Two or three cups of coffee a day linked to less dementia after 43 years of follow-up

What happened

• A long-term study of US adults associated regular caffeinated coffee with a substantially lower risk of dementia and better cognitive health with age.
• Follow-up extended up to 43 years.
• Adults drinking the highest amounts of caffeinated coffee had 141 dementia cases per 100,000 people per year, against 330 per 100,000 among those drinking little or none.
• The association was strongest at two to three cups a day.
• Tea was also linked to better cognitive outcomes, with a somewhat lower optimal amount; decaffeinated coffee did not show the same benefit.
• ScienceDaily circulated it on 7 November 2026.

Why it matters

That decaf did not show the effect suggests caffeine, or an associated component, could be part of the explanation. Either way the design cannot assert cause: coffee drinkers may differ in other habits, such as exercise or diet, that protect the brain. With dementia a growing public health problem, any everyday habit with a strong association draws attention, though caution is needed.

The number

141 versus 330. Dementia cases per 100,000 people per year among the highest caffeinated-coffee drinkers and those avoiding it.

Context

A cohort study follows people over time and records habits and disease but does not randomize consumption, so results show association, not effect. Reverse causation is possible, for example if people with early decline stop drinking coffee, though 43 years of follow-up reduces that risk. The text read names neither the journal, the participant count nor the adjustments, so those must be checked in the original paper. Related coverage: The brain processes odours by breathing phase and not by elapsed time, The parietal cortex readjusts how it encodes numbers according to the range being estimated, Alcohol makes the liver send inflammatory signals to the brain in Alzheimer mice, A PET scan with a labelled amino acid detects 91% of glioblastoma, against 53% for MRI.

In long-follow-up cohort studies the main difficulty is separating one habit’s effect from others that accompany it. Regular coffee drinkers may smoke more or less, sleep differently or have a different income. Analyses try to subtract these statistically, but confidence in the result depends on which factors were measured. That is why 141 versus 330 is a starting point, and the original paper should detail how it was adjusted.

What’s next

• The journal and methods must be identified to assess the analysis.
• Controlled trials of caffeine could test for a causal effect.
• Other studies can review the difference between coffee, tea and decaf.

Bottom line

The morning cup may accompany a healthy brain, but there is no data to prescribe it. For now it is a clue, not an indication.

Sources

• ScienceDaily, “Drinking 2–3 cups of coffee a day linked to dramatically lower dementia risk”, 7 November 2026. The text read does not identify the study’s journal or original paper date, so figures come from the popular summary.
• Cover image: Christoph, CC0, via Wikimedia Commons

Edited by Rodrigo Cornejo. How we select and verify: who writes these notes.