A prospective study that tracked more than 131,000 American health workers for up to 43 years found that people who regularly drank caffeinated coffee or tea had a modestly lower risk of developing dementia than those who rarely drank either. Decaffeinated coffee showed no such link, pointing to caffeine as the compound doing the work.
The findings, published in JAMA, come from the Nurses' Health Study and the Health Professionals Follow-Up Study, two of the longest-running cohorts in nutrition research. Researchers analyzed 131,821 participants, 11,033 of whom developed dementia during follow-up.
Compared with people who rarely or never drank caffeinated coffee, those in the highest intake group had a hazard ratio of 0.82, an 18 percent lower risk of dementia. The strongest signal appeared at moderate intake: roughly 2 to 3 cups of caffeinated coffee, or 1 to 2 cups of tea, per day.
What the numbers actually show
Rates of subjective cognitive decline, meaning self-reported memory and thinking problems, were 7.8 percent among regular caffeinated coffee or tea drinkers versus 9.5 percent in those who did not drink either. Objective cognitive tests in the nurses' cohort showed a small mean advantage as well, though the difference was modest.
Also read: The first hour after waking: 8 morning habits chronobiology says to rethink
The research team at Mass General Brigham, the Harvard T.H. Chan School of Public Health, and the Broad Institute also tested whether the association held in people at higher genetic risk of dementia. It did.
"We also compared people with different genetic predispositions to developing dementia and saw the same results," said lead author Yu Zhang, a PhD student at the Harvard Chan School and research trainee at Mass General Brigham.
Caffeine, not just coffee
Tea, which also contains caffeine, produced a similar association. Decaffeinated coffee did not. That pattern suggests caffeine itself, rather than coffee as a whole, is the ingredient tied to the protective signal in this data.
The authors caution against overreading the effect. Daniel Wang, an associate scientist in the Channing Division of Network Medicine and assistant professor at Harvard Medical School, framed the finding as one input among many that shape brain health over decades, not a single lever.
Also read: Long-term use of e-cigarettes may increase exposure to harmful chemicals
"While our results are encouraging, it's important to remember that the effect size is small and there are lots of important ways to protect cognitive function as we age," Wang said.
How the study fits with what we know
Dementia affects about 57 million people worldwide, according to the World Health Organization, with nearly 10 million new cases each year. Alzheimer's disease accounts for 60 to 70 percent of cases.
The WHO lists high blood pressure, diabetes, obesity, poor sleep, hearing loss, physical inactivity, smoking, harmful alcohol use, and air pollution among the modifiable risk factors.
Coffee is not on that list, and this study does not put it there. Because the data are observational, the analysis can show an association between habits and disease, not a causal effect. People who drink 2 to 3 cups of coffee daily may differ from non-drinkers in other ways that a statistical model cannot fully account for.
Also read: Early warning signs of colorectal cancer that younger adults should not ignore
Still, the scale and length of follow-up, more than 40 years, make this the strongest long-term dataset on the question so far. For readers who already drink coffee or tea, the practical takeaway is narrow: moderate intake looks compatible with long-term brain health, and swapping caffeinated for decaf takes away the ingredient this study links to the benefit.
Anyone concerned about memory, thinking, or dementia risk should speak with a doctor, particularly if caffeine interacts with existing conditions such as high blood pressure, arrhythmia, or a sleep disorder.
Also read: The hidden cholesterol sign you might spot in your legs