Whether it’s the first coffee of the morning or a comforting cup of tea in the afternoon, many of us rely on these daily rituals to help us feel more alert and focused. Coffee and tea are among the most popular beverages worldwide, but could they do more than help us through the afternoon slump? Could they also influence how our brains age?
Caffeine, one of the best-known bioactive compounds in both coffee and tea, improves alertness by blocking adenosine receptors in the brain, reducing the sensation of fatigue (Fredholm et al., 1999; Nehlig, 2010). Beyond these short-term effects, however, its role in long-term cognitive health remains uncertain.
Over the past two decades, observational studies have reported conflicting findings. Several large prospective cohorts and meta-analyses have suggested that moderate coffee or tea consumption is associated with a lower risk of dementia or cognitive decline (Chen et al., 2020; Liu et al., 2016). In contrast, other studies have found little evidence of a benefit, considerable variability according to genetic background and habitual consumption, or even potential adverse effects of high caffeine intake (Kapellou et al., 2023).
Interpreting this evidence has been challenging because many studies relied on a single dietary assessment at baseline, had relatively short follow-up, did not distinguish between caffeinated and decaffeinated coffee, or were vulnerable to reverse causation, whereby early, undiagnosed cognitive decline may itself influence coffee and tea drinking habits. Consequently, it has remained difficult to determine whether these associations are causal or simply reflect differences in lifestyle and health between people who drink more or less coffee and tea.
In this new study, Zhang and colleagues report that people who regularly consumed moderate amounts of caffeinated coffee or tea had a lower risk of developing dementia and experienced slightly better cognitive ageing than those who drank little or none. Interestingly, these associations were not observed for decaffeinated coffee, raising the intriguing possibility that caffeine – or perhaps other compounds found in caffeinated beverages – may play a role in maintaining brain health. But how convincing is the evidence?
Methods
Zhang and colleagues analysed data from two large prospective US cohorts:
These cohorts included more than 130,000 participants who were followed for up to 43 years. Unlike many previous studies that measured diet only once, participants completed validated food frequency questionnaires every 2-4 years, allowing the researchers to capture long-term changes in coffee, tea and caffeine consumption.
The researchers separately examined caffeinated coffee, decaffeinated coffee, tea and total caffeine (calculated from coffee, tea, soda, and chocolate consumption, using standard caffeine content values), and related these to dementia, subjective cognitive decline and objective cognitive performance.
The analyses accounted for a wide range of potential confounders – including age, smoking, physical activity, diet quality, cardiovascular risk factors and socioeconomic status – and included several sensitivity analyses to test the robustness of the findings.
Results
Coffee consumption
During up to 43 years of follow-up, 11,033 participants developed dementia. Compared with people who rarely drank caffeinated coffee (<1 cup/month), those consuming 2-3 cups per day had a 17% lower risk of dementia (Hazard Ratio [HR] 0.83, 95% CI 0.75 to 0.92). Drinking more than three cups per day did not appear to provide additional benefit, suggesting that the association plateaued at moderate consumption levels.
Tea consumption
Tea consumption showed a similar pattern. Participants drinking 1-2 cups of tea per day had a 16% lower risk of dementia compared with those who rarely drank tea (HR 0.84, 95% CI 0.77 to 0.92). Again, higher levels of tea consumption were not associated with substantially greater reductions in risk.
Total caffeine intake
When caffeine intake from all dietary sources was analysed, the lowest dementia risk was observed at around 300 mg of caffeine per day, approximately equivalent to 2-3 cups of coffee. These findings suggest that moderate consumption of both coffee and tea was associated with lower dementia risk, although the optimal amounts differed between the two beverages.
Decaffeinated coffee
Interestingly, decaffeinated coffee was not associated with a reduced risk of dementia, despite showing a similar dose-response analysis. This was one of the most striking findings in the paper and led the authors to suggest that caffeine may contribute to the observed associations. However, the study cannot determine whether caffeine itself is responsible, or whether it simply acts as a marker for other compounds present in caffeinated beverages or for differences in lifestyle between people who choose caffeinated versus decaffeinated coffee.
Cognitive ageing
Because dementia represents the final stage of a long neurodegenerative process, the researchers also examined subjective cognitive decline (participants’ own perception of worsening memory or thinking) and objective cognitive performance (using standardised cognitive tests) as intermediate outcomes that provide valuable insight into the cognitive changes that may precede clinical diagnosis.
