Index Home About Blog
From: ((Steven B. Harris))
Subject: Re: The USDA Food Pyramid
Date: 11 Jul 1995
Newsgroups: sci.skeptic

In <3troc4$> (Lawson English)

>Any nutritionist at your local hospital should be able to verify the
>Actually, if you think about it, most adults in this country should NOT
>eat very much meat at all.
>At least not beef or pork. Poultry and fish, maybe, but not beef or pork.
>Pure vegan diets are kinda stupid, also, but if you look at the
>vegetarianism of the more successful vegetarian religions, they generally
>aren't opposed to eggs, cheese, milk, and sometimes not even fish, but
>mostly red meat.

ROFL.  They are what Dr. Castelli calles "ovo-lacto-pesco-pollo"
vegeterians.  In other words, not vegetarians at all.  Chicken has as
much saturated fat in it as most beef.  And pork ("the *other* heart
attack meat") does also.

                                   Steve Harris, M.D.

From: B. Harris)
Subject: Re: Dietary, Blood Cholesterol Relationship
Date: 18 Dec 1996

In <> (Noel A.
Taylor) writes:

>>IMOHO and from what I've read from various sources, its deposition on
>>arterial walls is a natural function of age, things like smoking and
>>obesity tend to accelerate age. Genetics also plays a part in its
>>deposition (and ageing).
>Ornish's work clearly demonstrates, with over 25 years of detailed
>research, that this "natural function of age" is actually a natural
>function of poor diets. He makes clear that cholesterol is not the
>culprit; but the outstanding effectiveness of his diet makes clear that
>arterial plaquing is not a natural function at all.
>						--Noel

   On the contrary, the reason Ornish's diet is vegetarian (vegan) is
that he has deliberately designed it to have zero cholesterol, and a
vegan diet is the only type of diet which does.  If Ornish were not
afraid of dietary cholesterol, his diet would certainly contain
shellfish and deep water fish, which do not add cholesterol-raising
fats to a diet, but only pre-formed cholesterol.

   It is true that the major determinant of cholesterol blood levels is
saturated fat intake.  However, dietary pre-formed cholesterol, and
particularly pre-formed oxidized cholesterol, also raises cholesterol
blood levels.

                                             Steve Harris, M.D.

From: B. Harris)
Subject: Re: Dietary, Blood Cholesterol Relationship
Date: 21 Dec 1996

In <> (dalford) writes:

>In article <599mko$>,
> B. Harris) wrote:
>>    It is true that the major determinant of cholesterol blood levels is
>> saturated fat intake.  However, dietary pre-formed cholesterol, and
>> particularly pre-formed oxidized cholesterol, also raises cholesterol
>> blood levels.
>The correlation is weak. South African chicken farm workers digest avg.
>1200 mg. cholesterol/daily and yet have serum cholesterol levels lower
>than average Americans.
>Further, the lowest serum cholesterol levels may not be optimal.

   The effect of cholesterol from eggs has been shown to be sizable in
metabolic ward studies adding eggs to low fat diets.  See for example,
Mattson et al (Am J Clin Nutr 25:589-594, 1972) where going from
cholesterol intake of 0 to 245 to 550 to 730 mg per day, raised
cholesterol from 160 to 173 to 184 to 201 mg per dl.  This is not a
trivial amount, as it goes from a level where heart disease is very
rare, to a level which represents the average level of the US, and one
at which heart disease starts to get more common.  As for where the
optimal level of blood cholesterol is, that's a good question.  In the
largest population study known to me (done in China on 250,000 people),
there is no clear best cholesterol for overall death rate, and this
study was done in a mostly vegetarian population for which serum
cholesterol AVERAGED about 125, and went below 100 for a sizable
fraction of persons.

Addition of eggs to diets higher in fat has had lesser effect-- but
then, who said it didn't?  The argument is whether or not dietary
cholesterol has major effects on serum cholesterol under any
circumstances.  Clearly, if you're eating a very good low fat diet, it
does.  The same results, BTW, had also been shown in some very
interesting studies in which eggs were added to the diets of
"primitive" peoples eating paleolithic diets.

                                       Steve Harris, M.D.

From: B. Harris)
Subject: Re: Dietary, Blood Cholesterol Relationship
Date: 21 Dec 1996

In <> (dalford) writes:

>In article <>, wrote:
>>I must add that you may very well add meat to a healthy diet with no
>>certain negative effects and some positive (since meat is a good source
>>of iron, zinc and vitamin B12). Hunter-gatherers eat more meat than
>>westerners and have about 50% lower serum cholesterol levels and
>>possibly no atherosclerotic heart disease. It seems reasonable that
>>humans are adapted to meat which they most certainly have been eating
>>for more than two million years.
>That's right.

  It's only right if you're talking about game meat.  Not too many
people eat only game meat.

   Nevertheless, the failure of game meats to raise cholesterol levels
much, is one fact which points to the relative importance of oxidized
vs unoxidized cholesterol in raising serum cholesterols.

                                          Steve Harris, M.D.

From: B. Harris)
Subject: Re: Dietary, Blood Cholesterol Relationship
Date: 21 Dec 1996

In <> (dalford) writes:

>In article <59fk2e$>,
> B. Harris) wrote:
>>   It's only right if you're talking about game meat.  Not too many
>> people eat only game meat.
>>    Nevertheless, the failure of game meats to raise cholesterol levels
>> much, is one fact which points to the relative importance of oxidized
>> vs unoxidized cholesterol in raising serum cholesterols.
>Not just game meat, but other lean meats such as fish, pork, turkey, etc.
>Also, recent research indicates game meat and fish meat not only doesn't
>raise serum cholesterol level, it actually lowers it.

   Sorry, game and fish, perhaps.  But if you believe that pork and
turkey lower blood cholesterol, you're just being taken in by the usual
advertising ("pork-- the other white meat...").

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 30 Apr 1997

In <> Jack Ludwick <> writes:

>I remember the "no heart attacks of people with cholesterol less than
>150" quote, but it referred to participants in the Framingham study.

   Correct!  I don't think anybody claimed that nobody on Earth with a
habitual cholesterol less than 150 has ever had an MI.   But such
things are *rare.*  Sometimes other factors in such cases can be
identified, such as extra low HDLs, a history of chest radiation, or
childhood coronary damage (Kowasaki disease).  Or maybe there are lots
of other risk factors combined, like dialysis, diabetes, hypertension,
and smoking-- and the poor low cholesterol is overwhelmed.  Other times
you can't tell what went wrong.  Still, my point stands.  If your
cholesterol is less than 150 all the time, odds are good that you're
going to die of cancer, not coronary artery disease.

                                        Steve Harris, M.D.

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 02 May 1997

In <5kdk4t$g6u$> (Maureen Soar)

>So, my question is this: I can't see ANY possible harm from trying the
>Ornish diet at all, so why not recommend it to patients?

    For just the reasons you might think:

A:  95% of people will not be able to follow it, so you're prescribing
something that probably will not work.

B: While they (and you) find out they can't follow it, they are at risk
which they wouldn't have been if you just gave them a cholesterol
lowering drug.

C: Insurance companies and MediCare DON'T reimburse for the
extraordinary amounts of time which dietary therapy takes to

   Bottom line: for the 95% of people who cannot do a very low fat
diet, quality of life is impacted unnecessarily.  So also for the
physician who tries to make them do it.  Never try to teach a pig to
sing, says Lazarus Long: it wastes your time, and it annoys the pig.
Save the Ornish routine for your very motivated and very anal retentive
patients, who can use it not only to save their lives, but to fill
their time with something that calms their anxieties.  For everybody
else, there's Pravachol and Zocor.

                            Steve Harris, M.D.

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 04 May 1997

In <> writes:

>>  For everybody
>> else, there's Pravachol and Zocor.
>>                             Steve Harris, M.D.
>And now, Lipitor.

    Which I just prescribed to a lady with triglycerides of 500 on
Zocor plus 15 grams a day of fish body oil (and who didn't drink and
didn't have a high sugar intake and so on....).  I'm keeping my fingers
crossed.  Fish oil is great, but it doesn't nail triglycerides every
time, for some reason.  In a few people it doesn't work at ALL.


From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 04 May 1997

In <> xquark <> writes:

>My earlier defense of Dr Harris does not include support his snide and
>insulting sneer at those who could follow Ornish's diet.  "...anal
>retentive.."   Uncalled for and insulting to patients who do have the
>fortitude to make monumental changes.   He knows better or should.

   I don't.

    I suppose you know many carefree and easygoing and devilmaycare and
free-spirited persons who've successfully switched to the Pritikin
plan.  SUURRE you do.

    I wasn't using "anal-retentive" pejoratively here.  I was using it
in the sense of somebody who engineers, makes checklists, keeps
appointments, etc, etc, etc.   The kind of thing that gets you though
grad school, or what makes a good secretary.  Incredible attention to
incredible detail.   The kind of thing that keeps you alive in bad
situations, and when you have a bad disease.  Some people can keep this
up for a lifetime, and some can't.

                                      Steve Harris, M.D.

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 04 May 1997

In <>

> Are you claiming that cholesterol loweing drugs are just as effective
>as the Ornish protocol, or just the best we can do for people who will
>cheat on the Ornish plan anyway?

   The cholesterol drugs are just as efective at lowering cholesterol.
As for cutting the risk of death or even MI, there's no data on the
Ornish diet-- just one study in which it decreases (slightly) the
amount of plaque.  But cholesterol drugs do this also.  There IS data
that cholesterol drugs decrease risk of both MI and death (the 4S
Simvastatin trial, and also a Pravachol trial), so the evidence that
these "work" is far better than for Ornish.  Simply a function of the
fact that drug companies have more money to spend for trials, I think.
But I can't before.  The evidence is better for drugs at this point--
that's all one can say.  We KNOW they cut risk of death, and for Ornish
that hasn't been proved (though I certainly suspect it).

>What about the exercise and stress management components of the Ornish
>protocol? How important are they?

   Nobody knows.  Not even Ornish.

>Do you have any studies comparing these different approaches for
>people who have been diagnosed with heart disease?
>Oliver Costich

    There AREN'T any!   Blame the Federal government for that.   The
NIH is very conservative, which means that NIH money follows seed
research.  Only drug companies have much money for seed research.
If you've read Ornish's books, you'll see what problems he had getting
together private money to do his initial study.  He has some government
money now, but you can bet he's STILL not testing his diet head to head
with drugs.  It's the Feds' job to make sure studies like that get
done, and they drop the ball, generally.  Which is why we don't know
more about conventional vs alternative approaches from most diseases.

                                       Steve Harris, M.D.

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 04 May 1997

In <5kd6ic$> (Chris Malcolm)

>> Still, my point stands.  If your
>>cholesterol is less than 150 all the time, odds are good that you're
>>going to die of cancer, not coronary artery disease.
>1. I'm sure this applies to people whose chl happens to be < 150, but
>does it also apply to people whose chl normally would be > 150, but
>who took special dietary or drug measures to reduce it?

    I don't think anybody knows.  Probably not.  Certainly risk falls
drastically when you get your LDL below 100, but it takes time for
atherosclerosis to go away.  And I'm not sure it ever does, completely.
While you have lesions left, you're still at risk for clot or spasm
caused by one.

>2. Isn't dying of coronary artery disease a lot less unpleasant than
>dying of cancer?
>Chris Malcolm         +44 (0)131 650 3085
>Department of Artificial Intelligence,    Edinburgh University
>5 Forrest Hill, Edinburgh, EH1 2QL, UK                DoD #205

    On average, probably, but certainly not necessarily.  You might
read the book _Heartsounds_ which generally tells it like it is.  If
you die of arrhythmias that's quick, but dying of slow congestive
failure with a lot of angina can be as bad as any cancer.  Just as much
pain, and just as much cachexia and general interference with vitality.
 And it can last a lot longer than any cancer, too.  The average
survival with bad congestive failure and a terminal cancer are about
the same, but heart failure has a tail of 10 or 20% of people who just
go on and on and suffer forever.  Two, three, four years.  They're the
reason it's hard to get heart patients into some hospice programs,
though unfortunately endstage heart patients NEED hospice care as much
as any terminal patient does.

