The 2013 NEJM Mediterranean Diet Study

Every now and then the topic of cardiovascular disease (CVD) and eating healthy comes up in my CrossFit box or in casual conversation at dinner parties. In such cases, the common notion is almost always to reduce fat and cholesterol intake. Invariably I will challenge the person who brings the topic up. I tell them to dig through the actual studies on CVD and fat/cholesterol and see if the data is strong and at all clear. If you follow this site and/or are familiar with Gary Taubes’ books, you will already have an indication that it is not. Recently, some people I know did some digging and agreed that it wasn’t nearly as clear as they thought. But at the same time they noted that the recent NEJM article on Mediterranean diets by Estruch and others (2013) seems to do a good job of making the claim. Does it? Let’s dig in!

This is a fairly large study which monitors its participants for a long duration (generally over 4 years). Participants consisted of men and women without cardiovascular disease (CVD), but they all had either Type 2 Diabetes or at least three of the following major risk factors: smoking, hypertension, elevated LDL cholesterol, low HDL cholesterol levels, overweight or obesity, or a family history of premature coronary heart disease. Interestingly, the authors initially expected an endpoint (cardiovascular-related mortality) of 12% in the control group, but after starting the study they found the actual occurrence was much lower (6.6%). Thus, they needed to add participants and carry followup longer to retain the intended statistical power. Keep all of this this in mind as we go forward. The diets for this study consisted of either 1) Mediterranean diet with extra-virgin olive oil supplementation, 2) Mediterranean diet with nut supplementation, or 3) control. Calories were not restricted, nor was exercise promoted, so these are unknown variables. The study showed that in terms of the primary end point (a composite of heart attack, stroke, and death from CVD), both Mediterranean diets were protective. The results indicated a relative risk reduction of about 30% among high-risk persons who were initially free of CVD. However, looking at the individual components of this primary endpoint, only stroke risk reached statistical significance and was reduced by the Mediterranean diets (heart attack, death from CVD, and death from any cause were not protected against).

So, what do we make of this? If you look at Table 1 you will notice that for both Mediterranean diets they recommend avoiding 1) soda and 2) commercial bakery goods, sweets, and pastries. They recommend avoiding the latter group in the control group as well. But since this control group is meant to be low-fat, other carbohydrates are permitted. Thus refined carbohydrates in bread, crackers, and pasta are allowed. The authors state that the main nutrient changes in the Mediterranean diet are reflected in the fat content and refer the reader to the Supplemental Tables. But if you examines these tables, particularly Table S8, you will see that the authors were referring to recommended (not mandated) nutrient changes. They also didn’t recommend levels of consumables like the refined carbohydrates in bread, crackers, and pasta. As such, it is not surprising to see that the control groups all had greater carbohydrate intakes than either Mediterranean diet. This is a major sticking point with these kinds of studies and Gary Taubes makes a point about this in his books and lectures. That is, whenever you conduct a study and restrict something like fat and don’t restrict the calories in any way, humans will likely make up the difference where they see fit and carbohydrates are often the calories of choice. Thus, the control group in this study, although designated as “low fat” is also high carbohydrate. True, the control group was asked to keep commercial bakery goods, sweets, and pastries low, but they were allowed to drink soda, and load up on other refined carbohydrates like pasta.

The authors note at the end of the paper that the total reduction in fat consumption was small compared to the control group. This puts a totally different spin on the study as most people approaching the work would expect the difference to be a large one. But the authors weren’t trying to reduce the amount of fat extensively, just shift the types consumed. And yet, the contribution of this shift is hard to assess when the total amounts differ by so little. In the end, even if we are generous and assume that the increase in carbohydrates in the controls is not responsible for a loss in protection, this study boils down to data where the protection effects observed are actually due to supplementation with olive oil or nuts. And again, this protection effect, which is small and did not work for all the examined endpoints, was only examined in study participants that were not 100% “healthy” – they were all “at risk” individuals in some form or another. It’s a crucial point to remember – this study doesn’t say anything about the benefits of this diet on healthy individuals currently not at risk for CVD who are trying to prevent it in the future.

Print Friendly