The International Network of Cholesterol Skeptics


Discussions Nov-Dec 2002.                                                                                                           Home  

Lesley Klevay
Malcolm Kendrick
Bogdan Sikorski

Malcolm Kendrick
Bogdan Sikorski

Kilmer McCully
Malcolm Kendrick
Kilmer McCully
Malcolm Kendrick
Morley Sutter
Fred & Alice Ottoboni
Kilmer McCully
Malcolm Kendrick
William Stehbens
Malcolm Kendrick
Gunnar Lindgren
Fedon Lindberg
Malcolm Kendrick
Fedon Lindberg
Cory Mermer
Fedon Lindberg
Morley Sutter
Fred and Alice Ottoboni
Martin Sturman
Leslie Klevay
Hello:  Perhaps the Ridker article (NEJM 347:1557,2002) tells us a little
more about pathophysiology.  The quintile range for CRP is 8-fold; this
range for LDL is c 1.5.  Yet (Fig 3)  the highest quintile for both reduces
8-yr survival from 99.2% in the lowest quintile to 96.6%, less than 3%
improvement. Clearly prediction of death is difficult even when you study
28,000 people.
        Apparently one has to have a great amount of inflammation to be
harmed.  Anyone who has looked through a microscope knows that parts of
some Western arteries are inflamed.  This work may assist in therapy, but I
doubt it will assist in IHD eradication.
                                                                              Sincerely, Leslie
Malcolm Kendrick
Leslie, It is clear, at least to me, that plaques develop in a series of distinct
episodes characterised by endothelial damage/rupture followed by thrombus
formation over the area of damage. One would expect that the resulting
healing/inflammatory process would last about five to ten days.

Most plaques are not dangerous, never rupture, and you can have them for
years and years - they just narrow the artery a bit. Occasionally, a plaque
ruptures and the resulting thrombus completely blocks and artery and kills

So you are not looking at a continuous inflammatory process. You are looking
at a process that comes and goes. If you are unlucky enough to have an MI,
the CRP will increase by a factor of 1,000 over the next few days.
Essentially, this is proof that the CRP is a measurement of tissue
healing/inflammation, not just inflammation.

Once again, the terminology that people have decided to use has thrown
everyone off in the wrong direction. By deciding that CRP is a measurement
of inflammation, and thatplaques are areas of inflammation, this suggests
that CRP is some kind of inflammatory factor - and should thus be reduced.

CRP is just telling you that something is going on in the arteries. Ridker,
I know, is going to claim that CRP causes inflammation and should be
reduced. He is about to turn cause and effect upside down - and will add
further to the whole confused mess of CHD research.


Bogdan Sikorski
Hello everyone.
Attached is the first page of an article in a recent Cell (vol 111), 305-317.
It appears to be a ground-braking finding, which incidentally confirms what
my favoured medic (JK) has been preaching for some years - i.e., vegetative
neuronal system is involved in pathogenesis of most diseases, which can be
roughly divided into parasympathetic or sympathetic syndromes, depending
which side of the system is out of balance. Most such ailments can be
normalized by a proper nutrition - yes - only fat can do it. (:)
Incidentally, osteoporosis disappears in most patients who adopt ON (with a
small daily pork-skin supplement, cooked of course) within 12 months - so I
am told by those who have their hands full of such cases.
Malcolm Kendrick
My thoughts on homocysteine and infection/inflammation are, as follows.

 The process that creates atherosclerotic plaques is a combinaton of
 endothelial 'damage' followed by thrombus formation over the damaged area.
 The endothelium re-grows over the thrombus, effectively drawing it into
 the arterial wall (an inflammatory/healing process). This process,
 repeated, causes plaques to grow and eventually rupture, causing arterial
 Anything that can damage the endothelium and/or promote thrombus formation
 will therefore have a deleterious impact on Plaque formation/growth and
 rupture. Homocysteine does this, and I would imagine that various
 infective agents can do this too.
 In short, both homocysteine and infective agents can both probably
 accelerate/cause/promote plaque formation. Possibly not by themselves, but
 in combination with other factors.

