Research explained: prospective cohort study
Caffeinated coffee and tea linked to lower dementia risk over up to 43 years in two large US cohorts, but decaf was not
A Harvard analysis of 131,821 nurses and health professionals found the clearest link at about 2 to 3 cups of caffeinated coffee or 1 to 2 cups of tea a day. It is observational and cannot show that coffee prevents dementia.
People who drank the most caffeinated coffee had a lower rate of dementia than those who drank the least, according to a long-running US cohort study published in JAMA (March 17, 2026 issue). The top quarter of coffee drinkers had 141 dementia cases per 100,000 person-years, compared with 330 in the bottom quarter, a hazard ratio of 0.82 after adjusting for other factors. Tea showed a similar pattern, but decaffeinated coffee did not. The study made headlines when it came out and is being shared widely again this month.
What the researchers did
A Harvard team pooled two cohorts: 86,606 women from the Nurses’ Health Study, followed from 1980 to 2023, and 45,215 men from the Health Professionals Follow-up Study, followed from 1986 to 2023. In total 131,821 people without cancer, Parkinson disease or dementia at the start took part. They reported what they drank on food questionnaires every 2 to 4 years.
Over up to 43 years of follow-up (median 36.8 years), 11,033 people developed dementia. Higher caffeinated coffee intake also went with fewer people reporting that their memory and thinking had slipped: 7.8% in the top quarter against 9.5% in the bottom. In the nurses, who took phone-based memory tests, heavy coffee drinkers scored slightly higher on one test, while the difference on an overall cognition score was not statistically significant. The link was strongest at moderate amounts, around 2 to 3 cups of caffeinated coffee or 1 to 2 cups of tea a day, and did not keep growing with more.
How strong is the evidence?
This is an observational study, so coffee and tea are linked to lower dementia risk here, which is not the same as proof that they protect the brain. Intake was self-reported, and coffee drinkers may differ from non-drinkers in ways the adjustments did not fully capture. Participants were mostly health professionals, which may limit how well the results apply to others. One author reported research support from the Analysis Group; no other conflicts were listed.
What it means for you
If you already enjoy coffee or tea, this study gives no reason to stop, and moderate amounts showed the clearest link. It is not a reason to start drinking more, or to push past what sits well with your sleep. Black tea and green tea are gentler on caffeine. More options are in drinks.