Widely Reported Dietary Fat Study Causes “Massive Confusion”

A recent meta-analysis of studies on dietary fat and heart disease (Ann Intern Med 2014;160:398-406) caused huge excitement in the media which reported it as showing that the conventional guideline to decrease saturated fat and increase polyunaturated fat has been debunked.

But, not only did the media get the study wrong, more importantly, the study got it wrong. . . .

The study did not find that it doesn’t matter what type of fat you eat. It found that:

1. People who eat the most trans fats have more heart disease than people who eat the least
2. Saturated fats and omega-6 fats have no effect–positive or negative–on heart disease
3. People who eat the most omega-3 fats have significantly less heart disease.

So, the study actually found that people who eat the most polyunsaturated omega-3 essential fatty acids have less heart disease than people who eat the most saturated fat.

So the media reports on the study were seriously misleading. 

But, much more importantly, and not reported at all by the media, the study itself is misleading. And, within days of its publication, the authors were forced to post a new version to correct their several errors. However, the media has not informed the public of these corrections. 

The paper contained an incredible number of errors. For example, it reported one study on omega-3’s effect on heart disease as slightly negative, when it was, in fact, strongly positive. Two out of six of the studies on omega-6 were analyzed wrongly in the study, according to Walter Willet, Chair of Harvard School of Public Health Nutrition Department, saying they slightly increase the risk of heart disease when the studies actually showed that they significantly decreased the risk. Willet and others also criticize the study for omitting relevant data from several other studies that also would have changed the conclusion. 

Crucially, Willet points out the serious flaw that, when saturated fat was reduced, the study did not take into consideration what it was replaced with. Of course, if it is replaced with calories from another source that is bad for the heart, like refined carbohydrates or sugar, reducing saturated fat won’t reduce heart disease. But if it’s replaced with monounsaturated and polyunsaturated fats–and that’s the relevant comparison–there is a lot of evidence that it will.

For example, a previous review of the research (Am J Clin Nutr 2009;89:1425-1432) has shown that reducing saturated fat by 5% and replacing it with polyunsaturated fats reduces the risk of coronary events by a significant 26%. But replacing it with carbohydrates increases the risk by 7% (the review did not take into consideration the type of carbohydrate and points out that the negative result may pertain to refined carbs but not to high fibre complex carbs).

Willet says that because of multiple serious errors and omissions, the conclusions of the study are seriously misleading and should be disregarded. He says that “it has caused massive confussion and the public hasn’t heard the correction.”

Since science becomes useful and valuable when the public is informed about it, and since the public has been informed about the misleading conclusion and not the corrected conclusion, the public has been mislead. And, in this case, since it is about heart disease, the public has been dangerously mislead. Willet says “A retraction with similar press promotion should be considered.”