                                      Steve Harris, M.D.

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 04 May 1997

In Message-ID: <>
"David L. Hanson" <> wrote

>Dr. Harris apparently doesn't understand the health benefits of good
>nutrition and is quick to prescribe drugs for everyone. When heart
>patients can easily reverse heart disease by changing their diet to a
>vegetarian one, he blindly does what most doctors do, keeps giving them
>poisons. I would guess that heart disease rates among doctors is just as
>high as in the rest of the population. If they ate very low fat
>vegetarian (which is very close to what the rural people in China eat)
>then doctors could live longer and healthier also.
> David L. Hanson


Sigh.  I don't understand the benefits of good nutrition.  Right.  As
it happens, Roy Walford and I, in the proceedings of the National
Academy of Sciences a couple of years ago, published a study of one of
the more profound dietary regimes on cholesterol and other health
parameters.  This was the home-grown basically vegetarian totally
organic diet inside Biosphere II, the sealed eco-commune in Arizona.
This diet was diet much like the Ornish/Pritikin regime, but low
calorie as well. The cholesterols of the Biospherians dropped from 190
average to 125, which as it happens is just about the average for rural
China, which fact we mentioned in the paper itself..  These Biosphere
folks weren't sneaking pizza.

Alas, one reason they weren't is that they were sealed into the
Biosphere, with a tamper-proof seal which could be broken only in
emergencies.  So they had pretty strong incentives to stick with it.
And they were all pretty motivated types to begin with.  Alas, again,
when they got out, their cholesterol levels went right back up again
(data still to be published).   The thing about a very low fat diet is
that humans will rarely stick with it if given a choice.  We're
bred to like calories, since evolutionarily, starvation was a lot
bigger problem for us than dying of MI at age 60.   Your genes, after
all, don't really "care" what happens to you past 50.  Maybe 40.  Your
palate is programmed accordingly, and the only people who can
eat differently but don't, are pretty odd.  In rural China
the average peasant is sort of stuck, too, when you think about it
The average patient I see in the office is neither as motivated as
the self-selected patients of Ornish, or the locked-in denizens of
China or Biosphere II..

I really have no idea what the average heart disease rate is for
physicians, although the Physician's Health Study results suggest to me
that the heart disease rate among older Harvard trained physicians is
quite low (see N Engl J Med 321:129-35, 1989 and N Engl J Med
334:1145-9, 1996 for descriptions of the problems with the unexpectedly
low heart disease rates in doctors, including the placebo group, caused
this preventive study of aspirin and beta carotene.)   No, I'm sure
that heart disease rates in American physicians aren't as good as
peasants in mainland China. But we're not stuck in mainland China,
either   Again, if the mainland Chinese got the opportunity to
liberalize their diets, they'd do exactly what the modern Japanese have
done.  And what emigrants from Japan did in Hawaii and California long
before that.  There's no making an entire population eat a healthy diet
without quite extraordinary coercion.

Again, eating is THE most basic human drive. Yep, it's more important
and powerful than sex, by far, and those jokes about the virtues of
chocolate vs men have a lot of evolutionary bite in them..  Of course,
the basic reason for all  this is that eating is about reproduction
too, but it's far more difficult to do  in a state of nature than
having sex. Your ancestors (particularly your female ancestors)  didn't
have nearly as much problem finding sex as they had finding enough to
eat, and your genes still "know" that.  Sexual attractiveness is one of
the few reasons people successfully starve themselves, and even that
doesn't work that well.  As for heart disease prevention, it doesn't
even show up as a blip on Mother Nature's health  list.   Forget
about it.  You can blame Western medicine all you like, but
you're just as silly in that as if you blamed me as an American
doctor for the country's teenage pregnancy rate.  The truth of the
matter is that unless you more or less lock them up, or
chaperone them all the time, people will do what people will do.
This is America, and I'm a doctor, not a jailer.  And due to
the large overhead in office staff, loan repayment, license fees,
malpractice insurance and whatnot, I'm afraid I charge too much
to be a babysitter also.  Much as some readers seem to expect it.

                                        Steve Harris, M.D.

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 05 May 1997

In <> "Ed Mathes, RPA-C"
<> writes:

>...yes. nutritional
>counseling is 'labor intensive' BUT, that's what they make dieticians
>Edward J. Mathes, RPA-C   ! "The greatest derangement of the mind
>Family Practice           !  is to believe in something because
>      !  one wishes it" L. Pasteur 1876

    I thought they made dieticians so that you can refer your patient
to them, observe the cholesterol drop by only 10 mg/dl in 3 months, and
the weight actually increase by 2 lbs due to starve-binge, so that THEN
you can then give the Zocor and maybe the Redux, without guilt.  That's
what dieticians are for.

    I seem to find all the gonzo alternative types on the net.  What I
really want is one in my area, so that I can refer my problem
cholesterol and weight patients to him or her. <g>.  It's so much more
appropriate than just TELLING them they are unrealistic.

    The problem with guys like Ornish is that they work with a small
subset of very motivated and probably differently wired people (in
terms of calorie and fat hunger), and the public wants to think that
their results apply to everyone, and blames ME when they don't.  At
least I'm not so base as to blame in turn the poor dietician.  Or even
the patient.  It's not your fault when being at your "healthy" body
weight causes such hunger that you can't function.  That's no life.
It's rather a disease condition that demands something be done, rather
like chronic pain.  If nothing works but medication, then medication it
needs to be.

                                   Steve Harris, M.D.

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 06 May 1997

In <> Lisa Nuttall
<> writes:

>In article <5kk1ad$>, "Steven B. Harris"
><> writes
>> If nothing works but medication, then medication it
>>needs to be.
>I don't think there can be any argument with this, but your first post
>seemed to imply that nothing will work but medication, therefore there
>is no point in trying lifestyle modification.  Have I misunderstood?

   On average, you are correct.  So do we waste the time of the 95%,
and their doctor, in order to find and optimally treat the 5%?  Or
maybe it's 3%?   I only know that as a geriatrician, the fraction of
persons I could get to seriously modify their eating habits was so
small, that I eventually gave up.  If you think I gave up too soon, I
invite you to open up a geriatric clinic and walk a mile, or a year, in
my shoes.  Then judge.

>I'm one of those patients you describe who can't/won't adapt to an full
>Ornish style regime.  I haven't tried, because I know I wouldn't stick
>with it.  But I have modified my diet towards that ideal, I sometimes
>eat a little meat, and quite a lot of fish, but overall have reduced
>the fat in my diet substantially.

    Yeah?  Did you make a substantial difference in your cholesterol
levels?  Let's speak truth, now.

>  Don't you think its worth trying the diet
>with your patients either before or as well as prescribing the
>cholesterol lowering drugs?


>By the way my doctor told me not to worry at the moment about my high
>cholesterol as doing anything about it would mean going on a "horrible
>diet I wouldn't like at all."  I have angina.

    Your doctor is wrong.  You can lower cholesterol and risk factors
quite effectively with drugs alone.  I know that doesn't appeal to our
Puritan sense of getting something quite valuable for comparitively
little suffering, but it's true.

                                            Steve Harris, M.D.

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 07 May 1997

In <> Lisa Nuttall
<> writes:

>My point, perhaps I didn't express it clearly, was that if by offering
>diet advice and support you can persuade some of the 95% to make some
>changes, as much as they will willingly undertake, then they will gain
>some benefits. I can't believe that it's as black and white as you seem
>to feel - that either the diet has to be 'seriously' modified or its not
>worthwhile. Maybe we'd need to agree what we mean by serious changes to
>be any clearer here.

    Probably.  Whatever changes you make have to get your LDL down to
best levels.  That's below 160 (US units of mg/dl) if you have no risk
factors, below 130 if you have two, and below 100 if you actually do
have coronary disease.

>I would imagine that geriatric patients are the hardest group to
>persuade - after all they have less potential gain to set against the
>effort to change eating habits.  Do you think that your opinion of all
>patients ability to change is influenced by the frustration you
>experienced treating that specific group?

    Probably.   However, I will say in my defense that doctors treating
younger people don't seem much happier about compliance.

>>    Yeah?  Did you make a substantial difference in your cholesterol
>>levels?  Let's speak truth, now.
>It's difficult for me to tell as yet since I can't persuade my doctor
to do a full cholesterol profile until six months have passed from the
last one (my first ever) - that is not till July.<

   Gah!  British medicine!  Bah!  In the US you'd simply fire that
doctor and get another.

>  A non-fasting cholesterol test last month showed that my total
cholesterol is down from 7.7 (risk factor 7.2) to 6.3. (UK units, I
don't know what they are called). I'm not sure whether a drop in total
cholesterol is meaningful, without the figures for HDL, LDL and TGs, it
seems to me that its quite possible that the risk factor could rise
despite the drop in the total figure.<

    Possible.  To convert from mmole/l to mg/dl for cholesterol, you
multiply by 39.  So you went from about 310 (I'm doing this in my head)
to 250.   A good drop, but probably not enough.  Especially if you have
coronary disease.  250 is the average for the average person who has an
MI in the US, I know.

>Just confirm that the elements of your prescription above are not in
>priority order and we're in total agreement!  How do you know who is
>especially motivated?

Easy.  I give them a list of high saturated fat and high hydrogenated
fat foods they can't eat anymore.  At all.  This takes 5 minutes.  If
their cholesterols haven't come down substantially in a month, they get
the drugs.  If they've made progress, we try fine-tuning them with the
nutritionist.  If they make it that way, well and good.  If not, they
get drugs.  Very few of my people escape drugs.  Again, for those who
think they've a better way, I'd like to see you demonstrate it.

>  Do you really have time to get to know your
>patients that well?  If so you are very unusual in my experience,
>which I admit is limited to the UK.

    After they've been seeing me regularly for awhile I get to know
them.  Tonight I had pie made by an especially greatful one, actually
(not a heart patient-- she survived bilateral pneumonia).  However,
getting to know my patients has not persuaded me that dietary therapy
is the answer.  If anything, the opposite.

>I wouldn't think of myself as especially motivated. I've ignored advice
>to lose weight for years and think of myself as a person totally without
>willpower when it comes to food.

>>>By the way my doctor told me not to worry at the moment about my high
>>>cholesterol as doing anything about it would mean going on a
>>>"horrible diet I wouldn't like at all." I have angina.

    Okay-- I imagine it might be quite motivating to have angina, or
what you fear might be, and be REFUSED drugs (or testing).  But that's
rather draconian, don't you think?

>>    Your doctor is wrong.  You can lower cholesterol and risk factors
>>quite effectively with drugs alone.  I know that doesn't appeal to our
>>Puritan sense of getting something quite valuable for comparitively
>>little suffering, but it's true.
>I wouldn't mind him being a puritan if he would at least do something
>about my cholesterol. He won't even think about diet advice or drugs
>until I've had an angiogram because he doesn't believe the results of my
>exercise treadmill test due to my age. The doctor who did the test said
>it was "conclusive" and I've pointed out that apart from my age and sex
>and the fact that I don't smoke I have every other risk factor for CAD
>that I know about, but he wants to wait and see. I hope he is right, but
>I'll have to wait until at least August to find out.

    Horrible!  Why you Brits put up with this kind of crap, I'll never
know.  Again in the US you'd just fire your doctor (if you were over
65, anyway) and find a better one.  Even if you had an HMO under
insurance, and were younger, you'd have more latitude than THIS.  Any
US  HMO would cath you immediately for a positive stress test.  Or, if
looking at a younger woman, would at least at minimum do a stress-echo
or thalium (i.e., something more specific).

