With our class just wrapping up epidemiology with Dr. Stellman, I thought it appropriate to drop a quick blog about the much talked about Mediterranean Diet study published in NEJM this week. And since most of us are recovering from studying for the final, I thought I’d summarize it for you. The article brings new light to our hummus Sundays during class weekends (as long as you don’t partake in the baklava). The study written by Ramon Estruch and colleagues from the Department of Preventive Medicine and Public Health in Pamplona, Spain compares two Mediterranean diets with a “low fat” control group with the end point of reduction in major cardiovascular events. These include heart attack, stroke, and cardiovascular death. Included patients had no cardiovascular disease (CVD) at entry but had either diabetes or at least three risk factors (such as high blood pressure, being overweight, or smoking, among others). This was a parallel-group, multicenter, randomized control trial (i.e. the pinnacle of studies).
So what were these diets? It’s much more than an extra helping of hummus, I assure you. The first group consumed at least four tablespoons of extra virgin olive oil (EVOO) per day, which was used for cooking and salads. The chosen oil was polyphenol-rich and supplied by a local Spanish vendor. This was in addition to fresh fruits (at least 3 servings/day), veggies (at least 2 servings) plus a prescribed amount of sofrito (tomato, garlic, herb sauce), legumes, and white meat instead of red. Interestingly, participants in this diet also consumed at least seven glasses of wine with meals per week (optional for “habitual drinkers”…you know who you are). Discouraged items included soda, commercial bakery goods, sweets, and pastries (i.e. not homemade), spread fats, and red/processed meats. The second group was essentially the same except that they were to consume one serving of nuts instead of the olive oil. And the control consisted of a low fat diet: at least three servings of low fat dairy, grains, fruits, veggies, and lean fish/seafood. The control was discouraged to consume vegetable oil (including olive oil), sofrito, as well as nuts (in order to differentiate them from the Mediterranean groups). They were also discouraged to consume red fatty meats, spread fats, and the same pastries. I liked that the diets were easy to adhere to (as opposed to previous “all meat, no apples” diets that were popular a few years back).
Diet was communicated via training sessions by nutritionists at baseline then quarterly and adherence was assessed via questionnaires with supplemental personalized advice based on surveys. I know what everyone is thinking – that questionnaires are imperfect ways to measure adherence and that people may state adherence when they’re really not. Let’s face it; diets are very hard to change. This turns out to be a huge plus for the study because they actually measured biomarkers such as urinary hydroxytyrosol levels (to confirm olive oil consumption) and plasma lapha-linolenic acid (to confirm nut consumption). Cool, no? These Spaniards take their olive oil consumption very seriously and have it down to a science!
For the class of 2014, who are knee deep in stats, you may be expected to do the power calculation if Professor Weinberg catches wind of this, so let me just give it to you now. They estimated that 9,000 participants would be required to provide a power of 80% to detect a relative risk reduction of 20% in each Mediterranean-diet compared to the control. And they almost made that, but falling short didn’t end up mattering. During October 2003 to June 2009, they had 7,447 participants. The groups were well balanced in terms of risk factors and drug-treatments at baseline (for the most part; there were some adjustments made for differences later). Participants were followed for a median of almost five years, during which there was similar adherence reported. In terms of results, the trial was ended early because the endpoint was achieved. In the olive oil group, the crude rate of CVD events per 1,000 person year was 96 compared to 83 in the nuts group and 109 in the control (resulting in significant p-values for both). This is a 30% relative risk reduction. After adjustments, the difference was still significant. In the secondary analysis, there was also a significant reduction in stroke, but not in heart attack or other secondary events. To quote Dr. Stellman’s response to the study, “One clear finding was that any advantage of the EVOO or nut supplement was most clearly evident in the obese (BMI > 30) and those with hypertension, but did not seem to favor any other subgroups.”
Why does all this matter? Heart disease is still the leading cause of death in the US for both men and women (and combined). Dr. Stellman went on to comment on the implications. “So where does that leave us? There is almost universal agreement in the world of preventive medicine that diet and nutrition are closely linked to adverse health endpoints, including heart disease and cancer.” He goes on to cautiously recommend, “In my view, the trial does provide us with valuable, if limited, data that show, as most of its predecessors have shown already, that compliance is achievable albeit with great effort and expense, but that single dietary manipulations rarely lead to unequivocal results. A major problem that I doubt will ever be fully resolved is that all such studies are plagued by imprecise measurement of diet, so that even if a trial were to demonstrate an undeniable superiority of one intervention over another it would be difficult to identify the component or components that made the difference, and that the magic ingredient, whatever it is, is more likely to be combinations of foods rather than single items.” I also asked our classmate, Dr. Alan Beinstock, who is our resident nutritionist/anti-obesity activist of sorts, to comment on the article. He stated, “I subscribe to a Mediterranean diet saturated with fish, fresh vegetables and fruits, and olive oil and nuts is a natural and healthy way of being for it promotes and maintains a natural, healthy, and organic lifestyle.” And that folks, is good evidence.