Saturday, July 25, 2009

MRFIT Mortality

The Multiple Risk Factor Intervention trial was a very large controlled diet trial conducted in the 1980s. It involved an initial phase in which investigators screened over 350,000 men age 35-57 for cardiovascular risk factors including total blood cholesterol. 12,866 participants with major cardiovascular risk factors were selected for the diet intervention trial, while the rest were followed for six years. I discussed the intervention trial here.

During the six years of the observational arm of MRFIT, investigators kept track of deaths in the patients they had screened. They compared the occurrence of deaths from multiple causes to the blood cholesterol values they had measured at the beginning of the study. Here's a graph of the results (source):

Click on the graph for a larger image. Coronary heart disease does indeed rise with increasing total cholesterol in American men of this age group. But total mortality is nearly as high at low cholesterol levels as at high cholesterol levels. What accounts for the increase in mortality at low cholesterol levels, if not coronary heart disease? Stroke is part of the explanation. It was twice as prevalent in the lowest-cholesterol group as it was in other participants. But that hardly explains the large increase in mortality.

Possible explanations from other studies include higher infection rates and higher rates of accidents and suicide. But the study didn't provide those statistics so I'm only guessing.

The MRFIT study cannot be replicated, because it was conducted at a time when fewer people were taking cholesterol-lowering drugs. In 2009, a 50-year old whose doctor discovers he has high cholesterol will likely be prescribed a statin, after which he will probably no longer have high cholesterol. This will confound studies examining the association between blood cholesterol and disease outcomes.


Ricardo said...

Low cholesterol levels are certainly not good for your health! There is epidemiological data associating lower cholesterol with an increase in impulsive or violent behaviors and some authors even suggest that "low-cholesterol may be a marker for risk of suicide or traumatic death". This is because there might be a link between cholesterol and serotonin synthesis. As we all know, brain cholesterol is essential to cell membranes and have a number of biological functions, so it is no surprise that low cholesterol have a role in the development of neurodegenerative disorders, like Alzheimers. Low cholesterol is also associated with some cancers, namely lung, liver, lymphatic and hematopoietic cancer. There is also the Smith-Lemli-Opitz syndrome and its deformities to remember us of the importance of adequate cholesterol synthesis. Cholesterol is also necessary for the synthesis of Vitamin D and Dr. William Davis reports a strong link between Vitamin D and HDL levels. They get really high with D3 supplementation ( Dr. Uffe Ravnskov suggests that high cholesterol may protect against infections and even atherosclerosis ( For all these reasons, a few doctors are now thinking that "the benefits associated with cholesterol reduction may not outweigh the risks in all patients with hypercholesterolemia. Cholesterol-lowering interventions should be recommended with caution in patients at increased risk of cancer, stroke, and depression". Here is a Pubmed list of abstracts about low or lowered cholesterol health implications -

Robert Andrew Brown said...

Thanks Stephan

Thanks Ricardo,

Great links more reading (-:

Glenn said...

Could be it harmful to eat significant amounts of starch and fat at the same time, like say a diet of 40% fat, 40% carbohydrate, and 20% protein?

I have read secondhand that Polish high animal fat advocate Dr. Kwasniewski considered such a diet in the "forbidden zone," and that it would cause cardiovascular disease.

This reminds me of a sports nutrition book I read quite a while back where the author was recommending people eat high fat/low carb one meal, and low-fat/high carb the next, claiming it was more balanced, and easier for the body to digest foods this way.

It got me thinking that perhaps there could be an evolutionary basis for this:

Perhaps paleolithic hunter gatherers ate a high fat diet while they were able, but once they lost track of the herd, or the herd died out, they switched to a low fat/high carb diet based on wild tubers until such time as a new herd could be located - thereby eating predominately either high fat, or low-fat most of the time.

I think there are examples of cultures that eat high carbohydrate and are healthy. Ditto for high fat diets. But do we have any examples of cultures mixing fats and carbohydrates in roughly equal amounts while being in excellent health?

Jason said...

Hello, Stephan.

I do not have a comment regarding your recent blog post. I was hoping I might post a link to my blog, as I have posted a few recent posts that relate to debunking one of the MP's main hypotheses; That exogenous 25-D can dock to and deactivate the VDR, thus handicapping innate immunity. I know you share a wide viewer base here with your blog and was hoping to see if any would be interested in reviewing my short posts and commenting with any feedback.

Please delete my comment if I am inappropriately hijacking your comments thread.

Thanks and regards.

Jim Purdy said...

