Coffee and Tea: Lowering Dementia Risk with Caffeine

Based on Zhang Y et al., “Coffee and Tea Intake, Dementia Risk, and Cognitive Function,” JAMA, published February 9, 2026

The Longest Look Yet at Caffeine and the Aging Brain

For decades, the idea that your morning coffee might protect your mind in old age has hovered somewhere between optimistic folk wisdom and genuine scientific hypothesis. A landmark study published in JAMA on February 9, 2026 moves that hypothesis considerably closer to credible — not by proving causation, but by assembling the most extensive observational data yet gathered on the question.

Led by epidemiologist Yu Zhang from Harvard’s T.H. Chan School of Public Health and a team from Mass General Brigham and the Broad Institute of MIT and Harvard, the study analyzed 131,821 participants drawn from two of the longest-running cohort studies in American medicine: the Nurses’ Health Study (NHS) and the Health Professionals Follow-Up Study (HPFS). Participants were tracked for up to 43 years — the median follow-up was 36.8 years — with dietary assessments repeated every two to four years using validated food frequency questionnaires. Over that span, 11,033 participants developed dementia, identified through physician diagnoses and death records. The result is an extraordinarily detailed longitudinal portrait of how caffeinated beverages relate to cognitive aging across a full adult lifetime.

What the Study Found

The headline finding is an 18% reduction in dementia risk among participants in the highest quartile of caffeinated coffee consumption compared to those in the lowest quartile. In absolute terms, the highest consumers had 141 cases of dementia per 100,000 person-years, versus 330 per 100,000 in the lowest — a difference that remained statistically significant after adjustment for an extensive list of confounders including age, body mass index, smoking, physical activity, alcohol use, total caloric intake, and educational attainment. The hazard ratio was 0.82 (95% CI, 0.76–0.89).

Tea told a similar story. Higher tea intake was associated with lower dementia risk and lower prevalence of subjective cognitive decline, with dose-response patterns closely mirroring those for caffeinated coffee. The sweet spot, according to the analysis, was 2 to 3 cups of caffeinated coffee or 1 to 2 cups of tea per day — the intake levels associated with the steepest inverse relationship with dementia incidence.

Critically, decaffeinated coffee showed no such association. Participants who drank decaffeinated coffee did not experience the same reductions in dementia risk, and in some cognitive sub-analyses, higher decaffeinated coffee consumption was actually associated with worse verbal memory performance in women — though this finding requires cautious interpretation. The contrast between caffeinated and decaffeinated beverages is arguably the study’s most analytically significant design feature: it strongly implies that caffeine itself, rather than other shared compounds like polyphenols or antioxidants, is the primary driver of the observed associations.

Cognitive Function: A More Modest Picture

Beyond dementia incidence, the study examined objective cognitive performance in the NHS cohort using the Telephone Interview for Cognitive Status (TICS) — a validated telephone-based neuropsychological measure. Higher caffeinated coffee intake was associated with a modest improvement in TICS scores (mean difference 0.11; 95% CI, 0.01–0.21). An association with a broader global cognition composite score trended in the same direction but fell just short of statistical significance (P = .06). Subjective cognitive decline — defined by self-reported concerns about memory and thinking — was meaningfully lower among the highest coffee consumers: 7.8% versus 9.5% in the lowest consumption group.

The cognitive effect sizes, while consistent in direction, are small. The authors themselves caution against over-interpretation, and expert commentators have uniformly echoed this. The 18% reduction in dementia risk is a real and clinically interesting signal, but it does not render caffeine a pharmaceutical-grade preventive agent.

Biological Plausibility: Why Caffeine Might Protect the Brain

The decaffeinated-versus-caffeinated split is not just analytically useful; it also narrows the mechanistic search. The most well-characterized pathway involves adenosine receptor antagonism. Caffeine competitively blocks adenosine A1 and A2A receptors in the brain — receptors that, when activated by the endogenous neuromodulator adenosine, suppress neural activity and reduce the release of neurotransmitters including dopamine, acetylcholine, glutamate, and serotonin. With aging, neurodegeneration, and accumulating dementia pathology, the activity of these arousal and memory-supporting neurotransmitter systems declines. Caffeine, as an inhibitor of adenosine signaling, counteracts this suppression.

