Mediterranean Diet, Physical

Activity Linked to Lower

Alzheimer’s Risk

Susan Jeffrey | 14th August 2009

Two new cohort studies come to slightly different conclusions about the merit of adherence to a Mediterranean-type diet on the risk for Alzheimer’s disease (AD).

In 1 study, researchers led by Nikolaos Scarmeas, MD, from Columbia University Medical Center in New York City, extend their previous findings, showing that both high adherence to the Mediterranean diet and higher levels of physical activity were independently associated with a lower risk for AD in a cohort of community-dwelling elders free of dementia at baseline.

In the other study, Catherine Féart, MD, from the Université Victor Ségalen Bordeaux 2 in Bordeaux, France, and colleagues, including Dr. Scarmeas, find that higher adherence to the Mediterranean diet was associated with slower cognitive decline when measured using the Mini-Mental State Examination (MMSE), but not when measured with other cognitive tests. No association was seen for adherence to the diet and the risk for incident dementia, although the researchers point out that the power to detect a difference was limited in this study.

Both reports are published in the August 12 issue of the Journal of the American Medical Association. In an editorial accompanying the publications, David Knopman, MD, from the Mayo Clinic in Rochester, Minnesota, assesses these results cautiously and points out that the benefits of an inherent tendency to healthy food choices in free-living individuals are most likely accrued over a lifetime and in concert with other kinds of healthy choices.

“I see diet as part of a larger pattern of healthy behaviors, and I see this result as supporting the idea that risk reduction for dementia begins at least in midlife, if not earlier,” he told Medscape Neurology. “I really believe that changing one’s dietary habits at age 70 is probably a good thing incrementally, but the greater benefit accrues the earlier a change in diet is adopted.”

Lower AD Risk With Diet?

The Mediterranean diet features a high intake of vegetables, legumes, fruits, and cereals; a high intake of unsaturated fatty acids, mostly in the form of olive oil; a low intake of saturated fatty acids; a moderately high intake of fish; a low to moderate intake of dairy products, mostly as cheese or yogurt; a low intake of meat or poultry; and finally, a regular but moderate amount of alcohol, usually wine, generally taken with meals.

Previous research has shown that following a Mediterranean diet is protective against a variety of conditions, including hypertension, coronary heart disease, dyslipidemia, diabetes, obesity, and certain cancers, and is related to a reduction in all-cause mortality in the general population. Previous work by Dr. Scarmeas and colleagues has shown that higher adherence to the Mediterranean diet is associated with a lower risk for AD, as well as prolonged survival in AD (Ann Neurol. 2006;59:912–921; Neurology. 2007;69:1084–1093).

Another, earlier, report from the Washington Heights-Inwood Columbia Aging Project (WHICAP) by this group, published in the February issue of the Archives of Neurology and reported by Medscape Neurology at that time, suggested that elderly subjects who followed a Mediterranean diet were less likely to develop mild cognitive  impairment and were also less likely to convert from mild cognitive impairment to AD (Arch Neurol. 2009;66:216–225).

The current report is an extension of these latter findings from WHICAP, this time looking at the relative contributions of physical exercise and the Mediterranean diet on the risk for AD.

Subjects in 2 cohorts totaling 1880 community-dwelling elderly people without dementia were followed up for a mean of 5.4 years, with standardized neurological and neuropsychological testing done every 1.5 years.

Adherence to a Mediterranean-type diet and amount of physical activity were derived from questionnaires and divided into low, middle, and high adherence to the diet and no, some, or much physical activity; all models were adjusted for a variety of factors including ethnicity, education, and apolipoprotein E genotype.

During follow-up, 282 incident cases of AD occurred. Compared with those participants with low adherence to the Mediterranean diet, those individuals with high adherence had a significantly reduced risk for AD. Similarly, those reporting much physical activity at baseline had a significantly reduced risk for AD vs those reporting no physical activity.

