Get Newsletter
AlzRisk Risk Factor Discussion

Risk Factor:
Risk Factor Type: Nutrition and supplements
Current Understanding:
Current Understanding The studies reviewed here generally provide support for an inverse relation between Alzheimer's disease and a healthy dietary pattern, defined loosely as a diet high in fruits and vegetables, low in red and processed meats, and favoring mono- and polyunsaturated fats over saturated fats. Although the studies varied somewhat in their definitions of healthy dietary pattern and used different methods to quantify adherence to that pattern, results were largely consistent. However, these studies have several methodological limitations that complicate their interpretation. Some studies adjusted for putative confounders like diabetes and cardiovascular disease that may in fact be causal intermediates, although effect estimates were similar with and without adjustment for these factors. No studies measured long-term dietary pattern and few measured dietary intake in mid-life, which are likely to be both more biologically relevant and less susceptible to reverse causation. Additional prospective studies addressing these limitations and evidence from randomized trials will help clarify the issue. In the meantime, a healthy dietary pattern, recommended for lowering risk of cardiovascular disease, may also lower risk of AD, particularly as part of a broader intervention including adequate physical activity and other lifestyle changes. For a review of the putative mechanisms by which dietary pattern may influence AD risk and detailed commentary on interpreting the findings below in a broader context, please view the Discussion.
Literature Extraction: Search strategy  * New *
Last Search Completed: 28 November 2016

Risk Factor Overview

Koyama A, Weuve J, Jackson JW, Blacker D. "Dietary pattern." The AlzRisk Database. Alzheimer Research Forum. Available at: Accessed [date of access]."

* * *


The tables in the Risk Factor Overview present studies investigating the association between a healthy dietary pattern and risk of Alzheimer’s disease (AD) and total dementia (TD). These studies include several observational studies, plus one clinical trial comparing a Mediterranean diet and a low-fat diet [1]. Although the studies varied in their definitions of a “healthy dietary pattern,” most emphasized high intakes of fruits and vegetables, low intake of red and processed meats, and mono- and polyunsaturated fats over saturated fats. Treatment of grains, alcohol, and added sugars varied, and a few studies defined healthy dietary pattern according to specific nutrients present in food consumed. Overall, these studies generally suggest that a healthy dietary pattern is associated with a reduced risk of AD and TD, but they must be interpreted in light of several methodological issues detailed below.

Mechanism of Action

Healthy dietary patterns might reduce AD risk through multiple mechanisms. In addition, several components of healthy dietary patterns are associated with a neuroprotective effect, either directly and through reducing risk of cardiovascular disease[1]. Cardiovascular benefits of these dietary patterns might include reduced weight (see Obesity) and decreased atherosclerosis and heart disease[2]. Healthy dietary pattern might also reduce AD risk by reducing diabetes risk[3] (see Diabetes). There is also suggestive evidence from a short-term pilot study that a diet high in saturated fats and a high glycemic index can alter AD-related proteins in the cerebrospinal fluid [4, 5]. Additional direct benefits vary by dietary component, and could involve non-specific effects on synaptic health as well as more specific effects at various points in the AD pathophysiologic process. Fatty fish, some nuts and some vegetable oils provide high levels of n-3 fatty acids, which in animal models are associated with synaptic health[6] and longer hippocampal neuron survival and a decrease in reactive oxygen species in the cortex and hippocampus[7, 8]. Fatty fish and various fortified foods provide vitamin D, which may protect against neurodegeneration through anti-inflammatory and antioxidant effects and through increased neurotrophin production[9]. Fruits and vegetables can provide high levels of vitamin C, vitamin E, flavonoids, carotenes, and other antioxidants that may have a neuroprotective effect through lower oxidative stress[10] and inflammation[11] (for a complete review, see Nutritional Antioxidants). Beans, nuts, whole grains, as well as some fruits and vegetables can be rich in phytoestrogens, which may the increase expression of antioxidant defense enzymes[12]. Moderate alcohol intake, included in some healthy diet definitions, in addition to its potential cardiovascular benefits, may protect neurons through other mechanisms such as an increase in hippocampal acetylcholine release[13] (for a complete review, see Alcohol).