Bogdan Sikorski
I have also enjoyed reading your article on "the English problem". Having
noted Paul's very valuable and accurate comments, I would also agree with
your assertion that stress is it (maybe not just it).
But I have few comments.
1. Your assumption that English and French societies are basically similar
is puzzling. When it comes to stress, French would definitely be a less
affected mob. Not only do they have a different mentality, but also a
different mode of Government (more relaxed), different law system,
different attitude to life (pleasure rules, apparently).

2. As I recall "get rich at all cost" attitude, so popular (predominant) in
the US, and England (and Australia), is not so dominant in many European
cultures, including French.

3. French work the shortest hours in a week (35) and if I recall correctly,
overtime is illegal. Thus, far less pressure at work and more work places
are preserved for the benefit of all. Totally unlike the UK, US or some
"advanced" societies. Note a recent paper in BMJ on relation between work
stress and cardiovascular problems.

4. When it comes to food intake, here again are some potentially important
differences. I might be wrong, but my recall is that main family meals in
France are eaten far earlier than in England or US. In Europe, people start
work earlier (factories 6 or 7 am) schools and offices at 8 am, and have
their main meal at 3-5 pm. English, US and Australian (not mine though)
main meal is normally taken later  (6-8 pm). And that has important
metabolic consequences, which could influence atherosclerotic morbidity.

5. In terms of a structure of society, my guess is that far more people in
France live of the land (peasants) on small plots of land. They are
basically self sufficient in food. They make it as they need it and
according to long family and regional traditions. How common is that in the
UK or US!!! (with few exceptions e.g., Amish).  It is, however, very common
in few other countries in Europe (e.g. Poland, former Yugoslavia) which
could potentially be a better comparison, if one wanted to compare specific
strata of society, rather than make such a wide inter-country comparison.

6. On that subject, I belive it would be an interesting and probably more
reveling exercise to compare two similar subgroups of a society living
either on the opposite sides of one country (Gascon region vs those from
Normandy), or in a close proximity in different countries (Alsatians from
France vs those from Germany). Now that would give some indication if high
fat intake is so bad, and if different work/political systems have
influence on health outcomes. But how would one get data for that?
Look forward to your next contribution.

Kilmer McCully
Malcolm: Your article on the French paradox is extremely well done, and I greatly
enjoyed reading it.  Homocysteine and the possible role of B6, B12 and
folate in the French paradox, however, were conspicuous by their absence in
your analysis.  The traditional French diet typically contains fresh meats,
seafood and organ meats, all of which are rich sources of B6 and B12.  Liver
contains the highest concentration of these three vitamins among commonly
eaten foods.  The French diet contains abuundant organ meats, such as liver,
brain, sweetbreads, kidney, etc., providing an exceptional source of the B
vitamins that keep homocysteine levels in the population in the 7-8
micromolar range. In addition, the tranditional Mediterranean diet contains
abundant fresh vegetables and fruits that are rich sources of B6 and folate.
As you point out, the typical diet in UK contains abundant sugar, white
flour and processed foods, that are depleted of B6 and folate.  Homocysteine
levels in UK and Northern European countries have typically been in the
10-11 micromolar range, probably because of the intake of these processed
foods that are deficient in B6 and folate.  The homocysteine levels in
countries like France, Spain and Japan are typically low, correlating with
their relatively low mortality rates from heart disease, and the
homocysteine levels in countries like UK, Germany, US and Scandinavia are
higher, correlating with their higher mortality rates from heart disease.
Of course, association does not prove causality.  Nevertheless, these
observations offer a reasonable explanation of the French paradox, in my
   Homocysteine causes intimal damage probably by increasing oxidative
stress within arterial cells and tissues.  This has been shown by
pathological changes in arteries of children with homocystinuria and in
animals given large amounts of homocysteine by injection or feeding.  In
addition injection or application of homocysteine in the form of the
reactive anhydride, homocysteine thiolactone, causes intense inflammation in
tissues of normal animals.  Also, recent experiments with cell cultures show
that homocysteine induces a release of cytokines and other mediators of the
inflammatory response.  Finally, hyperhomocysteinemia is associated with
evidence of endothelial dysfunction in human subjects. The endothelial
dysfunction, pro-inflammatory and pro-coagulant effects of homocysteine are
counteracted by folate administration, probably by lowering homocysteine
levels.  These observations suggest that the inflammatory response observed
in arteriosclerosis and heart disease is actually caused by the effect of
homocysteine on the cells and tissues of the arteries.  The current interest
in high sensitivity C reactive protein is confirmation of the inflammatory
theory of arteriosclerosis, as first enunciated by Rudolf Virchow in 1852.
His view of arteriosclerosis was that inflammation and intimal damage lead
to increased proliferation of smooth muscle cells, increased formation of
fibrous and calcified tissue, deposition of mucoid substances in artery wall
and also increased deposition of lipids that are derived from plasma
filtered through arterial wall.  The role of thrombosis over the surface of
damaged intima in arteriosclerosis was first emphasized by Rokitansky in the
1840s.  The recent findings with high sensitivity C reactive protein confirm
the observations of these pioneers of the 19th century. Malcolm, your
explanation of the pathogenesis of arteriosclerotic plaques closely follows
the lead of our predecessors, Virchow and Rokitansky!
Malcolm Kendrick