Participants consuming 2-3 cups of caffeinated coffee per day reported less subjective cognitive decline over time, while those drinking 3-4 cups per day achieved slightly higher scores on objective cognitive tests. However, the absolute differences were very small. For example, the improvement in the Telephone Interview for Cognitive Status (TICS) score corresponded to approximately 0.11 points, equivalent to delaying cognitive ageing by around 8 months.
Sensitivity analyses
Importantly, the associations remained broadly consistent across multiple sensitivity analyses, including different statistical models and adjustments for a wide range of demographic, lifestyle and health-related factors. This increases confidence that the observed associations were robust, but it does not prove that coffee or tea directly reduce the risk of dementia, as residual confounding and reverse causation cannot be completely excluded in observational studies.
Conclusions
The authors concluded that moderate consumption of caffeinated coffee and tea was associated with a lower risk of dementia and modestly better cognitive ageing, whereas no similar association was observed for decaffeinated coffee.
While this large, well-conducted longitudinal study provides some of the strongest observational evidence to date linking moderate coffee and tea consumption with a lower risk of dementia, but further research is needed to clarify the underlying biological mechanisms and determine whether these associations are causal.
Strengths and limitations
This study has several notable strengths that make it one of the most robust observational studies of coffee, tea and dementia published to date. First, the exceptionally large sample size (>130,000 participants), long follow-up (up to 43 years) and large number of dementia cases (>11,000) provided substantial statistical power. Second, unlike many previous nutritional studies, coffee and tea consumption were assessed repeatedly every 2-4 years rather than only at baseline, allowing the authors to better capture long-term drinking habits and reduce exposure misclassification. Finally, the analyses accounted for an extensive range of demographic, lifestyle and health-related factors and were supported by multiple sensitivity analyses, increasing confidence that the observed associations were robust.
Nevertheless, the findings should be interpreted within the limitations of an observational study. Although the authors adjusted for many potential confounders, residual confounding cannot be excluded. Coffee and tea drinkers may differ from non-drinkers in ways that are difficult to measure accurately, including healthcare-seeking behaviours, social engagement, sleep patterns or cognitive reserve, all of which may influence dementia risk. Similarly, reverse causation remains possible. Neurodegenerative changes can begin many years before a clinical diagnosis of dementia and may lead individuals to change their drinking habits long before symptoms become apparent.
Another interesting point relates to the interpretation of caffeine as the potential protective agent. The absence of an association with decaffeinated coffee is consistent with this hypothesis, but it does not prove that caffeine itself is responsible. Coffee and tea contain hundreds of bioactive compounds, and people choosing decaffeinated coffee may differ systematically from those drinking caffeinated coffee (e.g. underlying health conditions such as hypertension, insomnia or gastrointestinal disease) that encourage these participants to drink decaffeinated coffee. Furthermore, although the study analysed coffee and tea separately while statistically adjusting each for the other, it did not examine patterns of combined consumption (e.g. coffee only, tea only, or both beverages). Consequently, it remains unclear whether the observed associations reflect caffeine exposure, beverage-specific compounds, or combinations of both.
Overall, this is a carefully conducted and methodologically rigorous cohort study that substantially strengthens the evidence linking moderate coffee and tea consumption with a lower risk of dementia. However, it cannot establish causality. Future studies integrating longitudinal epidemiology with causal inference approaches, such as Mendelian randomisation of genetically predicted coffee consumption or caffeine metabolism, together with biomarker analyses, could help distinguish whether caffeine itself – or other bioactive compounds in coffee and tea – plays a causal role in protecting cognitive health.
Implications for practice
So, should we all start drinking more coffee or tea to reduce our risk of dementia?
This study provides reassuring evidence that moderate consumption of caffeinated coffee and tea appears compatible with healthy brain ageing. However, these strong associations are not evidence for causation. For clinicians, there is little reason to recommend changing current practice based on these findings alone. Equally, there is no evidence here to support advising healthy adults to avoid moderate coffee or tea consumption because of concerns about cognitive decline.
The findings are perhaps most relevant from a public health perspective. Dementia has a long preclinical phase, and no single dietary factor is likely to have a large impact on disease risk. Instead, coffee and tea should be considered within the broader context of healthy ageing, alongside established protective factors such as regular physical activity, smoking cessation, good cardiovascular health, adequate sleep and social engagement. It is unlikely that a few extra cups of coffee could compensate for these stronger determinants of brain health.
Caffeine has well-established effects on brain physiology, making a biological explanation for these findings plausible (Fredholm et al., 1999; McLellan et al., 2016; Nehlig, 2010). However, changes in neural activity, cerebral blood flow or brain structure do not necessarily translate into meaningful improvements in cognitive function or a reduced risk of dementia. Understanding how these biological effects relate to dementia risk remains an important research challenge.