> In the
>meantime I can do the diet changes *despite* the doctors!

    Well, there you go.  But guess what?  I can't use this kind of
motivation ("Do it or have chest pain; I'm on vacation").   Anything I
recommend can't by definition be "despite the doctor."

                                        Steve Harris, M.D.

>Even if the angiogram shows no narrowings, I'm going to carry on eating
>this way, I enjoy it.

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 08 May 1997

In <01bc5b37$fa4fe480$> "Martin
Banschbach" <> writes:

>Marty B:
>The accepted framework Steve is risk of having an MI. Granted it may not
>necessarily be fatal. The Journal of the American Osteopathic Association
>had an editorial in the Jan. 1996 issue covering what approaches a
>physician can use with their patient populations that have existing heart
>pathology and have already had at least one MI. This is the group that
>should be motivated to try something to stop subsequent heart attacks
>(one of which may finally get them). The essence was that DO's were
>remiss in not trying diet first. Here is a breakdown of what physicians
>in the US (both MD's and DO's) are using for these high risk patients
>versus the effectiveness of the treatment in preventing a repeat MI.
>These are all non-surgical options.
>Treatment		Use		% Reduction in MI risk
>Diet Change		20%		55

    This journal should have had a citation for their number in this
case.  What is it?  A rough relation between change in cholesterol
during treatment and change in risk of MI, is about 10% reduction in
cholesterol for every 12-13% in MI mortality, and 8-10% in total
mortality.  See Circulation 91 2274-2282, 1995 for the biggest meta
analysis of all this data that has ever been done.  (Epidemiologically,
BTW, the relationship is 2% risk for every 1% cholesterol change, but
treatment after one MI, aka secondary prevention, doesn't do this well,
naturally).  A 55% change in secondary prevention MI mortality
translates to about 40% cholesterol reduction, which when starting at a
cholesterol of 250, 100 points or so.  That's a HUGE dietary
modification, from 250 to 150 mg/dl.   You need the Pritikin or Ornish
regime to do that.  Before statin type drugs became available, typical
cholesterol lowering with diet plus drugs was about 5-15%, with cardiac
events reduced 22% (see NEJM 323:1112, 1990).  Diet, unless really
incredibly stringent, simply does not compare with statin drug effects.

>The diet does not have to be the Ornish diet, we now have the step 1 and
>step 2 diets.

    You may have them, but you're not going to get a 40% cholesterol
change on them.  To imply this is totally bonkers, and means you have no
idea what on Earth you are talking about.  Sorry, but there's no polite
way to put it.

> The point is that physicians have to be made to consider
>diet changes before they consider drug use (and they give better care to
>the patient when more than one approach is used).

   No, physicians don't have to be made to do anything.  They only have
to be presented with honest arguments.  Which you haven't done.  Prove
me wrong and present your citations to primary literature.

>Marty B:
>All you have to do is focus on the risk of having another MI.  You already
>had one and you didn't like it did you?  The best way to decrease the
>chance that you will have another is to change your diet.

    Untrue.  I don't care how many times you say it, and how many times
you quote unreferenced Osteopathic editorials.

>Steve Harris:
>>    Do you want me to lie for the sake of YOUR religious convictions?
>> Maybe YOU would, but why should *I*?
>Marty B:
>Unfortunately, it is almost like a religion. I've been saved and I want
>you to be saved too.

    ROFL.  Your cholesterol dropped by 40% on diet alone?  Present your
numbers, Bud.

>I think that part of the Ornish movement is the very radical nature of
>the diet (an extremely strict vegan diet). If the animal rights group
>can't shove the vegan diet down our throat using ethics, they will use
>health claims. What most people loose sight of is that step 1 and step 2
>diets are not very extreme and yet they do give significant improvement
>in quality of life (decreased risk of having another MI which could be

    But not by much

> but they are not as effective as the Ornish diet.

    We presume.  But we don't really know, because we don't really know
how effective the Ornish diet is at preventing MI.   All we can do is

                                                Steve Harris, M.D.

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 08 May 1997

In <> (Steve Chambers)

>Risk factors, yes.  But I wasn't aware that there was good evidence
>to show a consequent reduction in _mortality_ for cholesterol-
>lowering drugs.

    You are quite wrong.  See, for example, the 4S (Scandanavian
Simvastatin Survival Study) Trial with 4444 people with coronary
disease and with cholesterols between 220 and 300, were all given
dietary treatment, and then further divided more or less equally into
drug treatment vs non drug treatment groups.  Addition of Simvastatin
to diet lowered cholesterols by 25%, MI mortality by 42%, total cardiac
mortality by 35%, and total mortality by 30% (Lancet 345:1274-1275,
1995).   NOTHING like this has ever been done with dietary therapy.

>>The case for dietary intervention is much stronger.

    Cite the study.

>If whole Asian societies can maintain dietary habits that reduce
>heart-disease risk then I find it hard to believe that we in the
>West can't too.  But I guess what you're saying is that the
>doctor-patient "pit-face" is not the best place to promote that

   Yep.  In Asia they eat that stuff cause they have no choice.  Their
experience does not apply to medical treatment in the West.

                                         Steve Harris, M.D.

From: B. Harris)
Subject: Re: Lower Your Cholesterol Risk Factor
Date: 08 May 1997

In <5krdim$> (John Jamieson)

>My doctor had me cut way back on fat, meat, and cut out eggs.
>I had my cholesterol checked 3 months later and it had only gone down
>5 points with that major diet change. Not very encourageing. I was only
>eating a fraction of the meat, eggs and fat that I had been eating. If
>the changes I made only got me 5 points, meals would become a terrible
>experience if I tried to bring it down another 30 points with diet.

    Typical experience.

>Now I'm on Pravachol 40mg a day. My cholesterol is 204 and LDL is 146.
>I didn't go back to my old ways. I have cereal for breakfast don't eat
>cheese or snack on Ice Cream as I did before I retired. I do eat a bit
>more very lean meat than I did when I was trying to control cholesterol
>with diet.

> It would be nice to know what conditiom
>my arteries are actually in instead of just having those cholesterol
>numbers that indicate a problem. I don't like taking drugs but as I see
>it I don't have much of a choice.
>           John

    See if your doc will order you a stress-echo test.  That will tell
you at least if you have any significant blockage.  You can also get an
ultrasound of your carotids, which will give you a least *some*
indication of how much placque your body likes to form (although the
correlation between neck and heart isn't 100%).

   The only thing to remember with a carotid scan, is not to let them
operate on you no matter what it finds, so long as you don't have
symptoms.  Just say no <g>.  You're only using this as an indirect
"look" at your heart arteries.

    A few places have "fast" CT scanners which can see calcified
placques in coronary arteries.  This can also be useful.

    If you have a positive stress test, or even a positive CT scan, you
might want to have coronary catheterization.  It's a tough decission,
however, and involves more than a little philosophy.  How important is
it to you to know exactly what's going on?  That's an intangible.

                                           Steve Harris,  M.D.

From: B. Harris)
Subject: Re: anticholestrol drug (was: Allopathy (was: I need help))
Date: 11 Jun 1997
Newsgroups: alt.drugs,talk.politics.drugs,,,

In <> Jerry Mings/Rich Balding
<> writes:
>Steven B. Harris wrote:
>>     There aren't too many side effects worse than death.  If an
>> anticholestrol drug keeps people from dying, why shouldn't I use it?
>>                                         Steve Harris, M.D.
>But does it?  It was my understanding that anticholestrol drugs
>lowered the numbers (don't we all love numbers?), but had
>little effect on mortality.

   Nope.  Both the 4S study (Simvastatin) and the West of Scotland
study (Pravastatin) saw about 30% decrease in total mortality by
treating cholesterol with a statin drug.  These were both monster
studies, with thousands of people followed over 5 years.  Double blind,
randomized, placebo controlled.  And there was no increase in non-heart
related mortality in the drug vs placebo group.  No increase in cancer
or suicide, specifically.

   These studies finally put the last nail in the coffin of the
argument that cholesterol per se (or at least LDL cholesterol) isn't
important pathogenically in heart disease.

                                     Steve Harris, M.D.

From: B. Harris)
Subject: Re: Girls' Body Fat Isn't Related to Cholesterol Levels, Study Finds
Date: 19 Jun 1997

In <5o77ls$b1h$> (Archimedes Plutonium) writes:

>In New York Times , 17JUN97, page C5 it states: " The findings
>contradict the conventional wisdom about obesity and cholesterol, said
>Dr. Darwin Labarthe.. in the American Heart Association's journal
>Circulation. In girls, the proportion of body weight that is fat is
>unrelated to cholesterol level ....The cholesterol level for both boys
>and girls fell as they got older and gained weight.... There is a
>difference in boys and girls that has not been appreciated before... He
>said the next step was to see how body fat and cholesterol were linked,
>and learn the role of other factors like diet, sex hormones and physical
>activity. "
>  So pitifully little is known about cholesterol, yet the world pops
>cholesterol pills to the tune of billions of dollars.

   The above study is on adolescents, about which comparitively little
is known when it comes to cholesterol.  But the "worlds" people popping
cholesterol pills to prevent heart attacks are not teenagers.  In
middle aged adults, obesity and cholesterol are related.  And it is
middle aged and elderly adults who are dying of heart disease, by and

>  Perhaps someday the cholesterol pill lawsuits will be on the same
>scale as the present day tobacco lawsuits. Only in the case of smoking,
>it did give their owners some satisfaction. Cholesterol pills, as I am
>told by those who take them, seen to do or feel nothing. As if you do not
>feel any different.

    Why should you?

>  What if researchers find that cholesterol is vital? And that there
is no "bad cholesterol". And what if they find that the cholesterol
pills cause cancer or some other side effects.<

    What if it is discovered that the far side of the moon is made of
green cheese?

    There is evidence from clinical trials involving nearly 10,000
people that lowering cholesterol with pills (statin drugs) cuts heart
attacks by 40% and total deaths by 30% (see Lancet 344: 1383-1389, 1994
for the simvastatin trial, and N Engl J Med  333:1301-1307, 1995 for
the pravastatin prevention trial).  In addition, careful study of these
trials has found no evidence of increased cancer deaths or increases in
any other kinds of deaths (see Archives of Internal Medicine
156:2085-2092, 1996).  These drugs save lives, and there is no evidence
at this time that they are very dangerous.   There certainly is no
evidence that taking them is less safe than not taking them, for people
with elevated cholesterol.  All evidence at this time points to the

    How long are you going to keep spamming the nets with your
uninformed blather, Archimedes?  Do some reading and get back to us.

                                        Steve Harris, M.D.

From: B. Harris)
Subject: Re: Allopathy (was: I need help)
Date: 23 Jun 1997

In <>
(Talena90) writes:

>What study would you like to see Steve? The one that documents the side
>effects of a prescription for lowering cholesterol? Or would you like to
>see the results of my husband's blood tests documenting the change in his
>cholesterol count?

    Not interested.  You can lower your cholesterol from 250 to 150 in
two weeks by the simple expedient of not eating.   There are all kinds
of other things that will do it, from semi-vegetarian (Pritikin/Ornish
style) very low fat diets, to lots and LOTS of soluble fiber (Rexall
sells a pretty good product, but trying getting anybody to stay on it
for months and years).  Problem is, most people will not DO any of
this.  The ones that do (a minority) are all the time blaming the
doctors for the ones that won't, instead of putting the blame where it
belongs.  If we all ate like Bantus and stayed at ideal body weight, we
wouldn't need anticholesterol drugs (except for a very few people with
especially bad genetics, perhaps).  I certainly know that.  The problem
is, we don't eat that way.  My patients don't.  They don't even if I
tell them to.

