TCS Daily

Eating Some Crow on Fat

By John Luik - February 24, 2006 12:00 AM

"The great tragedy of Science -- the slaying of a beautiful hypothesis by an ugly fact."
-- Thomas Huxley

Of all the beautiful hypotheses in the temple of preventive medicine, the claim that low-fat diets could prevent cancer and heart disease is perhaps the most central. But the last few weeks have been more than a little unkind to the beautiful hypotheses of the lifestyle medicine crowd. In fact the ugly facts have been piling up fairly quickly.

In early January, a study published in the Journal of the American Medical Association reported that low fat diets produced only temporary, moderate weight loss.

Then along came another JAMA study early this year that showed that the consumption of omega-3 fatty acids did not significantly reduce the risk of cancer.

And finally, again in JAMA, there were the three attention grabbing studies which reported that low-fat diets failed to reduce the risks of heart disease, colorectal cancer and breast cancer.

There have been numerous attempts to spin these studies into something that they weren't -- clear, if weak, endorsements of low fat diets as protective for cancer and cardiovascular disease. But that evidence simply isn't there.

For example, in the breast cancer study, incidence rates between the intervention and comparison group did not show a statistically significant difference, meaning that any differences between the groups were indistinguishable from chance. Similarly, in the cardiovascular disease study, the "diet had no significant effects on incidence of CHD (coronary heart disease), ... stroke ... or CVD (cardiovascular disease)."

So whatever the hype about these studies showing "trends" toward the benefits of low-fat diets, the scientific results do not support those claims.

What's difficult to know is just which bit of the story about the new studies is most extraordinary. Is it the fact that the advice provided to Americans to eat diets low in fat in order to reduce the risk of heart disease and cancer is not supported by current science? Is it the fact that this advice hasn't had strong scientific support for the last 30 years? Is it the fact that this doesn't seem to disturb the public health community? Or is the fact that that community isn't going to change the "lifestyle" advice they provide?

The scientific evidence about the link between low-fat diets and heart disease has been tenuous for the last three decades.

In the Multiple Risk Factor Intervention Trial (MRFIT) reported on in JAMA back in 1982, 12,866 U.S. men who were at increased risk from heart disease were divided into two groups -- usual care and special intervention. Members of the special intervention group were given drugs to reduce their hypertension, told to stop smoking and encouraged to reduce their fat and cholesterol intake. Despite these factors, there was not a statistically significant difference in mortality rates from coronary heart disease or any other cause between the two groups, and there was a substantial increase in lung cancer in the intervention group, hardly an optimal outcome.

Or consider the Masai, a nomadic group of Kenyans who consider vegetables and fiber fit only for cows and whose diet consists of milk, meat and blood. Despite this high fat diet the Masai have some of the lowest levels of cholesterol ever measured, with rates about 50% less than the average American and extremely low levels of heart disease. (see G. Mann et al, "Cardiovascular disease in the Masai," Journal of Atherosclerosis Research, 4: 289-312, 1964)

This pattern wasn't confined to the lesser developed world. As Petre Skrabanek reported in his 1995 National Review article "Fat Heads," the Seven Countries Study also fails to support the notion that low fat=lower cardiovascular disease. Crete, with 40% of its diet made up of fat, had one of the lowest incidences of heart disease.

Or consider the results of the famous study from Finland -- the North Karelia Project (Puska et al, "Changes in coronary risk factors during comprehensive five-year community programme to control CVD," BMJ 2: 1173-1178, 1979) Five thousand subjects were recruited for a study that ran for five years and focused on reducing CVD risk factors such as dietary fat. Yet in both women and men there were no significant differences in cardiovascular mortalities.

A little closer to home, there are the results of two long and large studies.

The Framingham Heart Study, which started in 1958 by following 5,200 men and women, has resulted in more than 1,000 published analyses. Here, the data failed to provide any confirmation of a benefit from a low-fat diet in terms of reduced coronary mortality. Indeed, according to a 1987 study in JAMA those with higher levels of cholesterol over the age of 50 -- supposedly the product of a high fat diet -- had lower rates of coronary and total mortality than those with lower cholesterol levels, effectively turning the low-fat hypothesis on its head.

