Consider browsing this week’s JAMA over a glass of wine—not just because two papers seem to tout the Mediterranean diet, but because it may take time to digest and appreciate their nuanced conclusions. Both suggest that eating the Mediterranean way—emphasizing fruits, vegetables, nuts, grains, and olive oil, with some fish and wine—may help preserve late-life cognition. However, one paper shows the benefit pretty clearly, while the other has more ambiguous results.

In the first offering, Nikolaos Scarmeas and colleagues at Columbia University Medical Center, New York, found that both diet and physical activity independently associated with lower rates of Alzheimer disease in 1,880 New York City elderly who were followed for more than 14 years. For the second study, with a median follow-up of 4.1 years, researchers analyzed 1,410 seniors in Bordeaux, France, and found that Mediterranean diet devotees had better scores on the Mini-Mental State Examination (MMSE), but not on three other neuropsychological tests. In that analysis—done by Catherine Féart, INSERM U897, Bordeaux, and colleagues (including Scarmeas)—higher adherence to the diet did not appear to correlate with reduced dementia risk.

Each investigation built upon earlier work by Scarmeas’ group suggesting that maintaining a Mediterranean diet could help protect against AD (Scarmeas et al., 2006 and ARF news story) and mild cognitive impairment (Scarmeas et al., 2009). The New York study followed the same cohort (from Washington Heights, a neighborhood in north Manhattan) but addressed physical activity in relation to diet, to see whether each could separately influence AD risk. Scarmeas and colleagues found the expected correlation between diet and exercise habits: those who exercised more tended to consume a more Mediterranean-style diet, and those who were less physically active generally had poorer diets.

However, when it came to how these habits affected AD risk, “there was an effect of physical activity over and above somebody's diet, and there was an effect of diet over and above physical activity,” Scarmeas told ARF. In other words, “if you took two subjects and both were physically active, the person who was also following a healthier diet had higher protection.” The independent effects of diet and exercise on AD risk remained even after the researchers controlled for a laundry list of potential confounders—including gender, ethnicity, education, apolipoprotein E (ApoE) genotype, smoking, depression, total caloric intake, and body-mass index. A number of studies have linked lower AD risk with increased physical activity (see, e.g., Lautenschlager et al., 2008 and ARF news story), but very few have comprehensively examined both diet and exercise as independent contributors.

The French scientists did not include a physical activity component in their study but rather designed it to see whether the earlier findings of Scarmeas’s lab (Scarmeas et al., 2006)—the only study to date that had looked at Mediterranean diet and AD risk—could hold up in a different population. Féart and colleagues analyzed dementia-free elderly living in Bordeaux, in southwest France. They measured cognitive ability using four neuropsychological tests: the MMSE (global cognition), Isaacs Set Test (verbal fluency and speed), Benton Visual Retention Test (working memory), and Free and Cued Selective Reminding Test (episodic memory). Compared to weaker adherents, those who had stuck more closely to a Mediterranean diet made fewer mistakes on the MMSE, but did no worse on the other three cognitive measures and did not seem to develop dementia any more frequently.

Taking into account both JAMA reports, “I don’t think these studies are particularly inconsistent,” said Deborah Blacker, a geriatric psychiatrist and epidemiologist at Massachusetts General Hospital, Charlestown, who was not involved in either study. “One has modest power to detect an effect and get the signal, but it’s pretty garbled. The other gets a pretty clear signal.”

Blacker and others, including the study authors, point out several features that may underlie the apparent differences in what the two investigations conclude. “The first thing that absolutely leaps off the page is that there’s a much larger number of AD cases and vastly more follow-up time in the New York study,” Blacker told ARF. While both cohorts had comparable average age at baseline (77.2 years in New York, 75.9 years in Bordeaux), most of the French participants were re-examined once within five years (median follow-up 4.1 years), whereas the New York group was examined every 1.5 years over a 14-year period (1992-2006). “It means the people got that much older,” Blacker said. “The more time you have, the greater your risk [for developing AD].”

Over the course of follow-up, 99 of 1,410 seniors in the French study developed dementia, of whom 66 had AD. In the New York study, 262 people succumbed to AD. “They had four times fewer people who developed AD,” Scarmeas said of the French study. “This was part of the reason they could not show an association between this diet and risk for AD. They could only show an association with rates of cognitive decline.”