An additional consideration is whether the effects of caffeine are the same throughout life. This study focused on middle-aged and older adults, asking whether long-term coffee and tea consumption influences cognitive ageing and dementia risk. Whether these findings apply across the life course remains uncertain because the developing brain differs substantially from the ageing brain, and some evidence suggests higher caffeine consumption may be associated with poorer cognitive performance in children (Temple, 2009, 2019; Zhang et al., 2020). As caffeine consumption increases among younger people, particularly through energy drinks, this is an important area for future research.
One aspect of this study I found particularly thought-provoking is that it illustrates how difficult it is to move from association to causation. As epidemiologists, we are often tempted to ask whether coffee protects against dementia. Yet perhaps the more interesting question is how the observed associations arise. Is caffeine responsible, or do other bioactive compounds in coffee and tea also contribute? Why do coffee and tea appear to have different “optimal” consumption levels? Could these effects vary across the life course? Answering these questions could improve our understanding of cognitive ageing and identify new targets for dementia prevention. Future studies combining longitudinal epidemiology with causal inference approaches, including Mendelian randomisation and biomarker studies, may help answer these questions.
Tomorrow morning, many of us will once again reach for our favourite cup of coffee or tea. This study suggests we can probably continue enjoying that daily ritual without worrying that it is harming our cognitive health. Whether it is actively protecting our brains, however, remains an open – and fascinating – scientific question.
Statement of interests
Andrea Santangelo declares no conflicts of interest in relation to this blog. AI (ChatGPT) was used to support some of the paragraph’s structure and clarity after Andrea’s first draft, but all critical appraisal and conclusions are her own.
Editor
Edited by Éimear Foley. ChatGPT assisted with language refinement and formatting during the editorial phase.
Links
Primary paper
Yu Zhang, Yuxi Liu, Yanping Li, Yuhan Li, Xiao Gu, Jae H. Kang, A. Heather Eliassen, Molin Wang, Eric B. Rimm, Walter C. Willett, Frank B. Hu, Meir J. Stampfer, Dong D. Wang. Coffee and Tea Intake, Dementia Risk, and Cognitive Function. JAMA. 2026;335(11):961-974. doi:10.1001/jama.2025.27259.
Other references
Fredholm BB, Bättig K, Holmén J, et al (1999) Actions of Caffeine in the Brain with Special Reference to Factors That Contribute to Its Widespread Use. Pharmacol Rev 51:83–133. https://doi.org/10.1016/S0031-6997(24)01396-6
Nehlig A (2010) Is caffeine a cognitive enhancer? J Alzheimers Dis 20 Suppl 1: https://doi.org/10.3233/JAD-2010-091315
Chen JQA, Scheltens P, Groot C, Ossenkoppele R (2020) Associations between Caffeine Consumption, Cognitive Decline, and Dementia: A Systematic Review. Journal of Alzheimer’s Disease 78:1519–1546. DOI: 10.3233/JAD-201069
Liu QP, Wu YF, Cheng HY, et al (2016) Habitual coffee consumption and risk of cognitive decline/dementia: A systematic review and meta-analysis of prospective cohort studies. Nutrition 32:628–636. https://doi.org/10.1016/J.NUT.2015.11.015
Kapellou A, King A, Graham CAM, et al (2023) Genetics of caffeine and brain-related outcomes – a systematic review of observational studies and randomized trials. Nutr Rev 81:1571–1598. https://doi.org/10.1093/NUTRIT/NUAD029
McLellan TM, Caldwell JA, Lieberman HR (2016) A review of caffeine’s effects on cognitive, physical and occupational performance. Neurosci Biobehav Rev 71:294–312. https://doi.org/10.1016/J.NEUBIOREV.2016.09.001
Temple JL (2019) Review: Trends, Safety, and Recommendations for Caffeine Use in Children and Adolescents. J Am Acad Child Adolesc Psychiatry 58:36–45. https://doi.org/10.1016/J.JAAC.2018.06.030
Temple JL (2009) Caffeine Use in Children: What we know, what we have left to learn, and why we should worry. Neurosci Biobehav Rev 33:793. https://doi.org/10.1016/J.NEUBIOREV.2009.01.001
Zhang H, Lee ZX, Qiu A (2020) Caffeine intake and Cognitive Functions in Children. Psychopharmacology (Berl) 237:3109. https://doi.org/10.1007/S00213-020-05596-8
