Is Butter Really Back?

Saturated fat has been in the news quite a bit lately. A recently published study has sparked quite the controversy. The study in question appeared in the journal, Annals of Internal Medicine, in March of this year (1). This study was a systematic review and meta-analysis of the research on saturated fat and risk of coronary heart disease. The primary conclusion was that saturated fat has no relationship with risk of heart disease, and that the recommendation to limit saturated fat consumption is not at all scientifically justified.

As usual, the media went overboard with this headline, proclaiming that nutrition scientists have been wrong all along and don’t know what they are talking about.  My two favorite headlines were the Time Magazine article entitled “Eat Butter” (2), and the New York Times piece entitled “Butter Is Back” (3). So, what is going on here? Is butter really back?

The Study

The Annals of Internal Medicine study identified 49 observational studies and 27 randomized controlled trials that investigated the association between saturated fat consumption and risk of coronary heart disease. The data from these studies were combined and a summary risk was computed. When comparing the top third of saturated fat consumers to the bottom third, there was a non-significant 3% increase in risk for coronary heart disease.

Problems With The Annals Study

A good number of nutrition scientists disagreed with the study’s methods and conclusions. Dr. Walter Willett, the chair of the Nutrition Department at the Harvard School of Public health, was among them. He published a comment in Annals concerning the article just a few months after it was published (4). Here is a quick summary of some of the problems he listed: 1) There were gross errors in data extraction. In other words, the researchers didn’t use the most relevant risk estimates from the studies that they included. 2) The omission of important studies. The researchers did not include a number of important investigations, particularly those showing a benefit of polyunsaturated fat consumption on risk of heart disease.

Harvard also sponsored a special teaching session on saturated fat where Dr. Willett mentioned another really important problem with the Annals study: 3) The saturated fat variable was tested in isolation. In other words, saturated fat was investigated on its own, not in comparison to other nutrients.

This third point bears some explaining. When you just look at saturated fat as a variable on its own, it is like comparing it to the rest of the American diet, which is generally unhealthy. So if you are comparing 2 things that aren’t super healthy, one doesn’t look worse for you than the other. 

What is far more useful, is to compare saturated fat to another nutrient in a substitution model. Our total caloric consumption is pretty stable. If we eat less of something, we tend to eat more of something else to replace it. For example, when we reduced our fat consumption in the 1990’s and 2000’s, we replaced it with refined carbohydrate foods, which wasn’t good for our weight or our health. 

A beautifully designed study was published in the American Journal of Clinical Nutrition in 2009 using saturated fat in substitution models (5). This study was a pooled analysis of 11 American and European cohort studies that included a total of 344,696 subjects. When saturated fat was replaced with carbohydrate, there was no association with risk of heart disease. However, when saturated fat was replaced with polyunsaturated fat, there was a 26% lower risk of coronary death and a 13% lower risk of cardiovascular events. Strangely, this study was left out of the Annals review.

Recommendations

In summary, reducing saturated fat is still a good idea. Eating less butter, cream, cheese, and animal fat is still a good recommendation. Replacing those calories with healthy sources of fat, like nuts, vegetable oils, and avocados, is associated with a reduced risk of heart disease. Replacing those calories with carbs, is not associated with risk of heart disease.

Saturated fat may not be as harmful as originally thought, but it is not quite out of jail. Let’s say it is on probation. I still have my clients shoot for a maximum of 7% of their calories as saturated fat, but to include an abundance of healthy vegetable fats in their diet.

References

1) Chowdhury R, et al. Association of dietary, circulating, and supplement fatty acids with coronary risk. Annals of Internal Medicine 2014; 160:398-406.

2) Walsh, B. Eat butter. Time Magazine 6/23/14 pp. 28-35.

3) http://www.nytimes.com/2014/03/26/opinion/bittman-butter-is-back.html?_r=1

4) Willett WC, et al. Letter in response to: Association of dietary, circulating, and supplement fatty acids with coronary risk. Annals of Internal Medicine 2014; 161:453.

5) Jakobsen MU, et al. Major types of dietary fat and risk of coronary heart disease: A pooled analysis of 11 cohort studies. American Journal of Clinical Nutrition 2009; 89:1425-32.