
In this same review,38 the authors summarized additional studies that
indirectly compared breast-feeding practices associated with cow's milk
consumption and Type 1 diabetes. This compilation involved fifty-two
possible comparisons, twenty of which were statistically significant. Of
these twenty significant findings, nineteen favored an association of cows
milk with disease, and only one did not. Again the odds heavily favored the
hypotheSized association, something that the authors failed to note.
I cite this example not only to support the evidence showing a cow's
milk effect on Type 1 diabetes, but also to illustrate one tactic that is
often used to make something controversial when it is not. This practice
is more common than it should be and is a source of unnecessary con-
fusion. When researchers do this-even if they do it unintentionally-
they often have a serious prejudice against the hypothesis in the first
place. Indeed, shortly after I wrote this, I heard a brief National Public
Radio interview on the Type 1 diabetes problem with the senior author