Your morning coffee habit might be doing more for your brain than you realize. A study published in JAMA in June 2025, drawing on more than 130,000 participants tracked for up to 43 years, found that adults who consistently drank two to three cups of coffee per day had a 35 percent lower risk of developing dementia compared with those who drank none. The protective association was strongest among people younger than 75, suggesting that decades of moderate coffee drinking during midlife and early old age may offer a meaningful cognitive advantage.
The sheer scale and duration of this research set it apart. But the findings also raise sharp questions: Does coffee itself protect the brain, or do coffee drinkers simply live differently? And at what point does more coffee stop helping and start hurting?
What the study actually measured
The JAMA analysis combined data from two of the longest-running dietary studies in the United States. The Health Professionals Follow-up Study, launched in 1986, enrolled male health professionals who reported their diets every four years. The Nurses’ Health Study did the same with female registered nurses starting in the early 1980s. Together, the cohorts gave researchers something rare in nutrition science: repeated snapshots of coffee and tea consumption over decades, rather than a single questionnaire taken at one moment in time.
Cognitive outcomes in a subset of the Nurses’ Health Study were assessed using the Telephone Interview for Cognitive Status, a validated screening tool that has been used in large-scale research since the late 1980s. This allowed the team to evaluate objective cognitive function across a geographically dispersed population without requiring clinic visits.
According to the study’s PubMed entry, the 35 percent risk reduction held after the researchers adjusted for genetic predisposition to Alzheimer’s disease, smoking status, physical activity, diet quality, and other health variables. The dose-response pattern was not linear: one cup offered some benefit, two to three cups appeared optimal, and the advantage did not keep climbing with higher intake.
That pattern echoes earlier pooled evidence. A 2024 systematic review and meta-analysis published in Food and Function, cataloged under PMID 39054894, examined multiple cohort studies on coffee, tea, and caffeine and found a non-linear relationship between coffee consumption and dementia risk. The lowest risk clustered around one to three cups per day. Individual studies varied in their conclusions, but the overall trend favored moderate consumption over abstinence or heavy intake.
The idea that midlife habits shape late-life brain health is not new, either. A Finnish population-based investigation known as the CAIDE study, published in the Journal of Alzheimer’s Disease, had already linked midlife coffee drinking to reduced dementia risk over roughly 21 years of follow-up. What the JAMA paper adds is substantially greater scale, longer duration, and the repeated dietary measurements needed to distinguish lifelong habits from temporary patterns.
Why the results are not as simple as they look
Even a 43-year observational study cannot prove that coffee prevents dementia. The most persistent concern is reverse causation. Neurodegenerative diseases begin silently years or even decades before diagnosis. People in the earliest, undetected stages of cognitive decline may cut back on coffee because of changing taste preferences, sleep disruption, or subtle shifts in daily routine. If that happens, the data would make abstainers look worse off even if coffee played no direct protective role.
The JAMA authors acknowledged this limitation and attempted to address it by excluding early dementia cases and using lagged exposure windows, but no observational design can fully eliminate the possibility.
Residual confounding is another concern. Coffee drinkers in these cohorts differed from non-drinkers in ways that go beyond the cup. They had different rates of smoking, exercise, dietary patterns, and chronic disease. Statistical models adjusted for many of these factors, but unmeasured variables, such as how routine-driven or socially engaged a person is, could still skew the results. Someone who maintains a steady two-to-three-cup habit for 40 years may simply be a more stable, health-conscious person in ways that are difficult to capture in a questionnaire.
There is also a ceiling on the benefit. Research using UK Biobank data has found that very high consumption, defined as more than six cups per day, was associated with higher odds of dementia and smaller total brain volumes on MRI. That finding introduces a clear tension: moderate intake looks protective across multiple cohorts, but heavy intake may carry its own risks. The threshold where benefit tips into harm has not been precisely defined and likely varies by individual metabolism, genetics, and how the coffee is prepared.
Then there is the question of what in coffee deserves the credit. Coffee contains hundreds of bioactive compounds, including caffeine, chlorogenic acids, and polyphenols. The JAMA study examined caffeinated versus decaffeinated coffee and compared coffee with tea, but the published results rely on summary statistics. Without access to individual-level data and full covariate adjustment tables, independent researchers cannot fully replicate the dose-response curves or rigorously test whether caffeine alone, non-caffeine compounds, or some synergy between them drives the association.
The study population also limits how broadly the findings apply. Both cohorts enrolled predominantly white health professionals with above-average education and consistent access to healthcare. Whether the same patterns hold across different racial and ethnic groups, lower-income populations, or countries with distinct coffee preparation traditions, such as Turkish, espresso, or boiled Scandinavian coffee, remains an open question.
What this means for your morning cup
The strongest evidence here comes from the JAMA paper’s combination of two independent, long-running cohorts with repeated dietary measurements and validated cognitive assessments. That makes the 35 percent figure more robust than a single-survey finding. But it still falls short of the gold standard: a randomized controlled trial assigning people to drink or avoid coffee for decades. Such a trial is logistically impractical, ethically complicated, and almost certainly will never be conducted.
Supporting data from the Food and Function meta-analysis and the earlier CAIDE study reinforce the direction of the association. When multiple independent datasets from different countries and time periods converge on the same pattern, the signal gains credibility even if no single study can close the causation gap. At minimum, moderate coffee consumption appears to be a marker of lifestyle patterns that favor healthier brain aging. It may also play a biologically active role, though the mechanism has not been pinned down.
The UK Biobank counterpoint prevents a simplistic reading. The evidence does not say “more coffee, less dementia.” It says roughly two to three cups per day sits in a sweet spot, and going well beyond that may not help and could hurt. For people who already drink coffee comfortably within that range and have no medical reasons to avoid it, such as uncontrolled high blood pressure, severe anxiety, or heart rhythm disorders, the data as of June 2025 are broadly reassuring.
For people who dislike coffee or cannot tolerate caffeine, the findings do not justify forcing a new habit. The absolute risk reduction for any single person is modest, and other interventions, including blood pressure management, regular physical activity, smoking cessation, and control of diabetes and cholesterol, have stronger and better-established effects on dementia risk.
Where the research goes from here
Several gaps remain. Longitudinal imaging studies could clarify how moderate coffee intake relates to brain structure and function over time, potentially revealing whether coffee drinkers show slower rates of hippocampal atrophy or white matter degradation. Genetic analyses focused on variants in the CYP1A2 gene, which governs how quickly the body metabolizes caffeine, might identify who benefits most and who should be more cautious. And larger, more diverse cohorts will be essential to test whether the patterns seen in predominantly white, well-educated health professionals hold up across the broader population.
Until those studies arrive, the most evidence-aligned position is cautious optimism. For many adults, particularly those under 75, a couple of daily cups of coffee fit comfortably within a brain-healthy lifestyle. But coffee is one piece of a much larger puzzle, not a substitute for the habits that decades of research have already proven matter most.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
