Coffee and Dementia: What Do New Scientific Studies Reveal?

Coffee is frequently mentioned in studies on brain health. Caffeine can affect the central nervous system, while polyphenols and compounds formed during roasting are of interest due to their potential roles in oxidative stress and inflammation.

However, a major question remains unanswered: Do frequent coffee drinkers have a lower risk of dementia than non-coffee drinkers?

In February 2026, two notable studies were published close together, providing additional data. One study in JAMA followed two occupational cohorts in the United States over many decades. The other study was a meta-analysis of cohorts with over 450,000 participants.

Interestingly, both studies found a non-linear dose-response relationship. However, they did not provide entirely identical interpretations for high consumption levels.

1. What is Dementia and Why Should We Distinguish It from Normal Memory Decline?

Dementia is a term referring to a group of syndromes causing sufficient cognitive function decline to affect independent living abilities.

Possible symptoms may include:

• Decline in memory and ability to absorb new information.
• Difficulty in planning, problem-solving, and decision-making.
• Disorientation in time and space.
• Changes in language, behavior, or ability to perform daily activities.

Alzheimer's disease is the most common cause of dementia, but not every dementia case is Alzheimer's. There is also vascular dementia, Lewy body dementia, and other forms.

Therefore, when a study evaluates dementia risk in general, the results cannot be automatically applied to Alzheimer's disease or all forms of cognitive decline.

2. JAMA 2026 Study: Following over 131,000 people for many decades

The study, titled Coffee and Tea Intake, Dementia Risk, and Cognitive Function, was published in JAMA on February 9, 2026.

The research team used data from two long-term cohorts:

• Nurses' Health Study (NHS): 86,606 women.
• Health Professionals Follow-up Study (HPFS): 45,215 men.

A total of 131,821 participants were involved. The maximum follow-up time reached 43 years, with a median follow-up time of 36.8 years. During the study, the authors recorded 11,033 new dementia cases.

This is a notable data scale because many previous nutritional studies only followed participants for a few years, limiting the evaluation of slow-progressing diseases like dementia.

Results for coffee with caffeine

When comparing the group with the highest caffeine consumption to the group with the lowest, the researchers found a pooled Hazard Ratio (HR) of 0.82, with a 95% confidence interval from 0.76 to 0.89.

This is a correlation with a lower risk of dementia of around 18% in the adjusted analysis model.

In the rate ratio analysis, the highest consumption group had 141 cases per 100,000 people per year, while the lowest consumption group recorded 330 cases per 100,000 people per year.

Which consumption level is most beneficial?

The dose-response analysis of the study found a non-linear relationship. A level of around 2-3 cups of caffeinated coffee per day was associated with the lowest risk of dementia compared to the non-consumers.

In the study, one cup was defined as 8 ounces, equivalent to approximately 237 ml. This is the volume unit of the study, not a fixed conversion for all Vietnamese coffee types.

Notably, exceeding this consumption level did not produce a clear advantage in the dose-response analysis.

What about decaf coffee?

Unlike caffeinated coffee, decaf coffee did not show a significant correlation with a reduced risk of dementia in the dose-response analysis of the study.

This finding makes caffeine a factor of interest, but it cannot be concluded that caffeine is the direct component responsible for the beneficial relationship. Other compounds may interact with caffeine or reflect a certain lifestyle pattern.

3. 2026 Integrated Analysis: Drinking a lot of coffee may no longer be beneficial

On February 12, 2026, Mazzoleni and colleagues published the study Coffee and tea consumption and risk of dementia: a dose-response meta-analysis of cohort and cohort-nested case-control studies in the Journal of Epidemiology and Population Health.

The research group searched for literature up to December 9, 2025 on PubMed and EMBASE, then combined ten studies with over 450,000 participants at the initial time point.

The average follow-up time of the studies was 11.5 years.

The results showed a U-shaped relationship between coffee quantity and dementia risk.

The authors noted the lowest risk in the range of 2-3 cups per day, equivalent to approximately 300-450 ml/day according to the analysis conversion.

For Alzheimer's disease, the analysis model showed no clear difference in risk until around three cups of coffee per day, after which there was a trend towards increased risk at higher consumption levels.

Two studies did not entirely contradict each other. Both suggested that moderate consumption might be the most favorable association. The difference lies in how the impact of high caffeine intake and the scope of the data were evaluated.

4. Why is caffeine related to brain health?

Caffeine primarily acts through adenosine receptor antagonism in the central nervous system.

Adenosine is a neuroregulatory substance involved in the process of feeling drowsy, regulating neural activity, and many other physiological responses.