Adenosine A2A receptor upregulation has been documented in the brains of people with Alzheimer’s disease, and blocking this receptor reduces the production and aggregation of amyloid-beta and tau proteins in animal models. Whether chronic caffeine consumption modifies these pathways over decades in humans remains unproven at the mechanistic level, but the biological rationale is coherent and under active investigation. The JAMA finding has already prompted interest in the adenosine pathway as a drug discovery target — a parallel avenue to the amyloid-clearing antibodies that have consumed much of recent Alzheimer’s research investment.

Caffeine also has anti-inflammatory properties and may reduce neuroinflammation through inhibition of microglial activation. Polyphenols present in both coffee and tea — though common to decaffeinated varieties as well — have their own antioxidant and anti-inflammatory effects, which may contribute modestly to long-term cognitive health even if they do not account for the dementia-specific findings here.

Genetic Risk: An Encouraging Sub-analysis

One of the study’s more reassuring secondary findings concerns genetic predisposition to dementia. The team examined whether the association between caffeine consumption and lower dementia risk varied according to participants’ polygenic risk scores for Alzheimer’s disease. The association held across high and low genetic risk groups with similar effect sizes — suggesting that the potential benefits of moderate caffeinated beverage consumption are not limited to those already at lower biological risk of dementia. As lead author Zhang noted in press materials, the consistency across genetic risk strata adds a layer of credibility to the finding and has direct implications for public health guidance.

The analysis also found that protective associations were stronger in participants aged 75 or younger at follow-up — a 35% reduction in dementia risk among that group versus approximately 19% in those older than 75. This age-dependent gradient is biologically plausible: earlier in the course of potential dementia pathology, modification of adenosine signaling or neuroinflammation may be more impactful.

Limitations and the Causation Question

The authors, peer commentators, and dementia research organizations have been appropriately measured in communicating what this study can and cannot establish.

This is observational data. The study cannot distinguish whether caffeinated coffee and tea reduce dementia risk from whether people who are cognitively healthier simply maintain higher caffeine consumption for longer. This is the reverse causation problem: early cognitive decline may cause reduced appetite, social withdrawal, or routine disruption that leads to drinking less coffee — making lower consumption a consequence rather than a cause of cognitive decline. The investigators attempted to address this through sensitivity analyses, including lagged exposure models that used coffee intake from earlier in follow-up rather than the period immediately before diagnosis, and the associations persisted. But reverse causation cannot be fully excluded in any observational study of this design.

Residual confounding remains a concern. Coffee and tea drinkers may systematically differ from non-drinkers in ways that questionnaires cannot fully capture — social engagement, dietary patterns, occupational stimulation, or access to healthcare among them.

The cohort is not fully representative. Participants were predominantly white, highly educated health professionals in the United States. Whether the findings generalize to populations with different dietary backgrounds, genetic ancestry, healthcare access, or cultural caffeine habits is uncertain.

Effect sizes are modest. An 18% reduction in dementia risk at the population level is epidemiologically meaningful, but it is not a guarantee at the individual level. As senior author Dong Wang stated, the effect is small and part of a broader landscape of protective lifestyle factors. Cognitive reserve, physical activity, sleep quality, social connection, cardiovascular health, and hearing management all contribute to dementia risk in ways that likely dwarf any single dietary habit.

Context: Where This Fits in the Literature

The JAMA study is not the first to suggest a link between caffeine and cognitive protection. A 2009 study in the Journal of Alzheimer’s Disease found that midlife coffee consumption was associated with a 65% lower risk of dementia in a Finnish cohort. A 2014 meta-analysis in the European Journal of Nutrition pooled data from 11 studies and found a modest inverse association between coffee consumption and dementia risk. Earlier work from the CAIDE (Cardiovascular Risk Factors, Aging, and Dementia) cohort in Finland also identified moderate coffee drinking in midlife as protective.