Meanwhile, in the massive Nurses' Health Study involving almost 99,000 nurses aged 30-55 and run by researchers from Harvard since 1976, an analysis in 1997 (Hu et al "Fat intake and the risk of CHD in women," 33: 1491, New England Journal of Medicine, 1997) of CHD risks and dietary fat found some association between 939 cases of developing coronary heart disease among 80,200 of these women reported over 14 years and animal and saturated fat intake, but nothing significant. If there was any effect, it was "consistent with the small-to-negligible effect predicted by metabolic studies." Another way of putting it: No big deal. On total fat and cholesterol intake there was no significant relation to intake and CHD.

Indeed, a 1998 review in the Journal of Clinical Epidemiology of the low fat-cholesterol-CHD thesis found that out of a total of 27 studies involving more than 150,000 subjects, in only three cohorts did patients with CHD consume more fat than the controls. In one cohort, those with cardiovascular disease actually consumed less fat than the controls. In all of the remaining cohorts, there was no statistically significant difference in fat consumption between those who had CVD and those who did not.

And a very large study of 28,000 Swedes -- the Malmö Diet and Cancer Study -- published just last year in the Journal of Internal Medicine noted: "Over the last decades evidence from large-scale epidemiological studies has been emerging, partly defying the previously believed hypotheses, with voices rejecting the fat-disease hypotheses becoming more prominent."

The aim of the Malmö study was to test the basis for the recommendation by governments and health authorities that people limit the percentage of the calories in their diets due to fat to less than 30%. What the study found was that with one exception "individuals receiving more than 30% of their total daily energy from fat (saturated, unsaturated, monounsaturated, polyunsaturated ) did not have increased mortality." Indeed, "receiving more than 10% of total energy intake from saturated fat did not have a significant effect on all-cause, cardiovascular or cancer mortality for men or women." The one exception was women who consumed the highest amount of total fat. They had a relative increased risk (RR) of 1.46 for cancer mortality, though even this result was barely statistically significant as it fell within the a 95% confidence interval of 1.04-2.04 indicating a small risk for subjects of the study dying from cancer in the first place.

These results have all been published in major journals and were certainly known by health professionals. As J. Bruce German and Cora J. Dillard commented in The American Journal of Clinical Nutrition in 2004, "studies on the long-term health benefits of consuming a low-fat diet ... are lacking, and low fat diets have been shown to exert a potentially deleterious effect on lipoprotein profiles in some persons." And as Lawrence Kushi and Edward Giovannucci observed in the American Journal of Medicine back in 2002: "Based on current epidemiologic knowledge, public health recommendations to decrease total fat intake for the prevention of cancer appear largely unwarranted."

As even Barbara Howard, one of the principal investigators for the low-fat studies, admitted to Gina Kolata of The New York Times for her story this month "Low-Fat Diet Does Not Cut Health Risks, Study Finds," ''We are not going to reverse any of the chronic diseases in this country by changing the composition of the diet.'' Except for smoking, all of the advice about healthy lifestyle is based largely on indirect evidence, she noted.

Indirect evidence in science, though, is often no evidence at all. With the JAMA studies this month we have, in what Michael Thun of the American Cancer Society called the "Rolls-Royce of studies," studies that should have found evidence if there was evidence to be found.

Yet, these facts never appear to disturb the public health community. And the strong advice about low-fat diets based on very flimsy evidence continues, with reporters paying more attention to the alleged flaws in the studies as opined by the high priests of the low-fat faith than to what the studies actually say.

In a JAMA editorial accompanying the study, authors Cheryl Anderson and Lawrence Appeal from Johns Hopkins wrote: "Despite null findings ... dietary changes can have powerful, beneficial effects on CVD risk factors and outcomes. To reduce the risk of CVD, individuals should maintain a desirable body weight, be physically active, avoid tobacco exposure, and eat a diet consistent with national guidelines."

Like what? After all, these studies found that significant increases in fruit, vegetables and grains, along with reductions in fat, did not change CVD and cancer outcomes. Eating a diet consistent with national guidelines is eating a low-fat diet -- the same diet that eight years of study and almost half a billion bucks found did not work.