When caffeine binds to adenosine A1 and A2A receptors without triggering an adenosine-like response, it can alter the activity of neural networks.

In preclinical studies, adenosine-related pathways are of interest because they may affect:

• Neural cell activity.
• Neuroinflammatory responses.
• Signal transmission regulation.
• Some oxidative stress pathways.

For coffee, polyphenol compounds and chlorogenic acid are also studied for their antioxidant properties and interactions with metabolic pathways.

However, it is essential to distinguish between evidence in cells, animals, and humans. A biological mechanism does not automatically prove that drinking coffee will prevent neurodegenerative disease in humans.

5. Why do studies on coffee and cognitive decline yield different results?

Difference in coffee quantity.

A 30 ml espresso cup is not equivalent to a 237 ml filtered cup. Caffeine content also varies depending on coffee type, Arabica-Robusta ratio, powder weight, extraction time, and brewing method.

If studies use different portion conversion methods, results on daily cups will not be entirely equivalent.

Difference in study subjects.

The JAMA study used data from US healthcare employees. Participants may differ from the general population in terms of education level, economic conditions, access to healthcare services, and lifestyle habits.

Therefore, results cannot be applied directly to all communities or demographic groups.

Potential for reverse causation.

A notable issue in nutritional research is reverse causation.

Individuals who begin showing signs of declining health, sleep disorders, or cognitive function changes may reduce their coffee intake before being diagnosed with cognitive decline.

When this happens, the low coffee consumption group may include individuals who are already experiencing initial health changes. This complicates the observed association.

Difference between general cognitive decline and Alzheimer's disease.

The cognitive decline is a group of symptoms with multiple causes. A study found a correlation with general cognitive decline does not mean that correlation appears in all forms of disease.

This is why you need to read separate results about Alzheimer's disease, vascular dementia, and other cognitive outcomes when data are available.

6. Important limitations of current scientific evidence

Two 2026 studies provide large data, but there are common limitations:

• No randomized trial: Participants self-select coffee-drinking habits, not randomly assigned to consume different levels of caffeine.
• Confounding potential: Diet, physical activity, smoking, underlying health conditions, sleep, and socioeconomic status may all affect dementia risk.
• Self-reported data: Coffee consumption is recorded through regular surveys, which may have recall bias and classification errors.
• Differences in brewing methods: Studies do not fully describe all variables such as roast level, extraction method, and chemical composition of each cup of coffee.
• No proof of preventive effect: Statistical correlation does not equate to potential replicable intervention efficacy.

7. How should coffee drinkers interpret these results?

For adults with established coffee-drinking habits and good tolerance, current results do not provide scientific grounds to eliminate coffee due to dementia concerns.

However, there is no basis to turn 2-3 cups per day into a brain-protective formula.

Please note the following practical points:

• Do not use coffee as a substitute for full sleep.
• Do not increase caffeine if insomnia, anxiety, or discomfort appear.
• Do not view coffee as a treatment for memory decline.
• Maintain physical activity, diverse diet, and control cardiovascular risk factors.
• Pay attention to blood pressure, blood sugar, sleep, and metabolic health in the long term.

Especially, do not convert international study cup numbers into an exact caffeine dose for everyone. Robusta, Arabica, espresso, and filtered coffee have vastly different caffeine levels.

SAVINA's scientific perspective

The notable point in the 2026 JAMA study is not that coffee has been proven to prevent cognitive decline, but the correlation between caffeine consumption and recorded cognitive health over a very long follow-up period.

Mazzoleni's meta-analysis adds a crucial perspective: the relationship between coffee quantity and dementia risk may be non-linear. Drinking more does not mean it is more beneficial.

Therefore, do not take a single number of 2-3 cups per day and turn it into a health recommendation. To determine if caffeine has a causal effect, long-term interventional studies with a more robust design are still needed.

Coffee can be part of a healthy lifestyle, but brain health is not determined by a single beverage. Sleep quality, exercise, nutrition, cardiovascular health, and other risk factors still play a foundational role.

References

1. Journal of the American Medical Association (2026). Coffee and Tea Intake, Dementia Risk, and Cognitive Function. Published on February 9, 2026. DOI: 10.1001/jama.2025.27259.
2. European Mazzoleni et al. (2026). Coffee and tea consumption and risk of dementia: a dose-response meta-analysis of cohort and cohort-nested case-control studies. Journal of Epidemiology and Population Health, 74(1), 203168. DOI: 10.1016/j.jeph.2026.203168.
3. National Institute on Aging. Alzheimer's Disease and Related Dementias – General Information on Dementia and Cognitive Health.

Content compiled from research published up to November 5, 2026. The results on coffee quantity are observational associations, not recommended preventive doses.