The last week in February (2013) saw headlines all over the world: “Mediterranean diet shown to ward off heart attack and stroke.” The Guardian ran with “Mediterranean diet ‘cuts strokes and heart attacks in at-risk groups‘.” The Sydney Morning Herald announced “Mediterranean diet cuts risk of first heart attack by 30%”.
The world headlines were all based on this article in the highly respected New England Medical Journal. The researchers’ own headline was “Primary Prevention of Cardiovascular Disease with a Mediterranean Diet”. The study is known as PREDIMED (Prevención con Dieta Mediterránea).
Someone I follow on twitter – Rob Lyons (@robspiked) captured it beautifully when I was having my usual rant about people not knowing what the real Mediterranean diet is. Rob replied: “@zoeharcombe no, that’s the diet Mediterraneans eat. This is the ‘Mediterranean Diet’, a mythological diet invented by US researchers.” How true!
Let’s get the facts on the table first. The study involved 7,447 people. 57% were women. The women were aged 60 to 80 and the men were aged 55 to 80. The Guardian was right that they were at-risk groups, as only people with type 2 diabetes or at least three other “major risk factors” (smoking, obesity, family history of heart disease etc) were included in the study. The Sydney Morning Herald was also right about the first heart attack, as only people with no cardiovascular disease at enrollment were admitted to the study.
This is where we need Rob in our minds. The real Mediterranean diet is high in meat (if it moves, it is eaten – rabbit, pork, beef, chicken, turkey, game, snails etc); fish; cheese; eggs; cream; vegetables & salads; fruits in season and white grains (white bread, white rice, white pasta). Those who eat more of the real food are slim. Those who eat more of the pasta become “Italian mammas”.
The first reference in the study is thus wrong. It claims “The traditional Mediterranean diet is characterized by a high intake of olive oil, fruit, nuts, vegetables, and cereals; a moderate intake of fish and poultry; a low intake of dairy products, red meat, processed meats, and sweets; and wine in moderation, consumed with meals.”
This is what Americans, who have never been to the Med, fantasise that the Mediterranean diet is. However, the truth is that the French/Italians/Greek etc are eating their body weight in red meat and cheese and “wine in moderation” would have an Italian rolling in the aisles. Here are the top wine drinking countries in the world. The prime Mediterranean countries, France and Italy, are in the top five. But for the staggering consumption of the Vatican City, they would be higher! ;-)
Anyway – allowing for the fact that American researchers don’t know what the real Mediterranean diet is, let’s see what the study diet actually was. The 7,447 people were divided into three groups. The study says that they were randomly assigned so, by luck, they appear to have been fairly equally distributed between groups so that there are no obvious confounding differences where, for example, one group has ended up with more than its share of smokers, older people, obese people etc. For once, we can look at the diet as the primary difference to observe.
Two groups were put on this Fictitious Mediterranean Diet (FMD from now on) and the control group were put on a low-fat diet. The two groups on the FMD were also told to avoid soda drinks, bakery goods, spreads, red and processed meat (apart from the red meat, this is excellent advice). The low-fat diet group was told to have at least three servings a day of bread, pasta, potatoes, rice etc – those nicely fattening products that raise triglyceride levels and damage arteries. The FMD group were told to have oily fish. The low-fat group were told to avoid it.
Group 1 doing the FMD was given additional olive oil and group 2 was given 30g of mixed nuts per person per day. The article details that one litre of extra-virgin olive oil was given to group 1 each week and that they were encouraged to consume 50g or more per day. The nut group were given 15g walnuts, 7.5g hazelnuts and 7.5g almonds daily. That’s at least 440 extra calories with virtually no nutrients for the olive oil group (some vitamin E and K but nothing else). The nut group would get approximately 200 extra calories with far more nutrients. Nuts have virtually every vitamin and mineral – many in good amounts. They have protein (olive oil doesn’t – it’s a pure fat).
The fat content in both oil (100%) and nuts (c. 65%) is huge – way higher than the 7% fat content of sirloin steak or the 4% fat content of pork chops. This was a daft study. Why tell groups to avoid natural fat in real food (meat, dairy, eggs) and to replace it with very high fat interventions? (We’ll see why when we see the funders of the study!)
The objective of the study was to measure “primary end points” defined as a major cardiovascular event – a heart attack, stroke or death.
There were 288 such incidents – 96 in the olive oil group, 83 in the nut group and 109 in the low-fat (current government dietary advice) group.
When the incidents in each group were presented as a percentage of person years (i.e. number of people in each group times the average years of follow-up), the incident rate was 0.81% for the olive oil group; 0.80% for the nut group and 1.12% for the low-fat group. As ever, the headlines are big, the real numbers are small. Barely 1 person in 100, in the highest risk groups for cardiovascular disease, at the worst possible age to have cardiovascular disease, actually had an incident during the 5 year study. For the nut and oil groups it was slightly less than 1 person in 100, for the low-fat group it was slightly more than 1 person in 100.
When the incident rate numbers are weighted so that the control group is 1.00, the oil group is 0.73 and the nut group is 0.72. This is relative risk, not absolute. It’s the oldest trick in the book to play to make numbers seem far more dramatic than they are. This is where the headline nonsense of “30% lower risk” comes from. Plus, we always need to remember that this is association, not causation.
The trial was intended to last 6 years. The researchers stopped at 4.8 years – this is usually positioned as “the differences were just so great that we could not morally continue to disadvantage the control group any longer”. A couple of points on this:
1) look at the graph on p8 of the New England Medical Journal paper. The gap between the intervention and control groups is starting to close at the point the experiment is stopped. I’ve seen this convenient ending of trials before.
2) the disadvantage for the control group is that they are following a low-fat diet – not that they are missing out on nuts and/or olive oil. That’s why the headline of the article should have been: “Low-fat diets are associated with cardiovascular disease” Which the media would have likely reported as: “Low-fat diets cause heart disease”. (Except they wouldn’t because the media just loves a good old eulogy about the Fictitious Mediterranean Diet.)
The bottom line
This could have been a very useful study. It could have been an original study to prove that our current low-fat, high-carb, starchy food, diet advice is doing more harm than good. It was essentially comparing a low-fat diet with a real food diet, but with an unnecessary messing around of natural fat delivery mechanisms (being told to avoid natural fat in meat and dairy and to replace it with unnatural levels of olive oil/nuts). Given the nutrition in nuts vs. olive oil, I was surprised that there was no difference between these two groups. This also reinforces the fact that the difference observed was about the low-fat diet being bad and not olive oil or nuts being good.
So why not just do a straight study comparing real food with its natural fat intact and our fake food/low-fat government dietary advice? Who would fund such a study? The funders of this study included: Hojiblanca and Patrimonio Comunal Olivarero (extra-virgin olive oil); the California Walnut Commission; Borges (almonds) and La Morella Nuts (hazelnuts). In the really, really, small print at the end of the article, we also discover that the author conflicts note: “Dr. Ros serves on the board of ‘his institution’ – the California Walnut Commission”; “Dr. Salas-Salvadó is on the board of and receives grant funding from ‘his institution’ – the International Nut and Dried Fruit Council”; Dr. Lamuela-Raventos receives funding from PepsiCo – their snack division does nuts; and Dr. Serra-Majem reports serving on the boards of the Mediterranean Diet Foundation (I never knew there was such a thing!)
As ever – follow the money and all shall be explained!