Thanks for your comments.

I do agree with you that homocysteine plays a role (probably accelerative)
in plaque development. I am never certain about how anyone knows how much B6
and folate people are eating in different countries. Every time I look at
the ingredients of anything from corn flakes to bread and yoghurts and even
choccy bars, they all seem stuffed with all the vitamins that I have heart

I suppose a more serious point is that of cause and effect - and which way
round things are.

If, like me, you believe that CRP (one example) is the sign of underlying
inflammation modulated by IL-6, and that plaque development is a form of
inflammation/healing. Then you would expect to see a higher CRP in those
with more active plaque development. However, the CRP is an effect of
underlying inflammation - not a cause. What about homocysteine levels? Are
you looking at cause, or effect?


Kilmer McCully
   It is true that foods have become increasingly fortified with multiple
vitamins in different countries.  It is also true that assessment of intake
of these vitamins in different countries is difficult to quantify.  In the
US pyridoxine (B6) and folic acid were first added to the food supply in the
1960s in the form of vitamin supplements and fortification of breakfast
cereals.  The CDC report of heart disease in the 20th century claims that
the mortality rate from "diseases of the heart" peaked in the late 1950s and
early 1960s in the US, followed by a greater than 2 fold decline in the
1970s and 1980s.  The decline in mortality rates in the US was exhaustively
examined by a national conference at the NIH in 1979.  No clear explanation
for the decline was identified among many factors examined, including
dietary fat, plasma cholesterol levels, exercise, smoking, and medical and
surgical care of heart disease patients.  As I suggested in my monograph in
Atherosclerosis Reviews, Volume 11, pages 157-246 (1983), the addition of
pyridoxine and folic acid offers an explanation for the dramatic decline in
heart disease mortality in the US beginning in the 1960s.  More recently, in
1998 the FDA  mandated addition of folic acid to fortified refined grain
products at a level sufficient to increase folate intake by 100 micrograms
per day.  The rationale was to prevent neural tube defects in children of
mothers with insufficient dietary folate.  Homocysteine levels have fallen
in the US since 1996, probably because of the fortification of refined grain
products.  It remains to be seen whether there will be a further effect on
mortality from heart disease in the US from this fortification.  It should
be mentioned that the RDI for B6 is set at 1.3-1.7 mg/day by the Food and
Nutrition Board of the National Research Council.  Results from the Nurses'
Health Study indicate that a B6 intake of 3.5 mg/day and a folic acid intake
of 400 mcg/day are needed to minimize mortality and morbidity from heart
disease.  Many processed foods are severely depleted in B6 and folic acid.
It must be emphasized that enriched grain products are not fortified with
B6.  Of course, sugar, oils and processed foods contain little if any B6 and
folic acid, explaining why populations with a high consumption of these
foods have higher homocysteine levels and higher mortality from heart
   The question of causality vs effect is difficult to settle at present
with finality.  The current methods for measuring protein-bound homocysteine
have shown that hyperhomocysteinemia is an independent risk factor for
vascular disease.  However, few investigators in the field believe that
protein-bound homocysteine causes intimal damage, pro-coagulant effects, and
other pathophysiological processes in atherogenesis.  It is much more likely
that homocysteine affects oxidative metabolism, causing increased free
radical production in arterial cells and tissues.  A theory of how
homocysteine is involved in oxidative metabolism is described in my
monograph in Annals of Clinical and Laboratory Science, Volume 23:477-493,
1993 and Volume 24:27-59, 134-152, 1994.  The many human prospective
interventional trials now in progress worldwide will add some evidence to
the question of causality when they are published over the next several
years.  To date the best controlled trial bearing on this question is the
Swiss Heart Study which showed prevention of restenosis in coronary heart
disease patients by high dose B6, B12 and folic acid therapy.  Of course,
the observations in children with homocystinuria, animals given
homocysteine, and the many in vitro studies point in the direction of
causality.  However, only further accumulation of evidence from many sources
will in the end shed light on the critical question of causality.
Malcolm Kendrick

