The New England Journal of Medicine on Wednesday retracted a widely cited 2013 study on the health benefits of the Mediterranean diet—but it also published a new analysis of the study's data that reached similar conclusions.
What's in the Mediterranean diet?
According to NPR's "Shots," the Mediterranean diet is high in fruits and vegetables, olive oil, nuts, and fish. Research has suggested people who maintain a Mediterranean diet are typically healthier—and partly because of such research, the Mediterranean diet ranked as U.S. News' #1 'best diet' in 2018.
However, researchers have found it difficult to determine whether people who follow the diet are healthier because of what they eat or because of other factors.
The original study
Spanish researchers in 2013 published a landmark study in NEJM claiming that following the Mediterranean diet led people to have fewer heart attacks and strokes.
For the research, Miguel Martínez-González of the University of Navarra and his colleagues conducted a clinical trial examining the association between cardiovascular risk and the Mediterranean diet among 7,447 study participants, ages 55 to 80, who were at high cardiovascular risk but did not have cardiovascular disease.
To determine a cause-and-effect relationship between the Mediterranean diet and cardiovascular events and reduce bias, the researchers determined they would use randomization in the clinical trial. The researchers claimed to have randomly assigned the participants to three diets:
- A Mediterranean diet with an ounce of mixed nuts;
- A Mediterranean diet with at least four tablespoons a day of extra virgin olive oil; or
- A traditional low-fat diet.
The researchers originally found that, after a median follow-up period of five years, participants who ate the Mediterranean diet with olive oil were 30% less likely to experience a heart attack, stroke, or death from cardiovascular causes than people who ate a low-fat diet. They also found people who ate a Mediterranean diet with mixed nuts were 28% less likely to experience a heart attack, stroke, or death from cardiovascular causes than people who ate a low-fat diet.
Problems with randomization
But Martínez-González and his colleagues had to retract their original study after John Carlisle, a British anesthesiologist, flagged issues with randomization in the Mediterranean diet trial. Carlisle had been recognized previously for finding randomization errors in another researcher's work.
A statistician at NEJM recommended Martínez-González and his colleagues reassess their methods, and they did so, confirming that certain study participants enrolled in the clinical trial had not been randomly assigned to a diet.
In particular, 425 participants lived in the same household as a previously enrolled participant, and they were both assigned the same diet without randomization. Another 467 participants had their diets assigned by clinical site rather than individually. Specifically, the lead investigator at one of the 11 participating sites assigned an entire village the same diet without telling other investigators.
The randomization error "affected only a small part of the trial," Martínez-González said. He added that when researchers looked at the data in the population not affected by the error, the results were the same.
The re-analysis affirmed the original findings, but the researchers could not, as before, claim the Mediterranean diet directly caused the reduction in a heart attack, stroke, or death from cardiovascular causes, because not everyone in the clinical trial had been randomly assigned a diet, "Shots" reports.
The medical community had mixed responses to the updated paper.
Barnett Kramer, director of the division of cancer prevention at the National Cancer Institute, said, "Nothing they have done in this re-analyzed paper makes me more confident."
Donald Berry, a statistician at MD Anderson Cancer Center in Houston, said, "These people were naïve. They were sloppy and didn’t know they were being sloppy." Berry said he would like to believe the results of the new analysis, but he is not convinced, because the new analysis did not address the problems with the original study.
However, Steven Nissen, a cardiologist at the Cleveland Clinic, said he found the new analysis convincing and he plans to continue to advise his patients to go on the Mediterranean diet. Nissen said, when the findings had been first published, "I was thrilled to see what seemed to be an impeccable trial." He said it was "sobering" to learn about the trial's errors, but he said, "I was reassured that the conclusions are correct."
Martínez-González said, "After all this long work, I am more convinced than ever" by the results of the study. He added, "Seldom has a trial undergone more scrutiny"(Kolata, New York Times, 6/13; Phend, MedPage Today, 6/13; McCook, "Shots," NPR, 6/13; U.S. News & World Report, 6/15).
Help your employees promote healthy habits—regardless of the newest diet fads
Programs aimed at promoting healthy habits among employees are likely to lead to improved employee engagement and productivity—but they're unlikely to reduce the total cost of care. To do that, you'll need to take a population health approach.