Those reporting no physical activity and low adherence to a Mediterranean-type diet had an absolute risk for AD of 19%, whereas those reporting much physical activity and high adherence diet scores had an absolute risk for AD of 12%.

Risk for AD by Adherence to Mediterranean Diet and Amount of Physical Activity

Low Diet Score Referent
Middle Diet Score 0.98 0.72 – 1.33
High Diet Score 0.60 0.72 – 0.80 .08
No Physical Activity Referent
Some Physical Activity 0.75 0.54 – 1.04 .08
Much Physical Activity 0.67 0.47 – 0.95 .03
High Diet Score Plus Much Physical Activity vs Low Diet Score and No Physical Activity 0.65 0.44 – 0.96 .03

Three-City Study

The second report by Dr. Féart and colleagues used data from the Three-City study, a prospective cohort study examining vascular risk factors for dementia in 1410 adults living in Bordeaux who were 65 years of age or older in 2001 to 2002. Adherence to the Mediterranean diet was again assessed using a food questionnaire, and cognitive performance was measured using the MMSE, Isaacs Set Test (IST), Benton Visual Retention Test (BVRT), and Free and Cued Selective Reminding Test (FCSRT). Assessments were done at baseline and at least 1 other time during 5 years of follow-up.

A total of 99 new cases of dementia were validated by an independent expert committee of neurologists. After adjustment for a variety of factors, the researchers found that a higher Mediterranean diet score was associated with fewer MMSE errors (β = .006; P = .04, for 1 point of the Mediterranean diet score), but not with performance on the other tests, particularly for those who remained free from dementia during 5 years.

However, adherence to the Mediterranean diet was not associated with the risk for incident dementia, although the researchers point out that their power to detect a difference on this endpoint was limited.

“The Mediterranean diet pattern probably does not fully explain the better health of persons who adhere to it, but it may contribute directly,” the authors speculate. “A Mediterranean diet also may indirectly constitute an indicator of a complex set of favorable social and lifestyle factors that contribute to better health. Further research is needed to allow the generalization of these results to other populations and to establish whether a Mediterranean diet slows cognitive decline or reduces incident dementia in addition to its cardiovascular benefits,” the authors conclude.

Findings to Be “Nibbled, Not Swallowed Whole”

In his editorial, Dr. Knopmen suggests that to say the Féart paper supports that of Scarmeas et al is “debatable,” pointing out that although the MMSE results in the Three-City study would seem to be in line with the WHICAP data, this was true only when it was considered as a continuous and not as a categorical variable.

“The lack of consistent association with the other cognitive measures, especially the FCSRT, is of concern if pre-AD pathology was the target of the Mediterranean-type diet,” he writes.

The studies reported in this issue, along with the earlier report by Scarmeas et al, provide only “moderately compelling evidence that adherence to the Mediterranean-type diet is linked to less late-life cognitive impairment,” Dr. Knopmen concludes. He cautions against the sort of “feeding frenzy” of media attention with these 2 studies that was seen after the initial report this year by Scarmeas et al, pointing out that the “nuanced science of these studies…should not be consumed so unabashedly.”

“The scientific value of these studies cannot be disputed, but whether or how they can or should be translated into recommendations for the public is the question,” Dr. Knopman writes. “For now, it is reasonable to nibble on these findings and savor them, but not to swallow them whole.”

The Washington Heights-Inwood Columbia Aging Project is supported by the National Institute on Aging. The authors have disclosed no relevant financial relationships. The Three-City Study is conducted under a partnership agreement between the Institut National de la Santé et del la Recherche Médicale (INSERM), the Institut de Santé Publique et Développement of the Victor Segalen Bordeaux 2 University, and Sanofi-Aventis. The authors have disclosed no relevant financial relationships. Dr. Knopman reports serving on a data and safety monitoring board for Sanofi-Aventis Pharmaceuticals (completed October 2008) and receiving personal compensation. Other disclosures appear in his editorial.