Does drinking coffee really help against dementia, or is that a myth?
Drinking 2-3 cups of coffee daily is associated with a lower dementia risk in several large studies, but this is not yet a proven causal relationship; adding tea appears to strengthen the association, and excessive consumption serves no additional purpose.
Drinking two to three cups of coffee a day is associated with an 18% lower risk of dementia compared with drinking little to no coffee. This finding comes from a large, long-running Harvard study of more than 130,000 people followed for up to 43 years. Caffeine appears to be the key factor: decaffeinated coffee showed no comparable effect in that same study.
Tea performs at least as well, and possibly better. A separate meta-analysis of 38 cohort studies found that each additional cup of tea per day was associated with a 4% reduction in dementia risk. The combination of 2-3 cups of coffee and 2-3 cups of tea per day was associated with 28% less dementia and even 32% fewer strokes, compared with drinking neither. That last finding comes from a UK Biobank analysis of more than 365,000 people.
There are, however, dissenting voices. That same meta-analysis of 38 cohort studies found no statistically significant protective effect of coffee on its own against dementia or Alzheimer's disease. An earlier meta-analysis of 8 studies likewise found no association. And the Rotterdam Study pointed to a subtle problem: people who are already quietly beginning to develop dementia sometimes start drinking less coffee at an early stage. If that is not properly corrected for in the analysis, coffee appears protective when the relationship actually runs the other way. That seed of doubt has not yet been fully removed.
How coffee might theoretically offer protection has been reasonably well studied in laboratory and animal research: caffeine and various plant-based compounds in coffee appear to protect brain cells. But mechanisms that work in a petri dish or in mice prove nothing in humans.
On the safety side, there is little to worry about at 2-3 cups per day. Excessive consumption, roughly 10 cups or more, can lead to anxiety, sleep problems and stomach complaints. There is therefore a practical ceiling on what you can do with it, even if an effect in humans is one day confirmed.
All claims are based on observational (cohort) studies and meta-analyses of observational research. No randomised controlled trials (RCTs) are available that can establish causality. The GRADE certainty-of-evidence rating is 'low' for the coffee-dementia link specifically (PMID 39054894). Reverse causality is a real methodological risk (PMID 25154552).