The attitude of we-know-best-no-matter-what-the-studies-say discredits both the scientific process and the requirement that health advice be evidence-based. It risks turning off a population already weary and wary of the lifestyle solutions offered up by the medical community. And it ignores a fundamental responsibility that attaches to providing risk advice: If the science changes, amend your advice. To do otherwise changes the enterprise from advice-giving to propaganda-making. However beautiful the hypothesis, the "ugly" facts are always preferable.

John Luik is writing a book about health policy.



Obesity and Diabetes also seems muddled by media hype.
Many newspaper reports warn that Obesity is "associated" with Diabetes.
This leads many to conclude that Obesity CAUSES Diabetes, so we had better stay thin.

However, when I dig deeper into medical websites, I learn that family genetics are the strongest determinant of who gets diabetes and that with the onset of diabetes, excess insulin promotes the accumulation of fat.

This tells me that Obesity is a SYMPTOM of Diabetes rather than a cause. If you are overweight, you should get tested for diabetes, because that's a sign that you may already have the disease.

So which is it?

No Subject
"Associated with" and "Studies suggest" are the weasil words of the pseudo-sciences that can't establish facts. The public, led on by unserious journalists and headline writers, take these associations and suggestions as clear causes of effects and, if fashionable enough, attempt to micro-manage their livestyles accordingly -- counting grams of fat or extolling the virtues of green tea -- in the mistaken belief that some statistical correlations among large populations have something to do with their particular, personal lives. It's all hubris and self-centered fashion mixed in with faulty reasoning, wishful thinking, and an over-reliance on those who claim to be experts at something.

The discovery of the role of vitamins in human health was, indeed, a marvelous advance. Unfortunately, vitamins merely forestall obvious ill health and death. Other substances may be required for good health and longevity but are too difficult to discover, especially considering a good deal of synergy may be involved, and synergy is almost undetectable by lazy or superficial analyses.

As for me, I'll eat a variety of foods, try to avoid too much over-refined stuff, and generally not eat too much. After that, it's up to genetics and luck, as it is for everyone else. I'm tired of being marketed to by journalists and scientists out for bucks or notoriety. Green tea or black tea, oat bran or some other soluble fiber, corn syrup or maple syrup . . . none of this really matters outside a few broad outlines.

I have also read that the deposit of fat in the belly is concurrent with diabetes, but not its cause. The same mechanism(s) produce both. What underlies that is a mystery I guess. A concurrent rising incidence of diabetes and obesity doesn't prove a link, certainly not a cause-and-effect one. Scientists don't necessarily know what they're detecting, only what they're measuring, which is determined beforehand, obviously, in the design of the study. That method works fine (maybe) in physics or chemistry, but not in sociology and other non-sciences.

I think it is.....
The last 30 plus years the dogma has been low fat. Translated to real language this means high carbohydrates which further reduces to high sugar.
Sugar rich diets cause both obeisity and diabetes. I think the corrlelation between obeisity and diabetes is to be found in the the high carbo diet insanity common to both conditions.
When I look around I can see the effects of the high carb life...lots of fat people. This just wasn't the case when I was a kid fifty years ago.
I was raised on a ranch and my diet has been high in animal protein and fat all my life--probably
70% of my diet. I am 61 now and weigh the same (6'1-1/2" & 175lbs) as I did when 18 years old. It may not be a "scietific" study but it's worked for me and all of my family...except the one sister and a daughter who turned to the high carb page.

African diet and cancer
Another example. African diets seem to produce lower death rates from Cancer. That is until you correct for age.

Cancer is basically a disease we get as we get older. Something like 95% of male deaths due to cancer occur after the age of 55. It's basically our immune system winding down. The only reason there SEEMS to be a cancer epidemic is that we're curing the other things that used to kill us earlier.

If you are in an area of the world where it's unlikely you'll live to be 60 due to starvation, Aids, wars, etc. you won't get cancer, but it has nothing to do with diet.

In 30 years in radiation therapy research, I worked with some real epidemiologists. Nowdays careful study takes a sloppy second place to getting the paper out the door so we can apply for next year's grant.

Statistics is often applied too broadly
There were 3 statisticians out duck hunting.

The first fires at a duck but leads it too much and misses.

The second fires at his duck but doesn't lead it enough and misses.

The third jumps up and yells "we got em!"