The methods used to assess dietary behavior may have also influenced the studies’ findings. Both categorized participants into tertiles (low, middle, or high adherence) based on how often they said they consumed foods from different components of the Mediterranean diet (e.g., vegetables, fruits, fish, alcohol). The researchers assigned a value of 0 or 1 to each food category depending on whether consumption was below or above the median, respectively, and summed these sub-scores to obtain a total diet score ranging from 0 to 9 for each participant. The key here is that dietary adherence is “not based on absolute values but on relative values within the local culture,” Blacker said. “Whatever that 0 to 9 means is something different in Bordeaux than in Washington Heights.”

Féart and colleagues also acknowledge this. They note that French people tend to eat more fruits and vegetables than people in the United States (Tamers et al., 2009). “Therefore, some overlap between ‘low’ and ‘high’ French and U.S. consumers of fruits and vegetables could occur: the low French consumers may be high U.S. consumers,” they write. “This may have added misclassification, leading to a decreased chance to observe an association between diet adherence and cognitive decline or incidence of dementia in our study.”

Despite these potential caveats, the two JAMA studies, along with earlier findings in the same New York cohort, seem to offer “moderately compelling evidence that adherence to the Mediterranean-type diet is linked to less late-life cognitive impairment,” David Knopman of Mayo Clinic, Rochester, Minnesota, writes in an accompanying editorial. The New York study comes with the added bonus of confirming the independent benefit of exercise. Scarmeas emphasized, though, that because the collective evidence comes from observational studies—not intervention studies or clinical trials—“we cannot say to patients, ‘Exercise and eat well, and you will be protected from AD.’” Still, he said “it may be worth trying to pursue both to the extent that your risk for getting the disease is not determined purely by genes but is partially affected by lifestyle and behavior.” Additional research is required to confirm that the findings from the New York cohort can in fact be extrapolated to other populations.

Those taking that advice to heart may want to check out another recent study on the Mediterranean diet. In this paper, published 23 June in the British Medical Journal, researchers at the Harvard School of Public Health, Boston, and at the University of Athens, Greece, report that certain components, more so than others, seem to confer most of the regimen’s longevity benefits. According to this study, eating lots of fruits, vegetables, nuts, legumes, and olive oil while curbing red meat and enjoying moderate amounts of alcohol appears more important than consuming grains and fish and minimizing dairy (Trichopoulou et al., 2009). And for more on the “you are what you eat” theme, take note of a report published online last week in the journal Dementia and Geriatric Cognitive Disorders (Solomon et al., 2009). This study of 9,844 people confirmed that high serum cholesterol levels at midlife correlates with increased risk of AD and vascular dementia 30 years later.—Esther Landhuis


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. Mediterranean Diet Slims Down Risk of AD
  2. Work Up a Sweat to Stay Sharp, Randomized Trial Suggests

Paper Citations

  1. . Mediterranean diet and risk for Alzheimer's disease. Ann Neurol. 2006 Jun;59(6):912-21. PubMed.
  2. . Mediterranean diet and mild cognitive impairment. Arch Neurol. 2009 Feb;66(2):216-25. PubMed.
  3. . Effect of physical activity on cognitive function in older adults at risk for Alzheimer disease: a randomized trial. JAMA. 2008 Sep 3;300(9):1027-37. PubMed.
  4. . US and France adult fruit and vegetable consumption patterns: an international comparison. Eur J Clin Nutr. 2009 Jan;63(1):11-7. PubMed.
  5. . Anatomy of health effects of Mediterranean diet: Greek EPIC prospective cohort study. BMJ. 2009;338:b2337. PubMed.
  6. . Midlife serum cholesterol and increased risk of Alzheimer's and vascular dementia three decades later. Dement Geriatr Cogn Disord. 2009;28(1):75-80. PubMed.

Further Reading


  1. . Mediterranean diet and risk for Alzheimer's disease. Ann Neurol. 2006 Jun;59(6):912-21. PubMed.

Primary Papers

  1. . Mediterranean diet and mild cognitive impairment. Arch Neurol. 2009 Feb;66(2):216-25. PubMed.
  2. . Adherence to a Mediterranean diet, cognitive decline, and risk of dementia. JAMA. 2009 Aug 12;302(6):638-48. PubMed.
  3. . Mediterranean diet and late-life cognitive impairment: a taste of benefit. JAMA. 2009 Aug 12;302(6):686-7. PubMed.