I like your science. But you are too USA'o'centric for me. Fortification of
foods with vitamins also happened in the RoW, and the pattern of CHD is very
different. In Scotland it peaked in the late seventies. In Northern Ireland
in the mid nineties. It's still going up in Sweden and down in Finland.

It quadrupled in Russia during the nineties. In France the rate of CHD
varies by a factor of 5 from the North to the South. No dietary - content -
differences that I have managed to find.

When it comes to science I am a Popperian (if that is a word). If your
hypothesis is that vitB6/folate deficiency is THE cause of CHD, then I want
to see a one hundred per cent correlation in every population. In my black
and white world, the discovery of one black swan destroys the hypothesis
that all swans are white.

I don't believe it is good enough to say that, in a number of populations we
looked at there is a positive correlation. If a factor is truly causative
then you need to find it in all populations, preceding the disease in time.
The disease should also never be present in those without the factor. I
don't think you can get anywhere near saying this with vitB6/folate.

This does not, however, take away from my view that homocysteine probably is
a factor, like smoking, that can accelerate the development of plaques. But
on it's own, is not enough.


Morley Sutter
I agree with your reply to Kilmer but offer the following additional
All of the so-called risk factors have weak effects.  Smoking inceases one's
risk of lung cancer about 10-fold after many years of puffing.  Which "risk
factor" produces a 10-fold risk of atherosclerosis?  I know of none.
Similarly, cessation of smoking reduces risk of lung cancer almost to
non-soking levels after 5 years.  None of the putative risk factors do this
for atherosclerosis so any causal relationship they might have is weak or

Should not we be speaking of necessary and sufficient or necessary but
insufficient causes of any disease?  Which of the "causal factors" of
atherosclerosis could be spoken of in those terms.  None that I know of.
Similarly it is a "cop-out" to label atherosclerosis as multifactorial.
Everything is multifactorial in terms of causality.  We should be looking
for possible causal relationships that  can be altered, doing so in a
prospective, randomised fashion and observing outcomes.  It is hard to
reverse emigratiion but it to seems to be a very "weak causal effect",
certainly not a necessary and sufficient cause and probably not a neccessary
but insufficient one either.
Fred and Alice Ottoboni
Dear Skeptics:

We would suggest that the work of Dr. McCully be studied in detail and
carefully considered regarding the causation and control of CHD.  His work
has removed much of the guesswork from this discussion.  Homocysteine is a
powerful risk factor and certain vitamins are known to control this risk.
Obviously there are other concurrent risks that somehow fit into this
picture such as dietary high glycemic carbohydrates and the almost universal
high ratio of omega-6 to omega-3 fatty acids.  But the science surrounding
the adverse effects of high homocysteine represents one very well done piece
of work.

The comments on the varing rates of CHD in various countries are
interesting, but confusing because attack rates, death rates, total cases,
and total deaths appear to be used indiscriminately when drawing
conclusions.  As was mentioned, it is reasonable to believe that death rates
are coming down because of vitamin use and medical intervention.  Attack
rates, that is the rate of newly diagnosed cases per unit of population at
risk, appear to be rising steadily.

The new cholesterol guidelines recently published by the US National Heart
Lung, and Blood Institute is one piece of evidence that suggest that these
attack rates are rising.  Under these guidelines, the number of individuals
on statin drugs is expected to triple.  Another piece of evidence is the
rising number of various forms of surgery and other invasive procedures on
the heart and coronary arteries.