Obesity & Diabetes *are* linked
If obesity was caused by diabetes, losing weight would not have an effect on insulin production. Clearing the effect does nothing for the cause. If obesity either caused or was a stressor that created problems for insulin regulation, losing the pounds would improve blood sugar results.

Try talking to doctors who have seen what happens when diabetics drop 30 pounds. My cousin had a light case of diabetes. He's now rail thin and completely under control with the doctor saying "just keep doing what you're doing and I'll see you in six months".

Sorry no evidence
you wrote: "Sugar rich diets cause both obeisity and diabetes. "

There is no eveidence that sugar CAUSES diabetes.

fat chance
This harkens back to an earlier gentler time when poeple were arguing over whether smoking causes cancer. The right wing corporate "junk science" types carried on long after the answer was clear to (almost) all scientists.

Now here we are again claiming there is no obesity epidemic, that fat people are not at greater risk for heart problems and diabedes. You can rail at this or that study (though not in a scientific or medical journal), but the answer is as clear as the smoking/cancer answer.

Socratic synthesis
High blood lipid is caused by excess carbohydrate metabolism, not fat consumption. We possess paleolithic, cave man hunter-gatherer metabolisms. We are designed (evolved), as our ancestors were, to chase over the savannahs after our dinner, or run long and hard to keep from being dinner, getting a low carbohydrate, high fiber diet that included lots of grit and hair. Gathered roots and fruits were nothing like the high sugar, high starch items in the produce section.

Then came agriculture. Folks could stay in one place, build permanent houses, and eat lavishly on high carbohydrate grains. Then came refined flour and sugar, and a sedentary life. Fat city! It is like trying to run a diesel engine on gasoline, possible but with lots of unwanted side effects. Now that we can afford it, we need to return to a rough appoximation of the paleolithic diet and activity. When babies are weaned, it should be without sugar, refined flour or starch and lots of protein, fats and low starch veggies. The little rascals don't know any better, and they'll be healthier. Their adult food cravings won't be perverted by early imprinting with a poisionous high carb diet.

No Subject
"Now here we are again claiming there is no obesity epidemic, that fat people are not at greater risk for heart problems and diabedes(sic)."

I don't know what article you're referring to, but this article is about fat in the diet (not obesity) versus heart disease and cancer.

food doesn't cause disease per say
I agree with don, high carb diet has serious consequences for certain groups of people, including european descent since their access to carb rich foods like fruits and vegies and grains were seasonal, their diets were animal product dominated, tho they did have carbs too like potatoes and other starches that were easier to preserve, but carbs didn't dominate, and animal products automatically raised the amount of fat you consume.

another reason I agree is I went on a low fat diet, trying to improve blood pressure and to see if i could possibly get my weight to budge, after only a few months of this my triglycerides and cholesterol skyrocketed, I never had a problem with it before in the last 15 years that I have been getting tested every year.

granted it might be the birth control pills, which I have been on before and didn't have a problem but these are newer not sure. my blood pressure did plummet, (I am on bp meds for over a year and yet my pressur never went below 140's/85 +, now it is a nice 130/70 which it hasn't been in 25 years since monitering it. I also cut back on bp meds a few months ago and take half of what I was taking.

now was it the lower fat eating? was it a slowing of metabolism due to nto eating enough fat? lowering of metabolism lowers blood pressure too but not due to better cardiovascular health. was it even severe anemia, which is reversing due to treatment. not sure which is which.

my question is why are the medical community and such so determined to stimagize fat, what do they got against it when it is shown to be a benign substance? what do they have to gain to keep pushing low fat or hating fat when studies and experineces shows that it is harmless, (unless one is a true glutton).

is money involved? I have found that science is not as open minded as it should be, science I was told in high school is about forming a theory or hypothesis on why something is the way it is, testing that, observing what happens and adjusting the theory accordingly to test furthur. it seems that science has been corrupted by money, and powerful people who have no interest in truth, and as a result science is no longer trustworthy anymore, you have to really take things with a grain of salt, and see if your experiences jive with what is being said.

too bad, science can really be a source of good information. but since you can't read the hearts of people in science or the media who report on it, you can only guess if they are being as honest as they can be.


TCS Daily Archives