From: B. Harris)
Subject: Re: medicine men - do what they preach?
Date: Thu, 14 Aug 1997

In <5sua2v$v99$> (George
Conklin) writes:

>   When you are speaking of probabilities, as in
>'cholesterol causes heart attacks,' you are speaking only of
>predicting 1 heart attack in 4 at best, since 3/4 do not
>have known risk factors.

    That is not true, unless you are dumb about where to set risk
factors.   But risk from cholesterol begins to rise after 160 total in
men, 190 in women.   The MEDIAN cholesterol of the average person with
an MI in the US, before the attack happens, is 240  (you can't look at
cholesterol during or shortly after MI, of course, because MI itself
causes levels to do down for a while).

    Again, half of people who are going to have heart attacks next
month in the US have total cholesterols < 240.

> So you cannot classify the 3/4 as
>'aberant' cases.  The real risk factors are unknown,
>although it now looks like homocystine (spelling?) was known
>all along for at least 30 years but suppressed by Harvard
>for political reasons.

   Nonsense.  Some risk factors are unkown but not the main ones.
Nobody in the Framingham study with cholesterol < 150 has died of a
heart attack yet.

                                         Steve Harris, M.D.

From: B. Harris)
Subject: Re: Benefits of Soy Milk
Date: Thu, 02 Oct 1997
Newsgroups: alt.folklore.herbs,,,,

In <01bcce76$3cc90100$5e2e63c3@uxlfdlvk> "Timothy West"
<> writes:

>Joseph Zorzin <> wrote in article
>> Stephen Wolstenholme wrote:
>> >
>> > On Sun, 28 Sep 1997 18:06:58 -1000, Stanward Sueo-Minh Oshiro
>> > <> wrote:
>> >
>> > >What are the advantages and disadvantages of drinking fresh soy-bean
>> > >milk?
>> >
>> > The main advantage is that it is usually OK for people who have
>> > problems with dairy products. The main disadvantage is that it does
>> > not taste as good as dairy milk.
>> The other disadvantage is it's more expensive.
>> But it doesn't have cholesterol like cow milk.
>Is this latter point supposed to be a disadvantage or an advantage?
>Cholesterol is vital is it not?

  Not.  You make all you need.  Too much in the diet raises levels in
your blood, which are not good for your arteries.  Oxidized cholesterol
in your diet is even more directly involved in plugging up arteries.
To minimize milk cholesterol intake, drink it skim.  Or there are soy
milks and even almond milks (which I also like).

                               Steve Harris, M.D.

From: B. Harris)
Subject: Re: JAMA article: how much & what kind of oil?
Date: Mon, 17 Nov 1997

In <> writes:

>One must also remember that a significant percentage (some say up to
>50%) of those dying from a sudden cardiac event have no (repeat NO) risk
>	A Sympathetic MD ;-)>

   I'd be surprised if that were true.  If your homocysteine's over 10
and your cholesterol's over 160, those are risk factors.  Show me a
study where half the people with sudden cardiac death have such
numbers, and I'll be impressed.   Instead, I'm guessing you're thinking
of studies where people with cholesterols under 200 are said not to
have a risk factor, and nobody measured homocysteines at all.  We know
better than this from the Framingham studies, though, don't we?

                                       Steve Harris, M.D.

From: B. Harris)
Subject: Re: Cholesterol Causes Heart Disease
Date: 24 Apr 1998 06:03:19 GMT

In <>
(RunnSwim) writes:

>So cholesterol does at least two bad things: first, it is what forms
>the lipid core of the plaque itself.  Second, it seems to play a role
>in the foam cells which may destabilize the plaque and directly
>lead to a heart attack.

    Yep.  Three bad things: high levels of lipids in the blood after a
fatty meal cause plately aggregation, which in turn may trigger off a
thrombosis at a placque site.  If you turn vegetarian, your risk for MI
or thrombotic stroke drops within hours, then over the following day
weeks and years, it drops by other mechanisms as well.  It's never too
late change your diet, and doing it will start to fix up your body
right away.

                                            Steve Harris, M.D.

From: (RunnSwim)
Subject: Re: Cholesterol Causes Heart Disease
Date: 26 Apr 1998 18:38:06 GMT

Michael Sierchio <> objects to commentary
by Harris:

>>But 100% of people with heart disease have cholesterols
over 10 mg/dl.  I'm not just being facetious,  I think
you're stating a fact that isn't as meaningful as you seem
to imply.  I'd wager that 90% of the people with NO heart
disease ALSO have a total cholesterol of over 150.  (Hint -
any teenager, though the HDL/ LDL ratios will usually be

>>Steven,  I appreciate your posts very much,  but I think
you are overstating the case.  I agree that cholesterol is
implicated in heart disease,  but it is not the cause. <<

At the risk of repeating myself...I am going to repeat


Regarding this whole debate, if I could just get everyone to
read three references, I think that a lot of this debate
could be clarified.  These references are:

1. Wald, NJ and Law MR.  Serum cholesterol and ischemic
heart disease Atherosclerosis 118 (suppl): S1-S5, 1995;  2.
Brown, MS and Goldstein, JL. Heart attacks: gone with the
century? Science 272:629,1996;  3. Grundy, SM. Cholesterol
and coronary heart disease: The 21st century.  Arch Intern
Med  157:1177-1184,1997.

I would love to excerpt long passages which would confirm
most of the points that I have been trying to make, but let
me just summarize:

1.  The relationship between serum cholesterol and heart
disease is perfectly log-linear, with correlation
coefficients exceeding 0.95 in multiple studies (r2 around
0.9).  Graphs shown in Wald, et al.  As stated by Wald, et
al: " [linear scales] give the impression of a threshold in
risk (since on a linear scale the dose-response relationship
is characterized by a relatively flat line at low
cholesterol levels with an increasing slope as cholesterol
increases) when there is no evidence of this."... "[with a
log-linear plot] the dose response  relationship is
remarkably linear, indicating that a given absolute
reduction in serum cholesterol is equivalent to a constant
percentage reduction in ischemic heart disease _regardless_
of the starting cholesterol level."

Scott Grundy (one of the most widely respected investigators
in this field and a proponent of a Mediterranean-type
diet...he is definitely _not_ a "low fat radical" like me)
states in his review that:

"an elevated serum ...LDL (low density lipoprotein
cholesterol) constitutes the foremost cause of coronary
atherosclerosis and, hence, of coronary heart disease.
Support for the fundamental role of LDL in atherogenesis
derives from several sources: epidemiology, animal research,
genetic forms of hypercholesterolemia, laboratory studies,
and clinical trials. The case against LDL is overwhelming.
Perhaps the most telling is the occurence of premature CHD
in individuals who have genetic forms of severely elevated
serum LDL concentrations in the absence of all other risk
factors.  Moreover, in several populations in which serum
LDL levels are very low, CHD rates are relatively rare, even
when the prevalence of other CHD risk factors is high."

..."recent advances in our understanding of the causes and
treatment of high serum LDL levels have been nothing less
than dramatic...statins act primarily by enhancing activity
of LDL receptors...Identification of  elevated serum LDL
concentrations as the primary atherogenic factor constitutes
a major therapeutic advance...placing the lowering of LDL
concentrations at the heart of CHD risk reduction creates a
framework on which preventive strategy can be
built....Although [other] susceptibility factors apparently
do _not_ cause CHD in populations that have very  low levels
of LDL, they assume increasing importance when LDL levels
begin  to rise as a result of dietary changes and sedentary
life habits."

This is what mainstream medical scientists think.

Now add to the above information what is now thought to be
the MECHANISM of plaque formation and heart attacks:


I'd strongly suggest checking out the following short

Lippy, P. et al. Current concepts in cardiovascular
physiology: the  role of LDL cholesterol in plaque rupture
and stabilization.  Am J Med 1998:104(2A):14S-18S,1998.

It explains a lot of the misleading stuff in Ravnskov's
Website, as well as shoots more holes in the theory that
cholesterol doesn't cause heart disease.

Turns out that the most dangerous lesions are not the
biggest ones. More than half of all heart attacks occur in
patients with less than 50% stenosis.  Only 15% of all heart
attacks occur in people with so-called "critical" stenosis
(>70%).  Actually, muscle proliferation may actually be a
good thing, as the proliferating muscle cells, though
increasing the stenosis, paradoxically stabilize the lesion,
by putting a thick "cap" between the arterial lumen and the
cholesterol core, which is filled with highly thrombogenic
(clot producing) lipids.

Anthony B suggested that statins really may work by doing
things such as inhibiting smooth muscle proliferation.

Do you want to inhibit smooth muscle proliferation and will
this lead to a short-term decline in heart attacks?

The body is actually pretty good at dealing with problems
inflicted on it by its unwitting owner.  Eat too much fat,
for whatever reason, and it gets deposited on the arterial
walls, in the same way that spilled oil fouls shorelines.
This is bad, because the lipid material causes clots, which
are the proximate cause of most heart attacks.  So cover it
up with a temporary fibrous/collagen "cap" to keep it walled
off from the platelets that want to glom onto it and create
a disasterous clot.  Then hope that it goes away.  But if it
doesn't, then make the job more permanent by growing muscle
cells over it to make a permanent wall.

It may actually be that people who are really good at
walling off their plaque get fewer heart attacks than people
who aren't. And the ability to wall off plaques may be
different in different people, explaining why there is no
consistent association between DEGREE of atherosclerosis and
serum cholesterol level, while there is clearly a consistent
and strong association between heart attacks and deaths from
heart attacks and serum cholesterol level.

Paradoxically, the less dangerous lesions are the large
lesions, covered by a thick, muscle-rich cap.  These are not
prone to rupture.  It is the plaque rupturing which exposes
the arterial lumen to the highly thrombogenic cholesterol
underneath the plaque.

Rather, the most dangerous lesions may well be the much
smaller ("immature") lesions, which are covered only with a
thin fibrous cap.  This cap can rupture easily.
Contributing to the rupture of the fibrous cap may be
interstitial collagenase enzymes, not produced in normal
coronary arteries, but present in abundance in macrophages
and smooth muscle cells which are filled with cholesterol
("foam cells").

This physiology explains why lowering lipids has effects on
heart attacks and death which are disproportionately large
to the degree of "reversal" of established, large coronary
lesions.  Reduce the cholesterol and you probably get an
immediate benefit in the reduction of the foam cells which
are producing the interstitial collagenase which is
disrupting the plaque and exposing the arterial lumen to the
highly thrombogenic cholesterol inside the plaque.  Plus you
get a benefit in reduced formation and "growth" of the early
plaques, covered by only a thin fibrin cap, which are the
more severe problem with regard to producing coronary
thrombosis and a heart attack.

The most immediate benefits of lowering cholesterol through
diet or drugs are direct reduction of platelet aggregability
(as pointed out earlier in this discussion by Harris), then
reduction in the cap-busting foam cells. However, down the
road there will also be decreased cholesterol plaque
formation and finally reduction in the size of pre-existing

The case for the causal role of cholesterol in heart disease
is overwhelming.  It is supported by the clear dose-response
relationship between serum cholesterol and heart disease;
the substantially reduced cardiac and total mortality when
heart patients are put on a severely fat restricted diet;
the dose-response relationship between cholesterol reduction
and heart disease seen in the drug trials; by animal models;
and by an understanding of the most probable mechanisms by
which most heart attacks are produced by coronary artery

When you look at the big, overall picture, everything is
very clear and very consistent.

