February 08, 2006
8 Year Study Finds Small Risk Reduction From Lower Dietary Fat

Does lowering fat in the diet provide any health benefits?

Adopting a low-fat diet in later life and following such a regimen for nearly a decade does not appear to have a significant impact on reducing the overall risk of breast cancer, colorectal cancer or heart disease, according to a Women's Health Initiative study that involved nearly 50,000 postmenopausal women across the United States. The results of the federally funded dietary modification study will be published in a series of three papers – two with lead authors at Fred Hutchinson Cancer Research Center and all three involving co-authors from the Hutchinson Center – in the Feb. 8 issue of the Journal of the American Medical Association, or JAMA.

The study – the first attempt to test the health impact of a low-fat diet in a randomized, controlled trial, considered the gold standard of clinical and public-health study design – did, however, uncover some encouraging trends, according to Hutchinson Center biostatistician Ross L. Prentice, Ph.D., lead author of the JAMA paper that describes the impact of a low-fat diet on breast-cancer risk, one of the primary goals of the study.

"Women in the low-fat-diet group reduced their overall rate of breast cancer by about 9 percent as compared to the women who didn't change their eating patterns, but that difference was not statistically significant; it could have been due to chance. So at this point we're not able to say with certainty that a low-fat diet reduces the risk of breast cancer," said Prentice, member and former director of the Hutchinson Center's Public Health Sciences Division. A 9 percent reduction in breast-cancer incidence means that, out of 10,000 women, 42 in the low-fat-diet group and 45 in the comparison group developed breast cancer each year.

Prentice and colleagues did find, however, that a low-fat diet was associated with a statistically significant 15 percent reduction in estradiol, a form of blood estrogen that increases the risk of breast cancer.

My guess is that if people could manage to stay on diets for decades then we'd see larger differences in results.

The 30% reduction in a single subtype of breast cancer suggests they really did find an effect from lower fat diets but that only some mechanisms by which cancer is generated are affected much by fat level in the diet.

Women in the low-fat group also experienced a 30 percent risk reduction for a certain subtype of breast cancer: tumors that were progesterone-receptor negative. "This finding provides an interesting hypothesis for further development and reinforces that breast cancer is multifaceted; it is not a single disease," Prentice said. PR-negative tumors, while relatively rare, are difficult to treat and associated with a higher mortality rate because they are unresponsive to hormone-blocking drugs such as tamoxifen.

The larger the decrease in fat in the diet (due to starting with a higher percentage of fat to begin with) the bigger the measured benefit:

Significant results were seen also among women in the low-fat-diet group who began the study with the highest baseline fat consumption and among women who most strictly adhered to the study's dietary-fat goals. Women in these categories experienced a 15 percent to 20 percent overall reduction in breast-cancer incidence.

"The bottom line is that changing to a low-fat diet may reduce breast-cancer risk, especially among women who have a relatively high-fat diet to begin with, but we don't view our data as strong enough at this time to make a broad recommendation that all women initiate a low-fat diet for that purpose," Prentice said. "Additional follow up with these women may yield a stronger, statistically significant conclusion."

With regard to colorectal cancer, the study did not reveal a reduction of cancer incidence overall, but it did show a modest 9 percent decrease in self-reported colon polyps – a precursor to colon cancer – among the women in the low-fat intervention group, according to Shirley A.A. Beresford, Ph.D., lead author of the paper describing the colorectal-cancer findings.

"It is important to remember that cancers often take decades to develop, and we may only be seeing the early stages of the impact of a low-fat diet intervention on the risk of colorectal cancer and other diseases," said Beresford, a member of the Hutchinson Center's Public Health Sciences Division and a professor of epidemiology at the University of Washington School of Public Health and Community Medicine. "The reduction in polyps suggests a possible reduction in colorectal-cancer risk could emerge over a longer time period." No significant reduction in heart disease emerged among the women in the low-fat intervention group, who achieved only a 2.4 percent reduction in low-density lipoprotein, or LDL, the so-called "bad" cholesterol, and a 3 percent lower rate of heart disease.

The study did, however, find trends toward reduction in heart-disease risk among the subset of women in the low-fat-diet group who made the greatest reduction in consumption of saturated fat and trans fat, both of which can raise the risk of heart disease because they increase production of LDL cholesterol.

