This site is intended for health professionals only

At the heart of general practice since 1960

Can eating fish reduce the risk of dementia?

Lianna Ishihara and Dr Carol Brayne weigh up the evidence

The possible neuroprotective role of n-3 polyunsaturated fatty acids (PUFA) for dementia and Alzheimer's disease is a promising hypothesis that has been addressed by few epidemiological studies to date. N-3 PUFAs are found in some fish; they play a role in neuronal membrane fluidity and also possibly in memory. They may reduce free radical production and/or lower serum cholesterol and triglyceride levels, but their exact function is not known1,2,3.


We conducted a systematic MEDLINE search using a combination of key words related to dementia/Alzheimer's and fish oil.

Five epidemiological studies addressing the topic in adults were available in the literature. Of these, one was a cross-sectional study and four were prospective cohort studies (measured intake before the onset of dementia).

N-3 PUFAs are most plentiful in fatty fish such as salmon, tuna and mackerel, so studies investigating the hypothesis have used fish consumption either as the exposure variable or as a proxy measure of n-3 PUFA intake.

Fish intake was most commonly assessed by a food frequency questionnaire, which requires individuals to report their own food consumption over a specified period of time (tuna sandwich, fresh fish)4,5,6.

Another approach was simply to ask how often fish was consumed (daily, weekly)7.

The main difficulties with such measures are the possibility of inaccurate recall – particularly in the presence of cognitive impairment – changes in dietary patterns over time or in relation to onset of dementia, and differences in types or preparations of fish.

The detection and definition of dementia varies across the studies. The mini mental state examination (MMSE) is routinely used to measure cognitive decline and detect suspected dementia, and it is used in most of the studies. Dementia can be diagnosed using various methods, including written tests, interview or examination by a specialist.

Different criteria exist for diagnosing dementia and Alzheimer's; they involve clinical assessment with or without imaging and rarely with neuropathological confirmation.

Other studies

The hypothesis was also investigated by a cross-sectional study in Madrid8. This is the weakest evidence in terms of design because the time relationship between exposure and outcome is uncertain. No significant difference in fish intake was found between MMSE score groups within each age category, although in the 75-plus years category fish intake was lower in the group with MMSE scores below 28.

One cohort study examined cognitive decline only9 and others examined cognitive decline and a clinical diagnosis of dementia. The studies were all population based with moderate baseline measures; follow-up ranged from two to seven years; analyses were adjusted for a variety of factors. Apart from the six-year follow-up of the Dutch Rotterdam study, all studies were in the direction of a protective relationship.


The best conducted studies were from Rotterdam6 and Chicago4 because they used quantitative measures of fish intake, follow-up time was sufficient to obtain cases, and dementia and AD were diagnosed by neurologists using standardised criteria. However, the results of the two studies are contradictory. A possible explanation lies in the analytic methods: the Rotterdam analysis defined n-3 PUFA intake as a continuous variable in g/day compared with categories of fish intake.

Difficulties in interpreting the results include the biological meaning of an infrequent dietary exposure, leaving open the possibility that fish intake may be indirectly related to dementia risk.

The impact of publication bias is unknown, but if studies with non-significant or negative results are less likely to be published then any positive association between fish intake and reduced incidence of dementia could be exaggerated.


lAlthough several studies have addressed the question whether fish intake can reduce the risk of dementia, the verdict remains open

lThe general trend in published studies is for an association between increased fish intake and reduced incidence of dementia; these studies include three cohort studies with significant results

lThis and the biological plausibility of the hypothesis are reasons to examine the issue further

lStudies that use validated measures of fish intake and confirmed diagnoses of dementia and Alzheimer's disease would provide more conclusive evidence

Lianna Ishihara is an epidemiologist and

Carol Brayne is a neuroepidemiologist at the Institute of Public Health, University of Cambridge


1 de Wilde MC et al. The effect of N-3 polyunsaturated fatty acid-rich diets on cognitive and cerebrovascular parameters in chronic cerebral hypoperfusion. Brain Res. 30-8-2002; 947(2):166-73

2 Conquer JA et al. Fatty acid analysis of blood plasma of patients with Alzheimer's disease, other types of dementia, and cognitive impairment.

Lipids 2000;35(12):1305-12

3 Friedland RP. Fish consumption and the risk of Alzheimer disease: is it time to make dietary recommendations?

Arch.Neurol. 2003;60(7):923-4

4 Morris MC et al. Consumption of fish and N-3 fatty acids and risk of incident Alzheimer disease. Arch.Neurol. 2003;60(7):940-6

5 Kalmijn S et al. Dietary fat intake and the risk of incident dementia in the Rotterdam Study. Ann.Neurol. 1997;42(5):776-82

6 Engelhart MJ et al. Diet and risk of dementia: does fat matter? The Rotterdam Study. Neurology 24-12-2002; 59(12):1915-21

7 Barberger-Gateau P et al. Fish, meat, and risk of dementia: cohort study. BMJ 26-10-2002; 325(7370):932-3

8 Requejo AM et al. Influence of nutrition on cognitive function in a group of elderly, independently living people. Eur.J.Clin.Nutr. 2003;57

Suppl 1:S54-7:S54-S57

9 Kalmijn S et al. Polyunsaturated fatty acids, antioxidants, and cognitive function in very old men. Am.J.Epidemiol. 1-1-1997; 145(1):33-41

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say