- Larry Weisenthal

From: (RunnSwim)
Subject: Re: Cholesterol Causes Heart Disease
Date: 28 Apr 1998 16:54:41 GMT

>>Tell where you found this statement " by growing muscle cells over it to
make a permanent wall."<<

This is shown in figure 2 of the Libby article.  Look at the inset which shows
normal medial smooth muscle cells in the outer arterial wall and "activated"
intimal smooth muscle cells (identified by their HLA phenotype) growing
over the plaque.  This figure, was, in turn taken from the following article:

Libby,P. Molecular bases of the acute coronary syndromes. Circulation

The figure and description on pp. 14S and 15S show that the vulnerable
lesions have thin fibrous caps, large lipid accumulations underneath
the caps, and numerous macrophages filled with cholesterol.  These
vulnerable plaques are "frequently depleted of smooth muscle cells,
particularly in the area where rupture has occurred."  In contrast, stable
plaques are covered by a thick layer of smooth muscle cells (i.e. they
are "walled off"), so they may be (paradoxically) a lesser problem for the
host than are smaller lesions which leave a wider coronary artery lumen.

All of this is very relevant to the issue under consideration (i.e. does
cholesterol cause heart disease).

Some participants have complained that it is only a statistical association.
But when you look at the mechanisms and you look at other facts (i.e.
that most heart attacks occur in coronary arteries which are not severely
stenosed, i.e. those without a thick muscle covering, but with only a thin
fibrous cap), the absolutely central role of cholesterol in heart disease
is very evident.  As noted by Grundy, additional key points are the
frightfully high incidence of heart disease in people with a single genetic
mutation which causes high cholesterol, in the absence of other risk
factors and the very low incidence of heart disease in people with very
low cholesterols, even in the presence of multiple other risk factors.

- Larry Weisenthal

From: B. Harris)
Subject: Re: Cholesterol causes heart disease
Date: 29 Apr 1998 12:04:00 GMT

In <6i6sn4$>
(Anthony Brea D.C.) writes:

>If you want to call cholesterol the 'cause' then you can call copper
>deficiency the 'cause' also.

    Not unless it explains essentially the amount of varience that
cholesterol does.  It's possible that cholesterol is not as much of a
cause as the varience it predicts "explains" statistically, but it's
very unlikely that copper (or anything else) is MORE of a cause than
the varience it explains statistically.  I hope that's reasonable clear
to at least some people reading this thread.

   Larry's problem, I think, is that in the past he's pulled out all
these high R values, showing that cholesterol explains >90% of the
varience in heart disease risk.  But that accounts for the slope only,
and not the intercept (the constant of integration).  Which says that
there's a lot of heart disease out there cholesterol levels don't
acount for.  You're all pointing this out to Larry.  He's saying
cholesterol is the best of the factors so far identified, and you're
saying a lot of factors haven't been identified.  You're both right.
However, if you want to suggest that known factors explain heart
disease just as well as cholesterol, you're wrong.

>Central role doesn't equal cause. Copper deficiency plays a central role,
>since copper deficiency leads to the altered lipid patterns in both
>animals and humans.

   NO!  If that were true, copper levels or activity or something about
copper, would have to be known to correlate better with cholesterol and
CAD than they do with each other.   No such evidence exists.  It might
be a theory of yours.  It might one day turn out to be true.  But right
now, there's not adequate statistical support for the magnitude of the
effect you assert here.

                                     Steve Harris, M.D.

From: B. Harris)
Subject: Re: Cholesterol causes heart disease
Date: 5 May 1998 08:40:51 GMT

In <>
(Alf Christophersen) writes:

>Correlation coeff of 0.95 only says that the mathematical model you
>tries to fit the data to explains 95% of the observations.

   Not even that.  It means that the model explains 95% of the varience
in the observations.  The constant of integration (in this case, the
baseline incidence of heart disease at the lowest cholesterol values),
is up as an unexplained constant.  And it's there where the theory
"fails," and you need some more explanatory power.

    When Larry says that the correlation is log linear, he doesn't mean
it's log linear down to zero cholesterol and zero heart disease.  It's
log linear down to cholesterols of 150 or so.  Below which, studies
like Framingham simply don't have enough people (enough numbers) to
validate ANY theory.  In the China Health Study, by contrast where the
cholesterols go from about 90 to 150 or so, and the average is 125,
there's no correlation *at all* between cholesterol and heart disease.
Nor would I expect one.  Probably even Larry doesn't.  This does not
mean that the Chinese peasants have no heart disease, however.

                                            Steve Harris, M.D.

From: B. Harris)
Subject: Re: Cholesterol causes heart disease
Date: 13 May 1998 23:46:14 GMT

In <> Brian Sandle
<> writes:

>Steven B. Harris <> wrote:
>> In <> Brian Sandle
>> <> writes:
>>>> Very low incidence of heart disease in people fortunate enough to
>>>> have a very low cholesterol, despite the presence of other
>>>> risk factors.
>>>Very fortunate - no bandages, no cuts.
>>    How does having no bandages influence number of cuts?
>I am suggesting that the cholesterol (bandage supply) may be low
>*because* there are no arterial lesions for the body to have had to
>put them on.

   You're having problems with your logic.  It MIGHT happen that way,
but the way you tell which is the dependent variable is then to
artificially lower the cholesterol, which should have no effect on
number of cuts, or damage.  But doing this (with drugs or diet) does
influence athersclerosis and all the bad things (strokes, MIs) which
flow from it.  Now can you understand why cholesterol is considered

>Bandages influence your number of cuts on a graph in which bandages
>are the independent variable and cuts the dependent variable.

   No, you don't just get to pick which is which by drawing a different
graph for the natural association.  You find out which is dependent in
reality (mechanistically) by changing it, and seeing if the other
changes in consequence.

   And, by the way, there are many artificial ways to damage the
insides of arteries.  A favorite model in rabbits is simply to run a
brush up the artery and scrape the inside.  The rabbit will then
develop atheroscelosis at the site.  But doing this does NOT change
serum cholesterol levels.  So your theory about which is the dependent
variable has some problems.

>The question I am asking is whether the cholesterol is causing are being
>caused by the arterial disease. We need a mechanism to say which is
>dependent on which. A correlation does not provide that.

   That is right.  But medicine has based its decision in the matter on
much more than mere correlation.

>And you introduce the word influence whereas I am only talking about
>the connection of the two points on a graph. The word influence cannot
>safely be used in that connection since it often means an actual force
>for change.

   It can safely be used if you have experiments showing that you get
an actual force for change, when you change cholesterol.

                                       Steve Harris, M.D.

From: "Steve Harris" <>
Subject: Re: Eggs again
Date: Mon, 23 Apr 2001 23:12:42 -0600

<> wrote in message
> Approximately one-third of all first-time heart attacks occur to
> people with 'normal' cholesterol, and no history of high cholesterol.

Which only goes to show that "normal" cholesterol in some populations is too

Assuming reasonable HDL numbers, heart disease risk does not change until
total cholesterol is 160-180, then begins rising slowly, then more rapidly
over cholesterols of 220.  Populations with average cholesterols of 150 (and
there are some) have almost no coronary atherosclerosis.

From: "Steve Harris" <>
Subject: Re: vegatarianism/ is meat important?
Date: Mon, 23 Apr 2001 13:34:08 -0600

"anonymous" <a@b.c> wrote in message

> If your cholesterol improved by abandoning a vegan diet, I don't think
> you know what a real vegan diet is.  It is not possible to have
> cholesterol problems from a vegan diet since such a diet could not
> possibly include ANY sources of cholesterol.

Your body makes its own cholesterol, doofus. And it's quite possible for you
to have bad enough genes to make too much, even if you don't eat any at all.

From: "Steve Harris" <>
Subject: Re: Help us!
Date: Sun, 6 May 2001 02:49:57 -0600

"Gault" <> wrote in message

> You're comparing 1951 medicine to today? You must be kidding.  Since I'm
> betting you weren't around then I urge you to take a look at medical
> books from the period and compare them to today's.  Or compare the
> available drugs from then with what we have today.  Modern medicine is
> *much* better than 50 years ago.  It certainly has its problems but
> comparing it to 50 years ago is ridiculous.

Hell, yes. 50 years ago there were about 3 antibiotics, maybe one drug for
high blood pressure, and no drugs at all for high cholesterol. Nobody was
even sure if cholesterol was important, even. There were big arguments about
whether or not smoking was bad.

Medicine improves over time. Really.


From: "Steve Harris" <>
Subject: Re: Help us!
Date: Sun, 6 May 2001 18:41:08 -0600

<> wrote in message
> "Steve Harris" <> wrote:
> >Nobody was
> >even sure if cholesterol was important, even.
> There are some that now think cholesterol is important, but there is
> not a clear proof of that.  I believe it is a Red Herring.

What clear proof would satisfy you? Reducing LDL cholesterol with a number
of drugs has decreased death in a number of studies. That used to be the
gold standard demanded by the critics, until we got it. Then they said
repeat it. Then they said they still didn't believe it.

So what do you want?

From: "Steve Harris" <>
Subject: Re: Humans as Omnivores (Was: vegetarian friend ate beef. -- lost  
Date: Mon, 14 May 2001 12:48:26 -0600

"Lyle McDonald" <> wrote in message
> An amusing piece of data in this regards is the plot of all mortality
> deaths vs. cholesterol.  While you see a definite increase with
> increasing cholesterol above a certain point (deaths occurring from
> CHD/stroke/etc, the stuff associated with high cholesterol), after the
> lowest point in the curve, all death mortality goes up again when
> cholesterol levels get *too* low.
> Even more amusingly, the deaths typically occurring at the lower
> cholesterol levels are weird deaths: folks stepping in front of buses,
> that sort of thing.
> Almost as if their mental capacities had been impaired by the extremely
> low cholesterol levels.
> Hmm.
> Lyle

If you look at the China Health Study, a massive study done 10 years ago of
a very large semi-vegan population in which the *average* total cholesterol
is only 150 mg/dL, and there is data down to cholesterols of 90 mg/dl or so,
you see something interesting.

First, you see that at these ranges cholesterol levels no longer correlates
to coronary heart disease deaths.

Second, the very lowest cholesterols are indeed correlated with different
causes of death, but these tend to be infectious in nature: pneumonia and
septic shock of various kinds. This is probably NOT due to low cholesterols
being markers for various other diseases, since the population of very low
cholesterol level people is such a large faction of the total population in
this study (like 20%).

One is left with the conclusion that very low LDL levels may be associated
with infectious deaths. They either are a marker for general malnutrition,
or else (there is some independent evidence for this) LDL is protective
against endotoxin mediated effects. It may even absorb endotoxin. I note
that dietary restriction studies have universally been done under
semi-barrier conditions in which the animals are not exposed to a normal
range of pathogens, and are not forced to heal from wounds, etc.

Of course, in a society where there is good medical care and good access to
rapid antibiotic treatment, these results may be of less interest, and
attempting to lower LDL "as much as possible" may still be the way to go for
the average middle aged person.


From: "Steve Harris" <>
Subject: Re: excessive dietary protein
Date: Fri, 8 Jun 2001 02:03:40 -0600

"Kelolo" <> wrote in message

> Steve, what do you consider to be the best studies that support the theory
> that saturated fats and/or cholesterol cause heart disease?

There are too many to count. If you want my personal epidemiologic favorite,
it would probably be the Seven Countries Study. I never really got over
those nice graphs where the line representing heart disease deaths was
nearly a linear function of saturated fat % cal intake in those countries.
As was LDL level.  This went from Japan where they had hardly any heart
disease, a 5% sat fat intake, and average cholesterol of 150. At the other
end places like Finland and Scotland with 20% sat fat, average cholesterol
250, and the highest coronary artery disease in the world.

>Are you a fan of Ancel Keys?  I haven't read enough to decide about him

Original old dietary restriction guy-- did some of the restricted diet
studies on consciences objectors in the late 1940's. In some ways a
precursor of one of my own mentors, Roy Walford.