"For heart-disease prevention, the data suggests that a greater emphasis on reduction of saturated and trans fats will be needed to have a major difference," Prentice said. Barbara V. Howard, Ph.D., president of MedStar Research Institute/Howard University in Washington, D.C., was the lead author of the heart-disease paper.

A reduction of dietary fat to only 20% of calories still falls well short of a Pritikin style diet where one might get only 10% of calories from fat. If they'd tried for a larger fat reduction they might have picked up a stronger signal in the data.

In my mind I hear Joe Jackson singing.

Everything gives you cancer
Everything gives you cancer
There’s no cure, there’s no answer
Everything gives you cancer

But wait. There are some major caveats when interpreting these results.

Forty percent of the participants were assigned to the low-fat diet, in which they were asked to reduce their fat intake to 20 percent of their total calories while eating five or more daily servings of vegetables and fruits and six servings of grains. The remaining 60 percent served as a comparison group and did not change their diet.

Although the primary hypotheses of protective effects of a low-fat diet on breast and colorectal cancer failed the test, the WHI researchers pointed out that the majority of women assigned to the low-fat diet didn't meet the 20 percent fat goal: On average, the women reduced their fat intake to 24 percent in the first year, but slowly increased their fat intake to 29 percent by the eighth year.

Furthermore, the study showed that women who had the highest fat intake at the study's outset showed greater evidence for reducing their breast cancer risk on the diet program. There was also a suggestive trend of breast cancer risk reduction for women who initially had the lowest consumption of vegetables and fruits and then increased their intake by one serving per day as part of the diet.

So the study was only for 8 years and yet our risk of cancer is the result of decades of accumulation of damage to our bodies. Also, the ladies didn't stay on their diets just like people do not stick to all the other diets they go on. Expecting people to stick with a diet for years isn't realistic.

A final point: I predict that in 10 years time we'll have plenty of genetic tests that tell us not only our genetic predisposition to various cancers (we already have some such tests) but also the genetic tests will tell us how much our individual risk will be modified by various dietary changes. Some people will be told that lowering their dietary fat won't matter much. Others will be told that lowering their dietary fat (or perhaps just specific kinds of fat) will have a big impact on their cancer risks. I hope the conductors of the study above took and stored DNA samples from the participants in this study. 10 or so years from now when DNA testing will be orders of magnitude cheaper testing of the DNA sequences of study participants could yield very useful information on how DNA variations and diet interact to influence cancer and heart disease risks.

Share |      Randall Parker, 2006 February 08 10:51 PM  Aging Diet Studies

Kelly Parks said at February 9, 2006 8:15 PM:

Dude, you've gotta fix your filter if you're gonna have posts like this. My brilliant comments about your 0besity post were disallowed because your filter doesn't allow w8t-loss comments.

remo williams said at February 10, 2006 1:58 AM:

It wont take 10 years...more like 5 to 7 years.

Kelly Parks said at February 10, 2006 9:43 AM:

You are familiar with the Flegal study, right? The Flegal study (http://jama.ama-assn.org/cgi/content/full/293/15/1861) showed, among other things, that obesity causes 25,000 excess deaths (not 400,000 as the CDC was reporting), that being underweight caused 37,000 excess deaths and that people who were overweight (but not obese) lived longer than people with "ideal" BMI's.

This study fits well with that one.

Kelly Parks said at February 10, 2006 9:50 AM:

Another related item. This article (http://tinyurl.com/9z6yl) is about a study that shows obese men can reverse type 2 diabetes and metabolic syndrome with lifestyle changes even though they’re still obese. They didn’t have to lose the weight to get healthy. Being sedentary is dangerous -- being fat is not.

Randall Parker said at February 10, 2006 1:15 PM:


When you hit filter problems you should email me the post you wrote and the error message. You can often get around it by avoiding dashes in the word you write.

Sorry I have to have a few thousand regular expressions looking at what gets posted. It is either that or get several thousand spam comments per day. I'm going to upgrade to a better way of doing comments that'll eliminate the bots and then I might be able to prune back the regular expression filters.

As for the studies: Obesity really is bad for you. Dr. Rudolph Leibel of Columbia University and other researchers have found that fat cells secrete dozens of hormones and some of the effects of these secretions are not desirable.

Kelly Parks said at February 10, 2006 7:16 PM:


I didn't say it wasn't bad for you -- the Flegal study did find 25,000 excess deaths per year related to obesity -- I just said it's nowhere near as bad for you as most people think. It's not the health crisis that it's portrayed as in the media.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©