Whole-Diet Approach, Mediterranean Diet, and Alzheimer Disease—Reply

In reply

Our study was conducted in a multiethnic urban cohort of New York City, which is unlikely to strictly consume the foods typical of Mediterranean countries. Therefore, and given the ethnic, geographic, and cultural differences in attitude toward olive oil consumption, it is not unexpected that the ratio of monounsaturated fatty acids to saturated fatty acids in our population was significantly lower as compared with that of Mediterranean populations. It is certainly possible that even higher ratios that exceed certain thresholds (such as the ones seen in Mediterranean populations) may have even more beneficial effects in cognition, but that does not negate the benefit of a Mediterranean-type diet in our cohort.

Use of nutrients in continuous forms has clear advantages, including higher power, more direct biological interpretations, and utilization of the data in their original form without use of transformations or cut-offs, which often contain subjective selection elements. However, potential disadvantages may also exist, such as possible underlying, nonnormal, skewed distribution (as for example in the case of alcohol that the authors themselves refer to¹ [a large proportion of subjects may need to be assigned a value of 0 since they do not consume it at all]) and absence of monotonic effects. Such potential disadvantages are often dealt with by categorization as in the case of the method used in the MeDi calculation. Clearly the choice of categorization can affect a subject's score assignment (median split being one of the commonest ones in medical literature), but Solfrizzi et al do not clarify (either in this correspondence or in their referenced letter to the editor) how exactly a possible resulting heterogeneity of variance between “high” and “low” groups could negatively affect the reliability and validity of the MeDi score.

Additionally, Solfrizzi et al do not clarify why this possible heterogeneity of variance between high and low groups could negatively affect transferability to other populations. We can appreciate that in our cohort the MeDi may vary from a true MeDi. Subjects with high MeDi adherence in New York may be potentially categorized as subjects with low MeDi adherence if viewed in comparison with Mediterranean populations, but their MeDi-attributable risk in relation to those populations may be difficult to calculate given the multiple genetic, environmental, and other differences among populations. That neither invalidates nor minimizes the significance of our finding since their relative AD risk in comparison with New Yorkers who are even further away from MeDi principles could be still lower.

Finally, transferability of the MeDi (calculated using similar methodology) advantages regarding mortality and other diseases has been clearly demonstrated in multiple previous studies.^{2 - 4} For example, adherence to the principles of the traditional MeDi has been associated with longer survival in both Anglo-Celtic individuals and Greek-Australian residents of Melbourne.³ In another study of 74 607 subjects in 9 European countries, increased adherence to the MeDi was associated with reduced mortality with no evidence of significant heterogeneity among countries in the association of MeDi with overall mortality.⁴

Given the known heterogeneity in diets across the globe, we fully realize that not all diets will fit the true MeDi, but if we can identify those dietary patterns that protect against diseases such as AD, we have made an important step in public health. We fully agree that successful replication of our findings regarding MeDi and AD in other populations (similarly to what has been already done for mortality^{2 - 4} ) would be more than desirable.

Correspondence: Dr Scarmeas, Columbia University Medical Center, 622 W 168th St, PH 19th Floor, New York, NY 10032 (ns257@columbia.edu).

Financial Disclosure: None reported.

Funding/Support: This study was supported by grants AG07232, AG07702, AG15294-06, 1K08AG20856-01, and RR00645 from the National Institute on Aging; the Charles S. Robertson Memorial Gift for Research in Alzheimer's Disease; the Blanchette Hooker Rockefeller Foundation; the New York City Council Speaker's Fund for Public Health Research; and the Taub Institute for Research on Alzheimer's Disease and the Aging Brain.