Yeah, all this stuff is true. Energy and sat fat and trans fat restriction
is good for you. To some extent you can eat higher protein diets than we
thought, and also mono-unsat fat, if you energy restrict.

I've seen the low fat energy restriction stuff work on lipids and other risk
factors, myself. If you look at that first paper from Biosphere II, I'm
second author.


From: "Steve Harris" <>
Subject: Re: excessive dietary protein
Date: Sat, 9 Jun 2001 20:41:11 -0600

From: "Quentin Grady" <>
Subject: Re: excessive dietary protein
Date: Friday, June 08, 2001 7:41 PM
>G'day G'day Steve,
>  In this respect you are like most of us impressed by linear
>relationships.  It suggests dose dependence.  Certainly the Seven
>Countries Study impressed me.  The scientific methodology didn't
>impress others who thought the statistical methods were shoddy and
>so selective as to be self serving.  Darn.  It was a bit like
>Mendel's pea experiments which subsequent analysis has sbowed to
>have been too perfect.  Of course it didn't mean Mendel had it
>wrong in some fundamental way, just that there is a temptation
>among researchers to gild the lily a trifle.

COMMENT: The "Seven Countries" results have held for decades for the seven
countries they looked at, so I hardly think you can accuse them of fudging
the stats (if you have any specific problems you should mention them, and
also send a letter to Circulation). If a look at 40 countries shows a few
outliers from the Seven Country trend, that doesn't mean that saturated fat
isn't (to first order) the bad stuff we thought it was. There aren't any low
fat countries with lots of heart disease. The outliers are a few high fat
countries with less than we thought they should, including the French.
Probably something protective in the diet, whether a lot of folate
(homocysteine lowerer), good things in red wine, or perhaps saturated fat
from sources other than dairy aren't quite as bad (dairy fat certainly kills
the Scandinavians).

>What I find strange is that so many of us bought the Seven Countries
>Study yet almost everyone blasted the study of dairy protein types
>and CHD performed by a New Zealand scientist. That was one heck of a
>straight line correlation he presented. He had more data points and
>didn't have the anomalies presented by the French who happily
>consumed more saturated fats than were supposed to be healthy.

The French are rare exceptions.  Mostly, however, the association between
sat fat and heart disease continues to hold. And of course it's been well
validated in animal models, we have a nice mechanism (LDLs), and good
intervention trials, so there's not much chance we're fooled by some
confounding effect.  Rather, the French result simply means that
atherosclerosis is more complicated than we thought. Rather like those few
people who smoke all their lives and don't get lung disease. They don't
prove a thing except that cigarettes are not sufficient cause in every case.

Circulation 1993 Dec;88(6):2771-9

Comment in: Circulation. 1994 Dec;90(6):3118-9

Differences in coronary mortality can be explained by differences in
cholesterol and saturated fat intakes in 40 countries but not in France and
Finland. A paradox.

Artaud-Wild SM, Connor SL, Sexton G, Connor WE.

Department of Medicine, Oregon Health Sciences University, Portland

BACKGROUND. For decades, the coronary heart disease (CHD) mortality
rate has been four or more times higher in Finland than in France
despite comparable intakes of dietary cholesterol and saturated fat.
A potential answer to this paradox is provided by this study of 40
countries and the analyses of other nutrients in the diets besides
cholesterol and saturated fat. METHODS AND RESULTS. CHD death rates
for men aged 55 to 64 years were derived from the World Health
Organization annual vital statistics. Dietary intakes were gathered
from the Food and Agriculture Organization of the United Nations
database.  Forty countries at various levels of economic development
and 40 dietary variables were investigated, including a lipid score
that combined the intakes of cholesterol and saturated fat
(Cholesterol-Saturated Fat Index [CSI]). The CSI was significantly
and positively related to CHD mortality in the 40 countries. The
countries with low CSIs had low CHD death rates.  Countries with high
CSIs had a wide range of CHD death rates. France, Finland, and other
Western industrialized countries had similar CSIs. After adjusting
for cholesterol and saturated fat, milk and many components of milk
(butterfat, milk protein, calcium from milk, and riboflavin) and
total calcium remained positively related to CHD mortality for all 40
countries. There were differences in the consumption of these foods
and nutrients in France and Finland. Milk and butterfat (fat from
milk, cream, cheese, and butter) consumption was higher in Finland
than in France. The consumption of plant foods, recently shown to be
protective against CHD (vegetables and vegetable oils containing
monounsaturated and polyunsaturated fatty acids), was greater in
France than in Finland. CONCLUSIONS. Over the years, France and
Finland, with similar intakes of cholesterol and saturated fat,
consistently have had very different CHD mortality rates. This
paradox may be explained as follows.  Given a high intake of
cholesterol and saturated fat, the country in which people also
consume more plant foods, including small amounts of liquid vegetable
oils, and more vegetables (more antioxidants) had lower rates of CHD
mortality. On the other hand, milk and butterfat were associated with
increased CHD mortality, possibly through their effects on thrombosis
as well as on atherosclerosis.

PMID: 8252690 [PubMed - indexed for MEDLINE]

From: "Steve Harris" <>
Subject: Re: Dangers of low cholesterol
Date: Fri, 3 Aug 2001 19:55:43 -0600

"Justin Bond" <> wrote in message

> Yes, the same thing goes with heart disease. In people that do not
> suffer from the genetic defect Familial Hypercholesteremia, its a
> better bet that heart disease is the result of chronic inflamation of
> the arteries, not cholesterol buildup. This is why most heart attacks
> happen to people with normal cholesterol levels.

No, most heart attacks only happen to people with normal cholesterol levels,
because "normal" in that case has been defined as "usual and average" in
countries where it's usual and average to die of a heart attack. You see the

The average cholesterol level of the average person having a first MI in the
USA is 240 mg/dL.

From: "Steve Harris" <>
Subject: Re: Dangers of low cholesterol
Date: Mon, 6 Aug 2001 20:27:26 -0600

"Justin Bond" <> wrote in message
> Ancil Keys, in a comment on the data noted "over most of the range of
> serum cholesterol...there was little or no relation between serum
> cholesterol and total mortality" (Keys, A., Letter to the Editor. The
> Lancet, 1987, 812). The comment is wrt total mortality, but CHD makes
> up the lion's share of the high cholesterol deaths, and the increased
> deaths at low cholesterol levels does not manifest until quite low -
> 160-180ish on Stammlers data. Hence the term "U-shaped" and not
> "V-shaped".

You might check out the China Health Study, which is the best epidemiologic
probe I know if into the ultralow cholesterol region. It's a truly gigantic
study of some hundreds of thousands of Chinese, and the *average* serum
cholesterol in the study is only 125 mg/dL.  Over their range, which
extended 40 mg/dL on either side of this, they found no correlation of
cholesterol to heart disease OR mortality. This is not suprising with CHD,
since almost the entire population was running cholesterols too low to begin
to break upward due to that being a risk factor (this happens at roughly 160
mg/dL for men and 180 mg/dL for women).

Interestingly, they didn't find any U-shape, either. Part of that was that
they threw out early deaths so as not to encounter the well known "occult
cancer" cholesterol drop. Interestingly, the CAUSE of mortality did change
somewhat over this low cholesterol range. At really low cholesterol levels,
deaths from infections become more common, especially pneumonia. I don't
know if this remained true when smoking is controlled for. However, there is
some modest evidence that LDL may absorb endotoxin an protect from septic
shock, so studies which find an excess of infectious deaths at really,
really low cholesterol levels, are especially interesting.

From: "Steve Harris" <>
Subject: Re: Dietary cholesterol and serum cholesterol: effect of baseline 
Date: Thu, 6 Sep 2001 13:57:17 -0600

"Jay Tanzman" <> wrote in message
> Now, to answer your question, using the above model, if one were to
> _remove_ dietary cholesterol, the magnitude of the effect on serum
> cholesterol would be dependent on the _final_ level of cholesterol
> intake, the level after the change was made.  That is, the largest
> changes in serum cholesterol would be in those who dropped their
> cholesterol intake _to_ the lowest level.  In other words, change the
> headings of the column above to "Final Dietary Cholesterol", "Decrease
> in Dietary Cholesterol," and "Decrease in Serum Cholesterol," and you've
> got your answer.
> -Jay

Yep, the biggest difference is going from semi-vegan to vegan.

There's a famous experiment where somebody or other fed 2 eggs a day to
semi-vegetarian hunter-gatherers and showed that their cholesterol blood
levels went up markedly. But two eggs a day on top of the average American
diet doesn't show up at all. That's probably one of the reasons why egg
consumption isn't a risk factor for CAD in most studies. It might be, if you
were vegan and was wondering whether or not to add them. Otherwise, forget


From: "Steve Harris" <>
Subject: Re: How many times should you check your cholesterol?
Date: Tue, 12 Mar 2002 18:35:10 -0700
Message-ID: <a6mag5$2gq$>

> "wuzzy" <> wrote in message
> > There are experts I think on this topic in this newsgroup:
> >
> >
> > How many times should you have your blood cholesterol tested
> > before you know you  have high cholesterol?

ANSWER: Twice, so long as you've fasted at least 12 hours each time, and no

From: "Steve Harris" <>
Subject: Isocaloric carbo for sat fat substitution (Re: sorry)
Date: Sat, 9 Jun 2001 19:04:12 -0600

>From: "Martin E. Lewitt" <>
>Subject: Re: sorry
>Date: Friday, June 08, 2001 10:09 AM
>I think Steve's points were (correct me Steve if I am wrong):
>1) saturated fat in diets are particularly bad in some way
>>I think he failed to make this point.  This would have to be
>>done on an isoCaloric basis, I think the reduced Calorie
>>diets might have been just as successful if not more, if sat fat >were
>substituted isoCalorically in for the carbs, HDL would have
>>been increased and TGs decreased, dramtically if the substitution
>>got into the low carb regime.

My point was that saturated fat raises cholesterol and if you replace it
with carbohyrates, cholesterol goes down. That's really what even isocaloric
studies show.

Not enough is known about extremely low carb diets to say if the
relationship holds, but certainly over a wide range around the normal
dietary intake of sat fat, you do well to replace calorie by calorie with
ANYTHING else, except maybe bad stuff like sucrose or hydrogenates. But a
straight substitution with carbohydrates is fine.

>>2) low fat reduced calorie diets can improve lipid risk factors
>>I think he succeeds here.  However, high fat reduced Calorie diets
>>might be better.

Nobody knows. The caloric effect is large enough to drown everything else
out, and nobody has isocalorically compared low-energy diets for content.

The following is the best meta-analysis I can find of isocaloric fat vs
carbohydrate replacement on coronary risk factors. It looks at 395 separate
metabolic ward trials. Metabolic ward means these people jolly well eat what
the researchers think they're eating, because they are stuck in a place they
cannot get extra food.  This is high quality data. It will be a long time
before we can review this many trials separately in this newsgroup.

The bottom line: isocaloric replacement of saturated fat by carbohydrate in
the average British diet results in both total cholesterol and LDL
reduction. Since 69% of the total reduction is reduction in LDL, it seems
very likely that lipid ratio remains roughly unchanged for most people who
replace saturated fat with carbohydrate. In any case, the countries of the
world with low incidence of heart disease, don't have low incidence because
they have particularly better total/HDL lipid ratios than the others.
Rather, they have average ratios (4:1) and their low CAD risk is due to
their low total cholesterols (< 160-180 mg/dL) and low LDLs. Only in
affluent countries with high incidence of heart disease does HDL/LDL ratio
become a better a-priori predictor of CAD risk than absolute LDL levels, and
this does not hold over into treatment changes. The reason is that some of
the ratio risk-difference is due to the difference in absolute HDL levels
between men and women, and is not addressable by diet (maybe by castration,
but not diet). It's irrelevant for our purposes. Treatment recommendations
in both coronary disease and diabetes target absolute LDL levels, not lipid
ratios, for a reason.

