[Obesity Discussion]

I'd like to take the time to welcome Epidemiologic Inquiry to Obesity Discussion. They have a wonderful website/blog at http://www.epidemiologic.org/


Here's a sample blog from their website. There are many good reads over there!

Thursday, October 05, 2006
WHI Low-Fat Dietary Trial: What a Billion-Dollar Trial Showed that Epidemiology Already Knew

As they say in baseball and criminal felonies- "3 strikes and you're out" - however, is this adage necessary true in medical research? In a recent blockbuster issue of JAMA, investigators from the decade-long Women's Health Initiative low-fat dietary trial simultaneously reported the results for breast cancer, colorectal cancer, and cardiovascular disease... in essense, results indicated no overall benefit of a low-fat dietary pattern.

Over a decade ago, when scientists first began proposing and planning this low-fat trial, FAT was the hottest craze in medical science. Total fat, not giving regard to types of fat, was deemed the key culprit to chronic diseases. However, around the same time, the majority of large prospective cohort studies had consistently found that total fat was consistently NOT associated with breast cancer, colorectal cancer, nor cardiovascular disease.

Nevertheless, the a low-fat trial was funding by the NIH via pushing and lobbying of Congress, despite mounting scientific evidence that it was not the total fat per se, but rather different types of fats that differentially influenced the risk of such diseases. Notably, trans-fat was already known in the early 1990s that it increased the risk of CVD, while other unsaturated fats were more beneficial for CVD. Differences in types of fat was also being recognized for cancer risks. Thus, 10 years late, the WHI trial ultimately confirmed what epidemiologists from long-running prospectives studies knew all along.

As for the WHI low-fat trial itself- it had many limitations. First of all, its study protocol specified that women reduce their fat intake to less than 20 percent of total intake, grains greater than 6 servings/day and fruits and vegetables greater than 5 servings/day. Nevertheless, only a minority of women hit these targeted goals, and even fewer hit all 3 targets. Furthermore, examining biomarkers of fat intake: serum triglycerides and HDL cholesterol, results indicated that these marker were barely changed at all by the intervention-- these issues raise suspicions of whether the trial accomplished its scientific aims to substantially change fat intake in the first place. Finally, because total energy intake was also decreased by fat in the trial, rather than isocalorically substiting of fat calories by other macronutrient sources, this is considered a major flaw in nutritional research. Essentially, one cannot tease out whether any effects of the intervention is related to reduction of fat or actually due to decreased calorie per se.

Overall, these issues open up lots of questions:
-Was there any true gain in informative knowledge from the trial?
-Were the limitations in study design and methods significant enough to warrant concern?
-Was the trial worth the multi-million dollar investment?
-Should this trial have been conducted in the first place?

In the end, perhaps medicine and public health should utilize a middle ground for equally relying on high quality prospective epidemiologic studies and randomized trials, both of which mutually complements each other study design's strengths and limitations.

Low-Fat Dietary Pattern and Risk of Invasive Breast Cancer: The Women's Health Initiative Randomized Controlled Dietary Modification Trial
Prentice et al.
JAMA. 2006; 295:629-642.

Low-Fat Dietary Pattern and Risk of Colorectal Cancer: The Women's Health Initiative Randomized Controlled Dietary Modification Trial
Beresford et al.
JAMA. 2006; 295:643-654.

Low-Fat Dietary Pattern and Risk of Cardiovascular Disease: The Women's Health Initiative Randomized Controlled Dietary Modification Trial
Howard et al.
JAMA. 2006; 295:655-666.