This is just one more reason why I don't regret that I stopped taking statins years ago.

Stephan Guyenet said...


I haven't seen anything that would suggest that's true. There are a number of healthy cultures in Africa that mixed fermented grains and dairy and seemed to do well. Although I don't have any detailed data on their cardiovascular health.


Sounds good. I like your point about 25(OH)D3 not being able to interfere with VDR activity because it doesn't get into the nucleus.

Glenn said...

Here's another question I have:

If wheat is really so unhealthy, how do we explain the supposed robust health and longevity of the population in Crete?

It's my understanding that they used wheat in one form or another as one of their staple foods.

I'm one step away from the double cheeseburger (butter instead of mayo of course) making a comeback at my house if I could just figure out what to do for the hamburger buns.

billye said...

I was led to your site and this post by Dr. Kenneth Tourgeman "" who recommends this post very highly. 10 months ago Dr. Tourgeman started treating me via nutrition and supplements only, while reducing many of the pharmaceuticles I was taking. He immediately stopped my Staten's. After a while my triglycerides went from 115 mg/dl down to 66 mg/dl. A remarkable drop, especially since this was done without the use of Staten's. He explained to me that the only reason Staten's, with all of their side effects work, is the fact that it raises vitamin D levels. Hence, since he told me to take 6000 IU vitamin D3 I did not need Staten's.

Incidentally, he cured my diabetes type 2 and I now regularly receive A1c levels of 4.7, 4.8, and the latest 5.0. I now regularly take high dose Fish oil gel caps, super vitamin K2, and kelp caps. As an added bonus I have lost 54 pounds so far.

He has made a new man out of me through his willingness to step out of the box and switch his nephrology practice to one of the diet of evolution nutrition only. He also recommends great blogs like yours and Dr. Davis's on a regular basis to all of his patients. His motto is "an educated patient is his best patient" His only motivation is to now reverse and cure illnesses instead of just treating them.

Anna said...

"if I could just figure out what to do for the hamburger buns"

Glenn, what's wrong with eating double cheeseburgers from a plate with a fork & knife? That's how our entire family eats burgers now, brat wursts and hot dogs, too. If you really must hold the burgers in your hands to eat them, have you tried wrapping with large lettuce or cabbage leaves? It works, though it can be messy (but so can buns if you use a lot of condiments and toppings).

And you can make delicious mayo without high n-6 oils in less than five minutes. Homemade garlicky aioli is even better on a burger!

Lisa said...

Thanks, Stephan, for yet another great series of posts. I was wondering what your current thinking is on blood lipid patterns perhaps being markers of diet composition rather than causal factors in CVD, as discussed in “Cardiovascular Risk Factors on Kitava, Part II: Blood Lipids”:

“I have a theory of the relationship between blood lipids and CVD that can explain these data. I believe that blood lipids, rather than causing CVD, simply reflect diet composition and other lifestyle factors. Both on Kitava and in the West, low HDL and elevated triglycerides imply a high carbohydrate intake. Low-carbohydrate diets consistently raise HDL and lower triglycerides. On Kitava, carbohydrate comes mostly from root crops. In the West, it comes mostly from processed grains (typically wheat) and sugar. So the blood lipid pattern that associates best with CVD and the metabolic syndrome in the West is simply a marker of grain and sugar intake.”

Does the differential effect of fructose vs glucose on lipids (as shown in the Stanhope et al. article that you looked at in “Fructose vs. Glucose Showdown” for example) bear on this issue at all do you think? It seems a high-fructose diet would result in a much more pronounced high carb pattern of low HDL and elevated triglycerides than a high-glucose diet..

Glenn said...


"He explained to me that the only reason Staten's, with all of their side effects work, is the fact that it raises vitamin D levels."

Billy, you might be interested in this Chris Masterjohn article where he argues statins may help, not because they decrease cholesterol, but because they inhibit the activation of Rho:

“High cholesterol levels may be a marker for something far worse — chronic over-activation of Rho. And cholesterol lowering with statins may be a marker for something far better — inhibition of the activation of Rho.”

He goes on to argue that the best strategy is decreasing inflammation:

“A more fruitful battle plan would be to attack the cause of the problem itself — inflammation.”

I think diets as recommended here (low n6 balanced with n3) should do that naturally.

sandra said...

I'm not clear on how mortality in this study was linked back to cholesterol levels... it sounds like levels from the initial screening were used. Couldn't levels have changed by time of death?

On the question of blood lipids reflecting diet: What could the combo of LOW triglycerides and LOW HDL mean (that's what I seem to least last time I checked)? I'm very concerned that my total cholesterol is way too low (135).

