Arguably the most important Mediterranean diet study has been retracted and corrected, drawing the same, yet less robust conclusion that it creates a lower risk of having a stroke, heart attack, and dying from heart disease in comparison to a low-fat diet.
The Report Criticizing The Study
In 2017, a report written by John Carlisle questioned the random assignment of participants in 11 of 5,087 analyzed studies from the New England Journal of Medicine. However, after the journal reviewed the studies, the only study the criticism of which was legitimate was Martinez-Gonzalez’s 2013 Mediterranean diet study.
There were three diets in the study: the Mediterranean + extra-virgin olive oil diet, the Mediterranean + mixed nuts diet, or the control diet (a low-fat diet). Over 7,000 participants with a relatively high risk of cardiovascular disease were supposed to be randomly assigned to one of the three groups.
The point of this randomization is to make sure that there aren’t similar factors across many participants in one group that could also be responsible for an improved or worsened health outcome and not just the diets they were given. Carlisle insisted that there were a number of red flags that showed a low possibility the participants were actually randomly assigned.
The research team was urged to reassess the data and found that two of the concerns pointed out by Carlisle were legitimate, while the rest were “based on mistaken assumptions.”
The Retraction & Correction
Gonzalez found that 14% of the participants were not randomly assigned. For around 10% of them, identical diets were given to people living under the same roof. For example, people coming in together, such as married couples, were given the same diet, which meant that the assignment was no longer random. Because people living together share similar environments, and if they were family members share similar DNA, it would be impossible to tell whether their health outcomes were because of their assigned diets or because of those similar factors.
In one of the 11 locations, one researcher had assigned an entire village to the same diet, because they also wanted free olive oil after finding out their neighbors were receiving from the study.
However, Gonzalez drew the same conclusions from the revised study, but with a less rigid experimental backing. The new report no longer states that all participants in the study were randomly assigned.