Overall conclusion is that a low-sat-fat low-cholesterol diet will reduce
British average cholesterols by 31 mg/dL, with 80% of the drop being LDL.
This is stated to be `10-15%' so you can infer that the average British
cholesterol is about 31.2/12.5% = 250.  Not good. But in agreement with
other epidemiologic studies of the UK.

Conclusion: a fairly prudent AHA-type recommended diet change would decrease
Brit cholesterol from 250 to 220 mg/dL (multiply mmole/L by 39 to get
mg/dL). A total cholesterol of 250 is about what the average never-treated
patient with the average first M.I. in the U.S. has.  Under 60% dietary
saturated fat and cholesterol restriction, level of HDL might go from (at a
guess, assuming risk ratio is an average 5) 50 to 44. Lipid ratio then goes
from 250/50 to 220/44 = 5 (unchanged). Still, and overall, a reduction in
coronary risk of 15% at least.  Probably not enough, but better than a poke
in the eye.
However inadequate, I know of no other isocaloric substitution changes which
show effects this large.

At any rate, these conclusions from nearly four hundred trials are enough to
lay to rest the idea that carbohydrate is worse than saturated fat when it
comes to coronary risk for the average person who will be having a heart
attack in your average Western country.

BMJ 1997 Jan 11;314(7074):112-7

Comment in:
 BMJ. 2000 May 27;320(7247):1469; discussion 1470

Dietary lipids and blood cholesterol: quantitative meta-analysis of
metabolic ward studies.

Clarke R, Frost C, Collins R, Appleby P, Peto R.

Clinical Trial Service Unit, Radcliffe Infirmary, Oxford.

OBJECTIVE: To determine the quantitative importance of dietary fatty
acids and dietary cholesterol to blood concentrations of total, low
density lipoprotein, and high density lipoprotein cholesterol. DESIGN:
Meta-analysis of metabolic ward studies of solid food diets in healthy
volunteers.  SUBJECTS: 395 dietary experiments (median duration 1 month)
among 129 groups of individuals. RESULTS:  Isocaloric replacement of
saturated fats by complex carbohydrates for 10% of dietary calories
resulted in blood total cholesterol falling by 0.52 (SE 0.03) mmol/l and
low density lipoprotein cholesterol falling by 0.36 (0.05) mmol/l.
Isocaloric replacement of complex carbohydrates by polyunsaturated fats
for 5% of dietary calories resulted in total cholesterol falling by a
further 0.13 (0.02) mmol/l and low density lipoprotein cholesterol
falling by 0.11 (0.02) mmol/l. Similar replacement of carbohydrates by
monounsaturated fats produced no significant effect on total or low
density lipoprotein cholesterol. Avoiding 200 mg/day dietary cholesterol
further decreased blood total cholesterol by 0.13 (0.02) mmol/l and low
density lipoprotein cholesterol by 0.10 (0.02) mmol/l.  CONCLUSIONS: In
typical British diets replacing 60% of saturated fats by other fats and
avoiding 60% of dietary cholesterol would reduce blood total cholesterol
by about 0.8 mmol/l (that is, by 10-15%), with four fifths of this
reduction being in low density lipoprotein cholesterol.

Publication Types:

PMID: 9006469 [PubMed - indexed for MEDLINE]

From: "Steve Harris" <>
Subject: Re: HDL and Triglycerides
Date: Mon, 3 Jun 2002 17:30:32 -0600
Message-ID: <adguc2$dq5$>

"wuzzy" <> wrote in message

> anyway most of us here have looked at data, some here have even looked
> raw data, and there is no question that cholesterol and triglycerides
> are markers of obesity, sedentary lifestyle and CHD,...  he would
> defend against my statement by saying "A high cholesterol is not
> dangerous by itself, but may reflect an unhealthy condition, or it may
> be totally innocent", like triglycerides and cholesterol are all
> markers of the metabolic syndrome, but which of them are the cause of
> heart attack..


Yeah, this is just like Duesberg saying that HIV infection is harmless, but
merely a marker for the true causes of AIDS, which are multiple drugs and
toxins (he's up to about 20 of these, including all antivirals and
antibiotics to have the same post-hoc explanatory power).

The problem is that such hypotheses are epidemiologically testable. Anybody
who offers LDL as merely a confounder or marker for something better "X",
either has a candidate for that something better "X" or not. If LDL is
simply a confounder and marker for true cause "X", then it's quite easy to
control for X in a multiple regression analysis. To say the least "X", if
merely a marker for harmless LDL, will have to correlate with both LDL and
heart disease better than they correlate with each other. I know of no such
identified "X".  For example, we know that LDL is not just marker for simple
things like obesity or insulin resistance, because we can control for these
statistically (ie, look at a lot of people of the same body mass index and
insulin response who have different LDLs and so on).

The alternative is the last refuge of the fanatical crank skeptic, because
it's a truly pussy hypothesis: the idea that LDL is a marker for a truly
harmful X that hasn't been found yet. That keeps the statisticians from
identifying LDL statistically as a confounder. AIDS was in much this same
position before HIV had been identified-- statisticians recognized that
things like a drug-use and hemophilia factor use might be confounders for
the *truly* harmful agent-- but without an HIV test, they couldn't
statistically tell. When an HIV test became available, they could
immediately segregate HIV-pos from HIV-neg, and thus control for the other
variables. Largely, they found that once they did this, the other variables
dropped out-- THEY were the confounders, not the HIV.

So what do we do if LDL is a marker for "X" that hasn't been discovered yet?
It's hard to refute this with purely retrospective studies. But you can do
prospective intervention studies aimed at LDL. This is much the same thing
as getting people to quit smoking to see what happens to lung cancer risk,
or giving anti-HIV drugs to see what happens to the symptoms of AIDS.

The answer from prospective studies in animal models of atherosclerosis, and
in randomized control studies of intervention into LDL levels in humans, is
that all methods of lowering LDL have shown evidence of reducing
atherosclerosis-- though some are better than others. The simplest studies
are in animals where atherosclerosis progression can be assessed directly.
Both dietary and drug manipulation of LDL are effective here.

In humans, the answer is the same. We have positive interventive studies for
heart disease in dietary manipulation of LDL by polyunsaturated fat feeding,
low fat diet feeding, and for 3 classes of drugs (fibrates, nicotinic acid,
and statins). Of these, only the statins have proven to extend life, largely
because the others have negative side effects (GI cancer for fibrates and
polyunsaturates), or else they haven't been studied well enough (niacin and
low-fat). None of the side effect stuff is relevant, however. If any of 5
different ways to lower LDL all result in less atherosclerosis, then either
LDL is primarily causal, or else it's so closely tied to the true causal
factor that no attempt to separate it has (or possibly can) work. In which
case, what harm in just treating LDL?

By the way, there is a second line of evidence that LDL is causal. We have
multiple pathological studies which have elucidated the pathway by which LDL
enters arterial intima, oxidizes, triggers off an immune/inflammatory
response in macrophages, and off you go into the first stages of
atherosclerosis. In animals, you can manipulate and tag LDL, and then  watch
this pathology happen at every stage.


From: Steve Harris <>
Subject: Re: Health Canada warning EZETIMIBE
Date: 5 Feb 2005 13:41:53 -0800
Message-ID: <>

>>Perhaps it is time for the medical community to reconsider that,

if the body needs Cholesterol to make up 2/3 of the brain, and
if the body needs Cholesterol to sheath the nerves, and
if the body needs Cholesterol to make up steroids in the mitochondria
of the
muscle cells, and
if the body needs Cholesterol to repair inflammation-induced damage to

maybe, just maybe, the body needs Cholesterol. <<


The medical community is well aware that the body needs cholesterol.
The medical community is also aware than in countries where the average
total blood cholesterol is 150 or lower (it was something like 90 in
rural China), there is next to no atherosclerosis, and a lot less
stroke. Nor do these populations seem to pay for it by having their
nerves, brain cells, and muscles fall apart. So whatever cholesterol we
need in our blood, it's less than that.

I'd be rather suprised if Zetia proves to have the side effects of the
statins. All it does is return your blood levels of cholesterol to
those that humans have lived with more millions of years.

Now, the next question is where or not this will get rid of most
atherosclerosis. I suspect it won't be nearly as effective as doing it
by calorie restriction and a high polyunsaturated diet (ie, what they
eat in those countries where heart disease is rare). Obesity itself is
an inflammatory state, and some of atherosclerosis is caused by an
inflammatory component, not just high blood cholesterol. Statins are
antiinflammatory, and Zetia isn't. For that matter, the side effects of
statins are not due simply to their lowering of blood cholesterol, but
go in part to mechanisms much deeper.


From: Steve Harris <>
Subject: Re: Health Canada warning EZETIMIBE
Date: 5 Feb 2005 17:54:13 -0800
Message-ID: <>

Nope, that's why we need prospective trials to see if lab numbers
correlate with clinical outcome. Which in the case of Zetia we do not
have. But the doofuses approved it anyway. Despite the fact that, if we
look outside of statin therapy and diet, other methods of lowering
cholesterol by pharmacological means have show a clinical outcome track
record that is either bad or (at best) very very uninteresting. Find me
the evidence that niacin, bile acid binders, or fibrates save lives.
The FDA's been burned before on using proxy markers in approving drugs.
They killed a lot of people with Mexilitine, for example. Nobody
apologized. And apparently nobody learned anything, either. Going by
near-term risk factor modification is a good paradigm for blood
pressure pills and even antidiabetic drugs, but that doesn't
necessarily mean it carries over into all other aspects of medicine.

The reverse is also true, to some extent. If you look at the effects of
beta blocker drugs or fish oil on PVDs in routine monitoring, you might
predict that both would be worthless as preventives in keeping people
from dying of sudden cardiac arrest due to malignant dysrhymias. And in
both cases you'd be wrong. Nothing substitutes for a good long
prospective clinical with real clinical endpoints. Of course, these are
fiendishly expensive. So the problem is the same as with all other
things in society: everybody wants the results of such trials, but
nobody wants to pay for them.

I have in mind Harvard president David Starr Jordan's aphorism: "If you
think education is expensive, try ignorance." To some extent the same
is true of clinical trials. They're expensive, sure. But the trillions
we waste in medicine by NOT doing them.....!


From: Steve Harris <>
Subject: Re: Health Canada warning EZETIMIBE
Date: 5 Feb 2005 16:12:38 -0800
Message-ID: <>

>>A sign at the medical laboratory I go to urges patients to have a CRP
test because in THEIR words "Half the people who have coronary artery
disease have normal or below normal cholesterol"

"Half" sounds like a 50 / 50 crap shoot to me.  According to the sign in
the Laboratory, above normal cholesterol is NOT related to cardio
vascular disease any more than normal or below normal cholesterol.  I
believe this has been verified by other researchers - especially if
people with hypercholestermia (sp?) are culled from the population in the
study. <<


You're going to ignore the best advise of medical science because you
saw a sign on a laboratory somewhere?? Who do you think put up the
sign? Did you also see a glowing finger writing "mene, mene, Tekel...."
in Hebrew?

Normal means "average" and the average differs from country to country.
In places like Finland and Ireland, where they have the highest heart
disease rates in the West, average cholesterol is about 260 mg/dl. In
the US, it's about 230 for people in middle age, and not much different
for people who have heart attacks. But it's "normal" to die in the US
of heart disease, if cancer doesn't get you first. Good health is
having the same diseases as all your neighbors.