Stephan Guyenet said...


The Cretans don't hold a candle to hunter-gatherers and certain non-industrial agriculturalists in terms of chronic disease. I'm not proposing that wheat is the only factor, simply that it's one factor. I acknowledge that there are cultures that eat wheat and are healthier than we are in the US. They typically have a better n-6:3 balance than us and eat less sugar.


That's amazing, thanks for sharing your story. I just took a look at Dr. Tourgeman's blog today, there's a lot of great stuff on it.


Wow, you remember my posts better than I do, haha. I still think that might be true. However, I've been learning more about it and I think there are other factors at play. I touched on it briefly in the post I just put up, and I'll be mentioning it again in an upcoming post.

I agree with your comment on fructose vs. glucose, I think that's a huge factor for the average American who gets almost a quarter of her calories from sugar.


The goal of the study was to gain information whereby a physician can measure a person's cholesterol and say: "statistically speaking, you have X risk of dying in the next six years and Y risk of dying of a heart attack". So their method of measuring cholesterol once and looking at six-year mortality is applicable to that question.

Ricardo said...

"Lowering cholesterol concentrations and mortality: a quantitative review of primary prevention trials" -

Dirk said...

uh oh , not Anthony Colpo again. That man is not smart enough to understand that total caloric intake , nor insulin ( insulin levels in the blood are nopt different in certain areas) can explain progressive lipodystrophy.

Nor can it explain animal husbandry where you breed a cow to be fat or lea, Are the lean cows bred to jog more and eat less grass?

He refuses to acknowledge Taubes' refutation of the calorie theory which is based on sound science.(it is just Taubes has no replacement hypothesis that is accurate)

I don't trust anything written by Colpo. He is not qualified and he does not understand that we need to know what is happening at the cellular level to prevent obesity and heart disease.

Dr.Eades $20,000 by the way.

trinkwasser said...

"Could be it harmful to eat significant amounts of starch and fat at the same time, like say a diet of 40% fat, 40% carbohydrate, and 20% protein?"

This is an opinion I've come to, from reading here and numerous other sources and the original papers.

Saturated fats may be dangerous if you're already eating toxic quantities of carbs (and generating the necessary quantities of insulin to deal with them, more with insulin resistance).

This is the dietary paradigm of all too many papers.

Without those carbs the saturated fat is processed quite differently: anecdotally it's commonplace to see saturated fats turned into HDL rather than LDL, but studies where the carb/insulin levels are low enough to show this are uncommon

O Primitivo said...

"Which Cholesterol Level Is Related to the Lowest Mortality in Population with Low Mean Cholesterol Level: A 6.4-Year Follow-up Study of 482,472 Korean Men" -

Stephan Guyenet said...

Nice find, Ricardo.

trinkwasser said...

"The MRFIT study cannot be replicated, because it was conducted at a time when fewer people were taking cholesterol-lowering drugs. In 2009, a 50-year old whose doctor discovers he has high cholesterol will likely be prescribed a statin, after which he will probably no longer have high cholesterol. This will confound studies examining the association between blood cholesterol and disease outcomes."

Now that is an excellent point! I suspect it will *appear* that people with lower levels of "cholesterol" will suffer from higher levels of CVD because of the difference between low levels caused by efficient metabolism and those caused by medicating down the effects of an inappropriate diet without mending the actual metabolic factors which are behind the disease process.

Thus people with lower "cholesterol" will appear to suffer more heart attacks, so "cholesterol" levels will need to be decreased further with more drugs. Wait a minute, that's already happening!

This also makes me think (can you tell I've been re-reading your blog again?) you, I and not a few other people appear to have higher HbA1c than calculated from spot glucose readings. All such people appear to be eating higher fat diets. Maybe the effect is that our blood cells live longer and don't need replacing so often, so pick up more glycation for this reason?

Such an effect might explain why "fat" appears in some studies to relate to increased insulin resistance if they are only looking at A1c and not actual glucose levels, which almost inevitably decrease on higher fat diets. The best indicator of IR, trigs/HDL, almost inevitably also improves on such diets. This is also a much better metric for CVD than Tchol.

In exactly the same way, reducing A1c using an inappropriate diet and high levels of medication produces worse results than reducing A1c through diet (ACCORD etc.)

One day people will look back on our "science" and laugh . . .

celine said...

Thanks for the post! I enjoyed it. I'm looking for easy and effective way to cholestrol reduction and I learned something new.