Epidemiological studies fairly convincingly suggest that higher intakes of fatty fish and n-3 fatty acids are associated with reduced risk of Alzheimer's disease (AD). DHA in plasma is normally positively associated with DHA intake. However, despite being associated with lower fish and DHA intake, unexpectedly, plasma (or brain) DHA is frequently not lower in AD. This review will highlight some metabolic and physiological factors such as ageing and apoE polymorphism that influence DHA homeostasis. Compared with young adults, blood DHA is often slightly but significantly higher in older adults without any age-related cognitive decline. Higher plasma DHA in older adults could be a sign that their fish or DHA intake is higher. However, our supplementation and carbon-13 tracer studies also show that DHA metabolism, e.g. transit through the plasma, apparent retroconversion and β-oxidation, is altered in healthy older compared with healthy young adults. ApoE4 increases the risk of AD, possibly in part because it too changes DHA homeostasis. Therefore, independent of differences in fish intake, changing DHA homeostasis may tend to obscure the relationship between DHA intake and plasma DHA which, in turn, may contribute to making older adults more susceptible to cognitive decline despite older adults having similar or sometimes higher plasma DHA than in younger adults. In conclusion, recent development of new tools such as isotopically labelled DHA to study DHA metabolism in human subjects highlights some promising avenues to evaluate how and why DHA metabolism changes during ageing and AD.