The recent controversy over methodological errors in the Spanish Mediterranean diet study, PREDIMED, was overblown.
Critics noticed that the patients were not strictly randomized, and called the authors to account. One study site had allocated all patients to the Mediterranean diet, and a protocol change not mentioned in the original publication resulted in some household members of participants who had been randomized to one of the diets also being assigned to that diet.
The authors retracted the study, re-analyzed it taking into account the methodological errors, and republished it in the New England Journal of Medicine. The results were essentially identical. The Mediterranean diet supplemented with nuts reduced stroke by 46% over 5 years in both analyses.
Although the methodological criticisms were strictly accurate, the errors made by the one study site did not affect the outcomes, and I think the decision to allocate all members of a household to the same diet was sensible and did not in any way impair the quality of the study. No household wants to prepare two different meals every day.
Furthermore, this Spanish study, though the latest and the greatest, simply confirms in primary prevention what many other studies have shown.
In the Seven Countries Study, the coronary risk on Crete was one-fifteenth what it was in Finland, where people eat more like North Americans, and only 40% of the risk in Japan, where only 10% of calories were from fat.
Recent books and articles have (with justification) demonized low-fat, high-carbohydrate diets but presented these as if they were the only alternative to a high-fat diet that is heavily based on consumption of meat and eggs. This represents a form of dietary suicide.
The Mediterranean diet is much misunderstood. It is not a high-carb, low-fat diet; it is a high-fat diet and therefore a low glycemic index diet. In the Seven Countries study, 40% of calories in Crete were from fat, mainly olive oil; whereas in Finland, 38% of calories were from fat, but mainly animal fat.
Most people think the Mediterranean diet is high in fish and chicken and low in red meat. This is partially true, but the quantity of animal flesh is misunderstood. In a retrospective article, Ancel Keys, PhD, the leader of the Seven Countries Study, described “the good Mediterranean Diet” as “a mainly vegetarian diet, favoring fruit for dessert …, low in meat and dairy.” The diet favors whole grains, nuts, legumes (beans, lentils, chickpeas, etc.), and high consumption of fruits and vegetables.
Thus, a daily serving of animal flesh of about 4 to 8 oz (or more), which might be the norm for most North Americans, does not conform to the Mediterranean diet. It would be more like 2 oz a day, mainly fish and chicken, or a serving the size of the palm of the hand about every other day.
The benefits of the Cretan Mediterranean Diet are probably of two kinds, negative and positive:
- Substituting olive oil for saturated animal fat and its evil companions — cholesterol, carnitine from red meat, and phosphatidylcholine from egg yolk.
- High fiber, the beneficial components of whole grain such as antioxidants and lignans, and a wide range of antioxidants from fruits and vegetables.
Many fruits and vegetables get their color and flavor from bioflavonoids: naringin in grapefruit, lycopene in tomatoes, resveratrol in red grapes, anthocyanin in blueberries, etc. A useful slogan is “we should eat fruits and vegetables of all colors.” This is why the dietary pattern is more beneficial than taking supplements with single or only several components, such as vitamin C and vitamin E. A recent meta-analysis confirmed that, with the exception of B vitamin complexes and folic acid (of which more later), there is no cardiovascular benefit of taking vitamin supplements.
A very important study conducted in Israel by Iris Shai, RD, PhD, and colleagues compared the Mediterranean diet against both a low-carb diet, similar to the Atkins Diet, and a low-fat diet. Weight loss was equal on the low-carb and Mediterranean diets, and both were significantly better than the low-fat diet. More importantly, the Mediterranean diet was clearly superior in reducing blood sugar, fasting insulin levels, and insulin resistance among diabetics. It is clearly the best diet for diabetes.
Whereas most dietary studies in free-living persons have high dropout rates, this study achieved 95% adherence at 1 year and 86% adherence at 2 years. This was accomplished because the study was conducted in a large institution where the residents obtained their meals from the cafeteria, and meals were color-coded so that residents would not take the wrong foods.
In the Lyon Diet Heart Study (1995), the Mediterranean diet reduced recurrent heart attacks and strokes by 70% in 4 years. By comparison, simvastatin in the contemporaneous Scandinavian Simvastatin Survival Study (1994) only reduced recurrent heart attacks by 40% in 6 years. There are also many studies showing that adherence to a Mediterranean diet markedly reduced risk in observational studies. A recent one of these came from the REGARDS study in the United States.
It is possible that a vegetarian diet might be even better than the Mediterranean Diet (though vitamin B12 supplements would be needed), but that hypothesis has not been well tested. For now, the best diet for cardiovascular prevention is the Cretan Mediterranean diet.
The American Heart Association reported in 2015 that only 0.1% of Americans consume a healthy diet, and only 8.3% consume a somewhat healthy diet, so there is much room for improvement.
J. David Spence, MD, is professor of neurology and clinical pharmacology at Western University in London, Ontario, and director of the Stroke Prevention and Atherosclerosis Research Centre at the Robarts Research Institute there. He is a Gold Fellow in the Stroke Council of the American Heart Association (AHA), and has served on leadership committees of both that Council and the Peripheral Vascular Disease Council of the AHA.