What distinguishes the new JAMA study is its scale, the duration of follow-up, the repeated dietary assessments over time, and — most importantly — the deliberate and systematic comparison of caffeinated coffee, decaffeinated coffee, and tea. Most prior research did not differentiate between these, making it impossible to isolate caffeine as a factor. The current analysis does exactly that, and the caffeinated-versus-decaffeinated contrast is its strongest methodological contribution.

What This Means in Practice

No physician or dietitian would prescribe coffee as a dementia prevention strategy on the basis of this study alone. The finding is associational, the effect is modest, and the literature has not yet produced a randomized controlled trial demonstrating that increasing caffeine intake in middle-aged adults reduces dementia incidence — a trial that would, in any case, be extraordinarily difficult to conduct over the timescale relevant to dementia development.

What the study does support is the conclusion that habitual, moderate consumption of caffeinated coffee or tea is not cognitively harmful — and may, over a lifetime, contribute modestly to lower dementia risk. For most healthy adults without contraindications (cardiac arrhythmias, anxiety disorders, pregnancy, or conditions exacerbated by caffeine), there is no evidence-based reason to discourage the daily 2 to 3 cups of coffee or 1 to 2 cups of tea that appear associated with the greatest benefit in this study.

For public health purposes, the broader message is about the cumulative value of modifiable lifestyle habits across decades of adult life. Dementia prevention is not a single intervention but a lifelong portfolio — and if one component of that portfolio happens to be a daily cup of coffee, the evidence, now substantially strengthened, suggests that is no bad thing.

Sources

Zhang Y, Liu Y, Li Y, et al. Coffee and Tea Intake, Dementia Risk, and Cognitive Function. JAMA. Published online February 9, 2026. doi:10.1001/jama.2025.27259
Mass General Brigham Newsroom. Consuming 2-3 Cups of Coffee Daily Associated with Lower Dementia Risk, Better Cognitive Function. February 9, 2026. Available at: https://www.massgeneralbrigham.org/en/about/newsroom/press-releases/consuming-coffee-associated-with-lower-dementia-risk
Harvard Gazette. Drinking 2-3 cups of coffee a day tied to lower dementia risk. February 9, 2026. Available at: https://news.harvard.edu/gazette/story/2026/02/drinking-2-3-cups-of-coffee-a-day-tied-to-lower-dementia-risk/
WBUR News. Top off your coffee: Study finds moderate caffeine intake may reduce dementia risk. February 9, 2026. Available at: https://www.wbur.org/news/2026/02/09/coffee-tea-caffeine-intake-dementia-risk-mass-general-study
Nature. Coffee linked to slower brain ageing in study of 130,000 people. February 9, 2026. Available at: https://www.nature.com/articles/d41586-026-00409-y
Science Media Centre. Expert reaction to study looking at tea and coffee intake and dementia risk. February 9, 2026. Available at: https://www.sciencemediacentre.org/expert-reaction-to-study-looking-at-tea-and-coffee-intake-and-dementia-risk/
Alzheimer’s Research UK. Could tea or coffee reduce dementia risk? February 9, 2026. Available at: https://www.alzheimersresearchuk.org/news/could-tea-or-coffee-reduce-dementia-risk-heres-what-you-need-to-know/
Eskelinen MH, Kivipelto M. Caffeine as a protective factor in dementia and Alzheimer’s disease. J Alzheimers Dis. 2010;20 Suppl 1:S167–74. doi:10.3233/JAD-2010-1404
Qi H, Li S. Dose-response meta-analysis on coffee, tea and caffeine consumption with risk of late-life cognitive disorders. Eur J Nutr. 2014;53(6):1573–1586. doi:10.1007/s00394-014-0744-0
Drug Discovery Trends. JAMA study finds moderate caffeine intake tracks with lower dementia risk, provides adenosine-pathway nudge for drug hunters. February 2026. Available at: https://www.drugdiscoverytrends.com/jama-study-finds-moderate-caffeine-intake-tracks-with-lower-dementia-risk-provides-adenosine-pathway-nudge-for-drug-hunters/