Methodological Issues


The reviewed studies varied in how they defined healthy dietary patterns, how they conceptualized and measured adherence, and how they measured intake. They also differed somewhat in when they measured exposure in the lifespan. Each of the approaches used has inherent advantages and disadvantages in terms of capturing the aspects of a healthy dietary pattern that may be relevant to AD risk.

Definition of a healthy dietary pattern. Although the studies varied in their definitions of “healthy dietary pattern,” all of them used definitions that have evolved since older views that underscored low total and saturated fat. Rather, the studies reviewed directly or indirectly emphasized regular consumption of fruit, vegetables and fish, along with intake of foods or oils containing monounsaturated and polyunsaturated fats. Most studies designated red or white meat and dairy products as unhealthy. Treatment of nuts and grains varied, with some studies explicitly designating these as healthy and others ignoring them. Alcohol intake was considered healthy under varying definitions of moderate intake, or ignored. The definitions of a healthy dietary pattern were often based on food categories, but a few studies defined a healthy dietary pattern according to specific nutrients present in the food consumed. Some studies examined a Mediterranean dietary pattern, with a focus on olive oil in addition to standard elements like fruits and vegetables and fish rather than red meats[14].

Duration and timing of dietary intake. All of the included studies measured dietary intake only at baseline, which may only capture short-term and recent dietary habits. Long-term cumulative dietary exposures are probably more relevant to the development of AD, which occurs over the course of decades. Multiple measurements of dietary intake would better reflect long-term dietary habits and also allow flexibility for sensitivity analyses to assess the role of incident conditions (e.g., diabetes, cardiovascular disease) on changing dietary habits. Moreover, repeated measures of diet can reduce measurement error[15].

Of particular note, few studies assessed mid-life dietary intake (e.g. in participants’ 50s), which may be more biologically meaningful[16]. Furthermore, measurements of dietary intake later in life may be subject to reverse causation, as preclinical cases or participants with other comorbidities of aging may change dietary habits, leading to biased estimates of association. Late-life measurements of dietary intake may also be particularly susceptible to reporting error due to existing functional and/or mild cognitive impairment[17, 18].

Measurement of Dietary Intake. Among the observational studies reviewed, the majority used a food frequency questionnaire (FFQ) to record participants’ dietary habits. One study used a set of qualitative and quantitative survey questions and another used a 24-hour dietary recall. Most but not all of the dietary assessments had been previously validated. However, the 24-hour recall method only captures very short-term dietary intake. This approach, as well as unvalidated approaches querying about any time period, are more likely to lead to non-differential misclassification of long-term exposure.

Determination and scaling of dietary patterns. Although the reviewed studies generally agreed as to which foods were and were not considered part of a healthy dietary pattern, the actual method of quantifying the healthiness of participants’ diets differed substantially across studies. Two primary approaches utilize either “a priori” or “a posteriori” methods. The “a priori” methods are based on predefined diet patterns, such as the Mediterranean diet score[14]. In contrast, the “a posteriori” methods utilize factor analytic approaches to dissect patterns naturally existing within the specific population under study, and then create scores reflecting adherence to those patterns. This method avoids the imposition of diet patterns on populations that may not adhere to that particular pattern (e.g. U.S. populations may not have wide distributions of the Mediterranean diet score, since few practice this diet). However, the “a posteriori” methods may yield patterns that are difficult to compare across cohorts because they strictly describe and quantify the diet of the population under study[19]. A more recent approach, called reduced rank regression, applies “a priori” knowledge of risk factors (e.g., nutrients previously associated with AD) to “a posteriori” examination of a specific population’s diet. Reduced rank regression does not find or describe naturally occurring diet patterns of those under study but rather explains variation within that population in nutrients believed to be biologically relevant to AD.

The observational studies reviewed here all compared higher intake of foods considered healthy and lower intake of foods considered harmful compared to the inverse pattern of intake. Overall, the consistency of the collective evidence despite the heterogeneity of approaches to exposure lends additional support for a protective role of a healthy dietary pattern.

Collective influence of dietary patterns vs. the influence of specific foods and food groups. Regardless of the method used to quantify a healthy dietary pattern, the majority of the included studies did not conduct further analysis on individual foods or food categories that may have driven the observed association. Furthermore, an inherent limitation of those studies using healthy dietary pattern scores is that a given score can be achieved through different consumption frequencies and combinations of food categories that comprise the score. For example, in some studies, a favorable fat intake profile emphasized oils containing polyunsaturated fats, whereas other settings emphasized olive oil (primarily composed of monounsaturated fats). Therefore, although a dietary pattern can be the basis of a more realistic intervention compared with individual nutrients, it may be difficult to pinpoint the mechanisms and interactions among foods that convey the optimal benefit for AD risk.