With regard to the effects of smoking on CHD, it is worth noting that in the
occupational health field, chronic carbon monoxide exposure has been a known
cause of heart disease.  A permissible limit on such harmful airborne
exposure was established years ago, even though the biochemical mechanism of
this form of carbon monoxide poisoning was not known.

A similar biochemical process apparently takes place with cigarette smoking.
Dr. McCully, in his book, "The Heart Revolution", describes this biochemical
mechanism and shows how carbon monoxide depletes certain vitamins.  This
depletion, in turn, allows homocysteine levels to rise, and the high
homocysteine levels cause hardening of the arteries (arteriosclerosis) which
set the stage for the development of atherosclerosis.  As you know, acute
carbon monoxide poisoning follows a different biochemical process involving
the formation of carboxy hemoglobin.

The process by which cigarettes cause lung cancer is different than the
process that causes heart disease.  The airways to and in the lungs are
lined with cilliated cells.  The corrosive gases in cigarette smoke cause
the cilliated cells to disappear and be replaced by squamous cells.  It is
thought that the carcinogens in the smoke (polynuclear aromatic hydrocarbons
such as benzopyrene) cause the squamous cells to undergo metaplasia and
become cancerous.  This process of change from cilliated cells to squamous
cell cancer can be seen by examining exfoliated cells using sputum cytology.
It is interesting that cilliated cells seem never to become cancerous.

Studies of smokers and former smokers using the same cytological methods
show that the cilliated cells return and replace the squamous cells that
line the airways within about 2 months after smoking cessation.  This is
thought to be the reason that studies have shown that if a smoker does not
get squamous cell lung cancer after quitting smoking, it is unlikely that
he/she will get the disease in the future.

Regards,  Fred and Alice
Kilmer McCully

In regard to your black swan analogy, this would correspond to a population
that consumes abundant dietary B6, B12 and folate, in the range of 3.5
mg/day of B6, 6 mcg/day of B12, and 400 mcg/day of folate.  If such a
population has a very high incidence of CHD, it would consitute the "black
swan" which provides evidence against the homocysteine theory of
arteriosclerosis.  In my review of literature in this area, I have been
unable to find such a "black swan."  Some of the changing patterns of CHD
incidence in different countries may be explained by fortification of the
food supply with B vitamins (for example, Hungary, which adds these three B
vitamins to bread), consumption of a diet with abundant sugar, white flour,
vegetable oils and processed foods (Northern European countries like UK), or
consumption of a diet with abundant fresh vegetables, fruits, and fresh
meat, seafood and organ meats (the Mediterranean countries like Southern
France, Spain, Greece, and Italy).  Also, the effect on CHD in the
population probably occurs over a period of years, judging by the US
experience.  A problem that I have encountered in this area is that of
obtaining well documented dietary intakes of B vitamins, allowing one to
draw conclusions about a relation to CHD in different countries.  There is a
recent study from Spain showing correlations between dietary patterns,
hyperhomocysteinemia, and CHD incidence that support the homocysteine

Thank you for your comments.

Malcolm Kendrick

A study by Despres in Canada demonstrated that syndrome X (if properly
identified, and not mixed up with type II diabetes - which is an almost
universal error) causes a 22 fold increase in the risk of CHD. Beat that!

Also, Asian Indians who emigrate to the USA increase their risk of CHD by,
up to, fifteen fold. Even better than smoking and lung cancer.

More seriously, before Kilmer starts to tell me that it's all due to
homocysteine (only joking), I agree with your general points.

My own current thinking is to think of a different model for CHD. I think
that CHD is, in fact, a series of endoethlium damaging/thrombus
forming/healing and inflammatary episodes. Each of which may, or may not,
cause plaques to form/grow/rupture. Everyone has these episodes which may be
triggered by any factor that can damage the endothelium and/or encourage
thrombus formation.

The endothelium can be 'damaged' by:

Homocysteine (in high levels)
Oxidised LDL/VLDL
Infection (with certain agents)
High levels of stress hormones etc.

Hypercoagulability can be caused by

Endothelial damage
Raised fibrinogen
Raised (other clotting factors)
Raised VLDL
Free radicals release etc.