There are countries where heart disease is much rarer to nonexistant,
and almost all of them have thinner people with much lower
cholesterols.  A much talked about "out-lier" is France, where the
people eat a lot of butter and cheese and have cholesterols about the
same as the Brits and Irish (240 mg/dl = about 6.1 mm/L) , but manage a
lot less heart disease than they should for that level.  This does NOT
mean that the cholesterol hypothesis is wrong, but rather that the
French probably eat some things which partly make up for their smoking
and bad diets (like red wine and more vegetables). If you look WITHIN
France and thus control for diet to some extent, you still find that
French with high cholesterol have much more heart disease than French
who have lower cholesterol (I'm talking of total cholesterol, and more
specifically of "bad" or LDL cholesterol). If the cholesterol
hypothesis was wrong, you wouldn't find any correlation in France at


From: Steve Harris <>
Subject: Re: Health Canada warning EZETIMIBE
Date: 6 Feb 2005 12:23:01 -0800
Message-ID: <>

>>Or perhaps cholesterol is secondary to the real cause.

I can understand people with coronary artery disease and doctors wanting
to "Believe" controlling cholesterol is the key.  It is very common for
people in general to want to believe that they can "control" things. It
is very difficult for people and doctors to admit that they do not
understand the disease process and there may be NOTHING they can do and
that they do not have control - at least with the present understanding.


Look, damnit. You can give a rabbit or a monkey terrible
atherosclerosis, which they ordinarily do not get, by feeding them
NOTHING more than added cholesterol to their control diet (on which they
do not get atherosclerosis, either). This was discovered in rabbits
almost a century ago. Fed cholesterol is the ONLY variable in these
experiments. And it can cause honest-to-god full-on
can't-tell-the-difference-from-the-human-kind of atheroslcerosis. All
by itself, witih nothing else. Period.

Now, do you GET it?  That doesn't mean cholesterol is the only variable
in the human process or even the main one. But we know it CAN be causal
*all by itself* of this disease, in animals. That means that it's
extremely unlikely not to be partly causal in humans (whether it
originates from the diet or the liver isn't important once its in the
blood), given the close correlation between disease and blood
cholesterol levels, the known pathogenesis of the disease which
involves macrophages filling up with cholesterol from the blood, and
finally the (duh) obvious facts that atheromatous plaques are filled
with cholesterol goo like the stuff inside of a creampuff.

The animal evidence that cholesterol is partly a causal factor in
atherosclerosis is actually better than the animal experimental
evidence that smoking is partly causal in lung cancer. If you really to
be perverse, why not attack the smoking lung cancer theory first?


From: Steve Harris <>
Subject: Re: pharma and cancer: who profits
Date: 7 Feb 2005 19:11:08 -0800
Message-ID: <>

>> Statins may cause liver cancer in rodents, but just about everything
>> does.
>I guess that is why they have been so successful evolutionary speaking.
>They must thrive on liver cancer ..

Mice in the wild are annual creatures, not much different from
grasshoppers. Average life expectancy is less than 6 months. They don't
live long enough to get liver cancer, which is rare in mice and rats
younger than 12 months, but common after 24 months. By 12 months the
reproductive life of a rodent is done, but mean life span for most
strains is at least 24 months, and max is as long as 33 months, even
with no restriction in diet.

Very few cancers hit mammals of reproductive age. Generally speaking,
humans "in the wild" don't live long enough to get 99% of cancers,
either. Cancer, by and large, is a disease of civilization, where
people live past 45. The yearly chance of a child getting cancer is
something on the order of 1 in 5000, but as people age the risk goes up
exponentially after about 30-- enough that amost one person in 4 will
eventually die of cancer.

Statins don't give young animals liver cancer either, FYI. Both mice
and rats are far more susceptable to cancer than humans, and their
cells transform much more readily in vitro (in the dish). They are very
poor models of tumor production in many ways.

>> However, the idea that much cancer is caused by
>> pharmaceuticals and pesticides and polution in general, really does not
>> stand up to scrutiny.
>Are you for real?


Yep. Are you?  This is one of those issues where political correctness
bears no relation to reality.

Here's a paper for you superbowl hot dip fans. Run in circles, scream
and shout.

In Vivo. 1992 Jan-Feb;6(1):59-63. Related Articles, Links
Carcinogenicity of lifelong administration of capsaicin of hot pepper
in mice.

Toth B, Gannett P.

Eppley Institute for Research in Cancer, University of Nebraska Medical

Center, Omaha 68198-6805.

Capsaicin was administered in a semisynthetic powdered diet at 0.03125%
level for the lifespan of Swiss mice starting from 6 weeks of age. As a
result of C treatment, tumors of the cecum were induced in 22% of
females and 14% of males, whereas the corresponding tumor incidences in
untreated female and male controls were both 8%. Histopathologically,
the tumors were classified as benign polypoid adenomas. Capsaicin is
the main pungent principle of hot pepper, which is consumed in high
quantities by humans worldwide. The capsaicin content of some chili
varieties ranges up to 0.53%.

Publication Types:
Review, Tutorial

From: Steve Harris <>
Subject: Re: high cholesterol and use of flax seeds or other foods
Date: 20 Jun 2005 10:38:44 -0700
Message-ID: <>

>>Have you got any references to back your claim that vitamin C reduces

A Brittish researcher named Spittle claimed this, but multiple attempts
to confirm it have failed. Vitamin C has little or no effect on blood


From: Steve Harris <>
Subject: Re: Natural Hygiene
Date: 8 Aug 2005 18:07:20 -0700
Message-ID: <>

montygram wrote:
> Actually, as I've posted here before, Ancel Keys, in his seminal and
> often cited magnum opus, "Seven Countries" (1979 book), stated clearly
> in this work that the lowest overall mortality was for people who's
> cholesterol levels were between 200 and 220, and this was before it was
> known that only oxidized cholesterol is dangerous!


This figure wouldn't mean much without differentiating by sex. Women
tolerate higher total cholesterols due to having higher HDLs. If you
just take a bunch of people and measure their cholesterols and
mortality rates, you'll see a big proxy effect just from the women in
your sample.

It is certainly NOT true that men have the lowest total mortality
between 200 and 220.


From: Steve Harris <>
Subject: Re: The brain needs cholesterol
Date: 26 Aug 2005 11:58:06 -0700
Message-ID: <>

MMu wrote:
> "Nick" <> schrieb im Newsbeitrag
> > Right on cubit.
> >
> > I would want to explore that.
> > Saturated fat is the precursor to cholesterol in
> > the blood. I supose that is true for the brain as
> > well.
> >
> > Anybody want to correct me?
> >
> yes, I do.
> the precursor to cholesterol is (2 units) acetyl-coenzyme A, not "saturated
> fat".
> AcCoA originates out of ANY kind of fat ("beta-oxidation") and out of
> glucose ("krebs cycle").
> I seriously doubt there is cholesterol deficiency in a healthy human being
> living in an industrialized country.


CORRECT. The dietary precursor to cholesterol is either fat,
carbohydrate or protein. :) Which is to say, calories. This means the
only people with really low cholesterols in their blood are slowly
starving to death, like some people with cancer, or some of the rural
Chinese with cholesterols of 90 in the China Study. Who had a very much
higher infection death rate--- surprise..


From: Steve Harris <>
Subject: Re: Digging Deep in the China Study (or how to discredit a vegan)
Date: 27 Aug 2005 14:21:03 -0700
Message-ID: <>

Max C. wrote:

[Snip a lot of stuff on the China Study]


Since most of the people in the China Study had LDL and total
cholesterols far below the target limits to prevent heart disease, it's
not surprising they have no very good correlations of these with heart
disease.  Rural Chinese in the first study had a total cholesterol
which averages about 125, with 2 standard deviations going from about
90 to 170. Over that range, it probably doesn't matter in terms of
heart disease what your cholesterol is. However, at the lower ranges,
such cholesterols are makers for severe (relative) calorie protein
restriction (by relative, I mean by comparison to how much work the
body needs to do. You can't get at this by looking at total calorie
intake, only by looking at body composition). This very lean state is
highly anticarcinogenic in rodents, and it wouldn't surprise me that it
is high anticarcinogenic in people also.

However, VERY low cholesterols (far below the 150 mg/dL that minimizes
heart disease) predispose to infection death, and they do in the in the
China study also. It's high infection death in rural people with low
animal product intake and very, VERY low cholesterols which drives all
the statistics you're talking about, including the anticorrelation of
TOTAL mortality and animal product intake, and the anticorrelation of
urban living and infection death Which last is VERY odd and impossible
to explain otherwise-- you'd think urbanites would get MORE
communicable diseases like TB, just from closer contact with each

Anyway, the China statistics are all screwed up because they are
looking at whole rural populations who are well under 150 mg/dL total
cholesterol, so any calorie and fat reduction from there on down to 90
mg/dL will not help vascular deaths, but will simply show up as less
cancer, but more infections. And the later will win out over the
former, without really good anibiotics and ICU care. Which rural
Chinese don't get. In THOSE circumstances, you'd best look for all the
animal fat and calories you can get. But rural Chinese are a highly
skewed and selected group which doesn't really look like anybody in the
food-rich US, except maybe anorrhexic cocaine-addicted Hollywood
fashion models.


From: Steve Harris <>
Subject: Re: More calculations on various cancers in the China Study
Date: 28 Aug 2005 10:16:54 -0700
Message-ID: <>

Max C. wrote:
> Well, the 1983 data (as I understand it, 83 mortality info is taken from
> around 1975) didn't include data for some of the cancers I wanted to
> evaluate, but here's some of the interesting points I found after
> running correlation calculations on it.  By the way, I'm getting faster
> with these correlation calculations.  These only took me about 30
> minutes using XL's built in correlation function.  :)
> Processed sugar intake slightly correlated with All Cancers + 18
> Income correlated with NASOPHARYNX AND OTHER PHARYNX CANCER + 27
> Population correlated with NASOPHARYNX AND OTHER PHARYNX CANCER + 30
> Processed sugar intake slightly correlated with stomach cancer + 19
> Interestingly, stomach cancer was negatively correlated with population
> - 22
> Income correlated with colorectal cancer + 24
> Population correlated with liver cancer in those under 34 years + 32
> And once again, we see VERY strong correlations for liver cancer in
> older people with regards to socio-economic factors.
> Sugar intake correlated with liver cancer in 34 to 69 + 38
> Income correlated with liver cancer in 34 to 69 years + 38
> Population correlated with liver cancer in 34 to 69 + 26
> Processed sugar intake slightly correlated with Leukemia in those under
> 34 years + 18
> Income correlated with Leukemia in those under 34 years + 22
> Processed sugar intake correlated with Leukemia in those 35 to 69 years
> + 25
> Income correlated with Leukemia in those 35 to 69 years + 25
> Max.


Again the China study contains a few things which just don't apply to
the west. Nasopharyngeal carcinoma (NPC) is an Epstein-Barr virus (EBV)
related cancer, which only rarely occurs outside China, even though EBV
is certainly common outside China. Even inside China, EBV NPC is a
disease of southern China and is hardly seen in Northern China. Since
North and South China differ greatly in rurality, this badly screws
statistics for this cancer with confounders, since it's probably mostly
genetic (Chinese-Americans have high incidence also, as well as
Greenlanders and Eskimos, and some North Africans, so this is NOT a
diet thing).

You mentioned the HepB and possibly HepC connection with liver cancer.
Again, without controlling for viral infection status, the dietary and
cholesterol information is worthless, because it's noise on top of a
greatly varrying infection carriage rate, which in China is heavily
socioeconomically class dependent, just as it is everywhere else. These
blood-borne viruses are sexually transmitted, vertically transmitted at
birth, and by all the same risk factors that apply to AIDS. Except
they've both been around a lot longer and aren't spreading in China
like HIV.  They're long endemic.  Again, without good data on them,
looking at diet and cholesterol is a waste of time. It would be like
trying to pick out correlations between lung cancer and diet and
cholesterol, when you know nothing about smoking behavior.


Index Home About Blog