Design and Analysis

Confounding and intermediate variables. Confounding by socioeconomic status (SES) cannot be discounted, as SES can be strongly associated with both dietary pattern and risk of AD[20, 21]. The majority of reviewed studies included years of education as a covariate in regression models, either as a continuous variable or in 3 to 4 categories, but this does not fully account for the potential impact of employment and income, cognitive activity, receptivity to health education, or access to health services[22-26]. It also does not address quality of education, which may be particularly heterogeneous within and across the many countries represented in the reviewed studies. Even with more comprehensive adjustment for SES, residual confounding may still remain due to measurement error, coarse categorization of socioeconomic factors, and use of aggregate data[27].

Physical activity is another potential source of confounding, as it can be associated with a healthy dietary pattern and appears to be an independent protective factor for AD[28, 29]. Most studies adjusted for self-reported physical activity, or for total energy intake (for which physical activity is a major determinant[30]). However, neither of these can accurately quantify the different dimensions of physical activity such as type, intensity, and duration, leading to a high likelihood of residual confounding. In addition, individuals may take up physical activity and healthy diet as part of a broader pattern of health-seeking behavior that reduces risks for cardiovascular disease and AD.

Some of the reviewed studies adjusted for cardiovascular risk factors. These factors are especially challenging to manage in the context of an observational study because dietary pattern can influence cardiovascular risk, thus making cardiovascular factors likely intermediaries in the relation of healthy dietary pattern to AD risk. Individuals may also change their dietary pattern (and other cardiovascular risk-related behaviors like physical activity) in response to cardiovascular risk. Thus, establishing the temporal order between diet and the participants’ knowledge of their cardiovascular risk is important to determine whether to adjust for cardiovascular risk factors in analyses of dietary pattern and AD risk. Findings within the reviewed studies were similar with or without adjustment for cardiovascular factors. Additionally, a related study specifically investigated the role of vascular mediation by adjusting for several vascular variables with no change in effect sizes [31]. However, it is possible that a lack of attenuation of the diet-AD association with adjustment for cardiovascular risk factors could represent actual mediation offset by measurement error, unmeasured confounders of cardiovascular risk and AD, and effect modification [32].

Results from Other Lines of Research

Numerous prospective studies have examined the association between a healthy dietary pattern and later cognitive function, with overall findings less consistent than the reviewed studies assessing incident AD or TD. Most observational studies measuring dietary intake in mid-life and cognitive function in late life have reported a positive association between a healthy dietary pattern and cognitive function[33, 34], while one study measuring diet in late life did not report any significant findings[35]. Observational studies addressing change in cognitive function over time have not been consistent, showing both a possible benefit of a healthy dietary pattern[36-39] as well as null findings[33, 40, 41].

In one randomized clinical trial (RCT), participants who were randomized to a Mediterranean diet emphasizing olive oil or nut consumption had a significantly lower risk of TD and better cognitive performance compared to participants randomized to a low-fat diet after six years of follow-up[42]. Other randomized trials have investigated the effect of a healthy dietary pattern on short-term change in cognitive test performance. One small trial showed a borderline significant effect of reduced cognitive decline over a 24-month period among initially cognitively normal retirement home residents, although the association was attenuated at 33 months, and randomization did not appear to be successful[43]. In a secondary prevention trial of cardiovascular disease, participants in midlife were randomized to the Dietary Approaches to Stop Hypertension (DASH) diet in combination with weight management therapy, the DASH diet alone, or a usual diet as a control[44]. Participants in both intervention arms exhibited better cognitive performance compared with the control group after 4 months, with a stronger effect in the combined diet and weight management arm.

Discussion and Recommendations

Overall, the collective findings from the studies reviewed suggest a protective effect of a healthy dietary pattern on risk of AD. However, the interpretation of many studies may be complicated by changes to diet in response to chronic disease or late-life retrospective assessment of long-term dietary intake. Regardless, a healthy dietary pattern is an established modifiable factor to lower risk of cardiovascular disease, and the available evidence suggests that it may also lower risk of AD, whether by cardiovascular mechanisms or through independent effects on AD pathology [1]. Any dietary intervention may be most effective as part of a larger lifestyle intervention including adequate physical activity and other healthful behaviors. Additional studies addressing the methodological issues in current studies and larger randomized trials may provide more robust support for a role for dietary pattern and other cardiovascular protective factors in reducing AD risk.