I have no idea how many of these factors need to act, in concert, to cause a
plaque to form. But you can see that it may be possible to have two
conditions, with no overlap, causing plaque formation. This means that CHD
may have no factor that is both necessary or sufficient to cause CHD. Which
means a re-think of Koch's postulates and Bradford Hill's canons for
causation - and Occam's razor is done for as well.


William Stehbens
Dear Colleagues,

Recent arguments and opinions expressed by some sceptics could well be severely criticized by other sceptics.  Some correspondents do not understand that CHD is not a disease but a non-specific clinical diagnosis often independent of atherosclerosis per se. 

I must again stress that it is essential to learn some basic pathology of atherosclerosis and to have some knowledge of the disease before making sweeping statements regarding its genesis and progression.

See 1. - misuse of coronary heart disease. Heart 1999;82:1-2.      
2. - diagnostic error of CHD. Lancet 1987;i:606-611.
3. - causality in medical science.Perspect Biol Med 1992;36:97-119.
4. - pathogenesis of atherosclerosis and two types of early lesions. Cardiovascular Pathology 1997;6:123-153.

Yours sincerely, WE Stehbens MD DPhil
Emeritus Professor of Pathology
Dept of Pathology and Molecular Medicine.
Wellington, NZ.

Malcolm Kendrick
Dear Prof Stebhens,

Terminological inexactitude is an endless problem in medicine. It is clear
that you are certain in your own mind what atherosclerosis means. But the
definitions here are constantly changing - atherothrombosis is now coming
into vogue.

I tend to disagree that CHD may not mean atherosclerotic plaques in the
coronary arteries. A recent editorial in the BMJ suggested that CHD means
coronary artery heart disease, meaning, atherosclerosis of the coronary
arteries. I think it was Graham Jackson who wrote it. You may agree, or
disagree, but....

the important issue is not what you call the condition, it is that everyone
understands what you are talking about when you use the term. Perhaps in
this group we should agree that when we say CHD, we mean atherosclerotic
plaques (wherever they develop)? Or maybe that's not what everyone thinks?


Gunnar Lindgren
In the Swedish magazine "Forskning&Framsteg", new results from different
fields of science are reported. In the current issue nr 8/02, there is an
interesting article about the bear and its winter sleep. The author (Gunilla
Heick) is pointing at the mechanism, how the bear will store fat for the
winter. They will gain about 30 % more weight in the months before the
winter sleep. From the article:

"We have recently found that insulin affects the accumulation of fat at the
bears. The level of insulin will rise at the end of the summer, which is a
result of the enormous appetite of the bear at that time. To fill the
stomach before the winter sleep the bear eats plenty of berries. These have
a high level of sugar. The cells storing fat need insulin to transport the
sugar through the cell membrane. The more sugar the bear consume, the more
insulin it needs. The sugar is then converted to much needed fat."

I think this is an example what happens when the diet is based on a large
amount of sugar or carbs with high GI.
Fedon Lindberg
Here's another interesting connection: D vitamin deficiency has been linked
to insulin resistance (see review article: B.J.Boucher. adequate vitamin D
status. does it contribute to the disorders comprising syndrome X? British
Journal of Nutrition (1998) 79, 315-327). Restoring Vitamin D deficiency
improves insulin resistance. In this way, lack og enough sunlight exposure
(or enough sun at all as in the case of the Nordic countries) during the
winter and due to heavy clothing may affect both insulin sensitivity, blood
sugar levels and eating patterns, causing weight gain during the darker
period of the year. It is also well known that immigrants from South East
Asia or Africa tend to get explosive rates of obesity and diabetes when they
move to Northern Europe and they also have a high prevalence of vitamin D
deficiency. May be then we can conclude that lack og sunshine contributes to
insulin resistance, diabetes and obesity. Not so strange, after all, humans
need quite a lot of sunshine...
Best regards
Fedon Lindberg
Malcolm Kendrick

Emigrant Asian Indians get hugely high rates of central obesity and syndrome
X (confused with diabetes) when they move anywhere - sunny or not. The rate
of (what is diagnosed as) type II diabetes is just as high in South African
Asian Immigrants, as it is in the UK.

Also Asian Indians who move from rural environment in India to the cities
increase their rate of (what is incorrectly diagnosed as) type II diabetes
from 1.5% to 9%.

It is not a lack of sunlight that causes (what is incorrectly diagnosed as)
type II diabetes, it is emigrating.


Fedon Lindberg

Dear Malcolm,

I did not mean that Vitamin D is the cause of insulin resistance, but it
could be a contributing factor. I am also aware that almost all that
emigrate to westernized countries increase their rates of obeity and insulin
reisutance-diabetes etc.
However, it has been shown that vitamin D deficiency is common also in their
native countries (in sunny, warm climates, people tend to avoid direct
sunlight exposure). In any case, I find it interesting that insulin
sensitivity improves after correction of vitamin D deficiency. Insulin
resistance is a condition where genetics and environment both play a role,
and Vitamin D deficiency may be one of many environmental factors.

Cory Mermer
Regarding insulin resistance in emigrants, perhaps I am thinking too simply
here, but I always thought that it mainly resulted from their adoption of a
Western diets, so high in sugar.  Their bodies go into "shock" from
overdosing on sodas, candy, cookies, etc.  Can the answer be as simple as

That is not to say that stress from relocation and vitamin D deficiency
can't be factors as well.  But wouldn't the fact that their bodies are being
forced to produce more and more insulin be a more immediate threat?

Fedon Lindberg
Definitely so Cory, but in fact, not all immigrants adopt a high sugar
lifestyle. Myself, I believe that
for sure a high glycemic load, less vegetables, stressful environment,
toxins, transfat, too much omega 6 too little omega 3, too little physical
exercise are main determinants of insulin resistance on the basis of a
genetic background that allows it, but I find the fact that Vitamin D
defficiency is a contributing factor interesting.
Morley Sutter
Ms (Martin and Malcolm0
I am compelled to comment on the debate regarding statistics and causality.  Statistics is a method of distinguishing between populations (or a sample); staistics can never demonstrate a causal relationship.  I would add that causality can never be demonstrated post hoc or retroactively, i.e., the study must be prospective and designed to be as free of bias as possible. The design of the study is all-important,   All epidemiological (non-interventional) studies can demonstrate correlations only and correlation is not equivalent to causation. (Sorry for the pedantic statements but I am trying to establish some ground rules).
Popper further maintains that causation (a hypothesis) can never be proven, only disproven or falsified.  I disagree slightly with this, but agree in general terms and it is a good starting point.  The upshot of these points of view is that establishing causality is illusive, complicated and the goal of all investigation but seldom achieved.  I think that our epidemiologists should be much more circumspect in their conclusions.  A big problem in the news media is that "associated with" is often interpreted to mean "causal". A rooster crowing is "associated" with the dawn.
Fred and Alice Ottoboni
Dear Morley:
You are correct.  Associations cannot prove cause-effect.  Many years ago we covered this issue thoroughly (and at our extreme peril) in an epidemiology class at the UC School of Public Health.
The answer is associations are great clues to what may be happening.  But to know for sure, one must turn to hard science.  A simple example is the classic "Church Picnic" where some of the attendees come up with diarrhea and vomiting 7 hours later.
First you do interviews and find the apparent association.  Those who ate the chicken salad got sick -- few of those who did not eat the chicken were affected.  This is a clue, not an answer.  The answer is to do stool samples or other laboratory tests on the victims (and the non-victims) to discover the actual organism responsible.
To prove the associations surrounding the statin drugs, controlled dietary feeding studies, and reliable biochemical, and physiological measurements are required.  Sometimes, these proof methods are impossible or extremely difficult.  But these difficulties do not justify our treating an association as a fact or near fact.
It has been very interesting following your discussions of these issues.


Martin Sturman

Thanks for your comments and for your appealing phrase-making. I think we are in general agreement about causality, and association as discussed in Morley's Sutter's superb letter.  Did you know, by the way, that 99% of people who die in auto accidents eat potatoes?

Here is the only place we seem to diverge semantically -and politely, of course, without any rhetorical extension of the legs:
I take your remark that "most trials are carried out pretty well."  to mean, done with the usual schematic orthodoxy. True, but do you not think that, along with mistaken interpretation of data, one of the other serious failures in clinical research is in constructing and planning a good trial and not simply conducting one?  As an aside, I quote Peter Medawar, "The problem must be such that it matters what the answer is."

Excellent discussion of Circulation paper, by the way, and much appreciated. 

Best regards,