1. Frisardi, V., et al., Nutraceutical properties of Mediterranean diet and cognitive decline: possible underlying mechanisms. J Alzheimers Dis, 2010. 22(3): p. 715-40.

2. Rees, K., et al., 'Mediterranean' dietary pattern for the primary prevention of cardiovascular disease. Cochrane Database Syst Rev, 2013. 8: p. CD009825.

3. Barnard, N.D., et al., Dietary and lifestyle guidelines for the prevention of Alzheimer's disease. Neurobiol Aging, 2014.

4. Blacker, D., Food for thought. JAMA Neurol, 2013. 70(8): p. 967-8.

5. Hanson, A.J., et al., Effect of apolipoprotein E genotype and diet on apolipoprotein E lipidation and amyloid peptides: randomized clinical trial. JAMA Neurol, 2013. 70(8): p. 972-80.

6. Wurtman, R.J., Non-nutritional uses of nutrients. Eur J Pharmacol, 2011. 668 Suppl 1: p. S10-5.

7. Hashimoto, M., et al., Protective effects of prescription n-3 fatty acids against impairment of spatial cognitive learning ability in amyloid beta-infused rats. Food Funct, 2011. 2(7): p. 386-94.

8. Wang, P.Y., J.J. Chen, and H.M. Su, Docosahexaenoic acid supplementation of primary rat hippocampal neurons attenuates the neurotoxicity induced by aggregated amyloid beta protein(42) and up-regulates cytoskeletal protein expression. J Nutr Biochem, 2010. 21(4): p. 345-50.

9. Wrzosek, M., et al., Vitamin D and the central nervous system. Pharmacol Rep, 2013. 65(2): p. 271-8.

10. Nones, J., et al., Flavonoids and astrocytes crosstalking: implications for brain development and pathology. Neurochem Res, 2010. 35(7): p. 955-66.

11. Shukitt-Hale, B., F.C. Lau, and J.A. Joseph, Berry fruit supplementation and the aging brain. J Agric Food Chem, 2008. 56(3): p. 636-41.

12. Borras, C., et al., Genistein, a soy isoflavone, up-regulates expression of antioxidant genes: involvement of estrogen receptors, ERK1/2, and NFkappaB. FASEB J, 2006. 20(12): p. 2136-8.

13. Fadda, F. and Z.L. Rossetti, Chronic ethanol consumption: from neuroadaptation to neurodegeneration. Prog Neurobiol, 1998. 56(4): p. 385-431.

14. Trichopoulou, A., et al., Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med, 2003. 348(26): p. 2599-608.

15. Willett, W., Nutritional epidemiology. Third edition. ed. Monographs in epidemiology and biostatistics. 2013, Oxford ; New York: Oxford University Press. ix, 529 pages.

16. Reiman, E.M., et al., Brain imaging and fluid biomarker analysis in young adults at genetic risk for autosomal dominant Alzheimer's disease in the presenilin 1 E280A kindred: a case-control study.
Lancet Neurol, 2012. 11(12): p. 1048-56.

17. Pope, S.K., et al., Cognitive ability is associated with suspected reporting errors on food frequency questionnaires. J Nutr Health Aging, 2007. 11(1): p. 55-8.

18. Lafay, L., et al., Determinants and nature of dietary underreporting in a free-living population: the Fleurbaix Laventie Ville Sante (FLVS) Study. Int J Obes Relat Metab Disord, 1997. 21(7): p. 567-73.

19. Feart, C., et al., Potential benefits of adherence to the Mediterranean diet on cognitive health. Proc Nutr Soc, 2013. 72(1): p. 140-52.

20. Darmon, N. and A. Drewnowski, Does social class predict diet quality? Am J Clin Nutr, 2008. 87(5): p. 1107-17.

21. Meng, X. and C. D'Arcy, Education and dementia in the context of the cognitive reserve hypothesis: a systematic review with meta-analyses and qualitative analyses. PLoS One, 2012. 7(6): p. e38268.

22. Bennett, D.A., et al., Cognitive and social lifestyle: links with neuropathology and cognition in late life. Acta Neuropathol, 2014. 127(1): p. 137-50.

23. Davey Smith, G., et al., Education and occupational social class: which is the more important indicator of mortality risk? J Epidemiol Community Health, 1998. 52(3): p. 153-60.

24. Kaup, A.R., et al., Older Adults With Limited Literacy Are at Increased Risk for Likely Dementia. J Gerontol A Biol Sci Med Sci, 2013.

25. Kelleher, J., Cultural literacy and health. Epidemiology, 2002. 13(5): p. 497-500.

26. Sattler, C., et al., Cognitive activity, education and socioeconomic status as preventive factors for mild cognitive impairment and Alzheimer's disease. Psychiatry Res, 2012. 196(1): p. 90-5.

27. Kaufman, J.S., R.S. Cooper, and D.L. McGee, Socioeconomic status and health in blacks and whites: the problem of residual confounding and the resiliency of race. Epidemiology, 1997. 8(6): p. 621-8.

28. Osler, M., et al., Dietary patterns and mortality in Danish men and women: a prospective observational study. Br J Nutr, 2001. 85(2): p. 219-25.

29. Scarmeas, N., et al., Physical activity, diet, and risk of Alzheimer disease. JAMA, 2009. 302(6): p. 627-37.

30. Colbert, L.H., et al., Intensity of Physical Activity in the Energy Expenditure of Older Adults. J Aging Phys Act, 2013.

31. Scarmeas, N., et al., Mediterranean diet, Alzheimer disease, and vascular mediation. Arch Neurol, 2006. 63(12): p. 1709-17.

32. Kaufman, J.S., R.F. Maclehose, and S. Kaufman, A further critique of the analytic strategy of adjusting for covariates to identify biologic mediation. Epidemiol Perspect Innov, 2004. 1(1): p. 4.

33. Samieri, C., et al., Long-term adherence to the Mediterranean diet is associated with overall cognitive status, but not cognitive decline, in women. J Nutr, 2013. 143(4): p. 493-9.

34. Kesse-Guyot, E., et al., Adherence to nutritional recommendations and subsequent cognitive performance: findings from the prospective Supplementation with Antioxidant Vitamins and Minerals 2 (SU.VI.MAX 2) study. Am J Clin Nutr, 2011. 93(1): p. 200-10.

35. Psaltopoulou, T., et al., Diet, physical activity and cognitive impairment among elders: the EPIC-Greece cohort (European Prospective Investigation into Cancer and Nutrition). Public Health Nutr, 2008. 11(10): p. 1054-62.

36. Tsivgoulis, G., et al., Adherence to a Mediterranean diet and risk of incident cognitive impairment. Neurology, 2013. 80(18): p. 1684-92.

37. Cadar, D., et al., The role of lifestyle behaviors on 20-year cognitive decline. J Aging Res, 2012. 2012: p. 304014.

38. Tangney, C.C., et al., Adherence to a Mediterranean-type dietary pattern and cognitive decline in a community population. Am J Clin Nutr, 2011. 93(3): p. 601-7.

39. Wengreen, H.J., et al., Diet quality is associated with better cognitive test performance among aging men and women. J Nutr, 2009. 139(10): p. 1944-9.

40. Samieri, C., et al., Mediterranean diet and cognitive function in older age. Epidemiology, 2013. 24(4): p. 490-9.

41. Cherbuin, N. and K.J. Anstey, The Mediterranean diet is not related to cognitive change in a large prospective investigation: the PATH Through Life study. Am J Geriatr Psychiatry, 2012. 20(7): p. 635-9.

42. Martinez-Lapiscina, E.H., et al., Mediterranean diet improves cognition: the PREDIMED-NAVARRA randomised trial. J Neurol Neurosurg Psychiatry, 2013. 84(12): p. 1318-25.

43. Kwok, T.C., et al., A randomized controlled trial of dietetic interventions to prevent cognitive decline in old age hostel residents. Eur J Clin Nutr, 2012. 66(10): p. 1135-40.

44. Smith, P.J., et al., Effects of the dietary approaches to stop hypertension diet, exercise, and caloric restriction on neurocognition in overweight adults with high blood pressure. Hypertension, 2010. 55(6): p. 1331-8.