Why Coffee Drinkers Have Lower Dementia Risk — What The Research Actually Shows
Quick Read Multiple large studies show that people who drink coffee regularly have lower rates of dementia and cognitive decline.
We write a new article on supplements every day , all detailed, with citations.
Once a week we send an email containing a list of the last 7 days articles. Please sign up and also feel free to share the articles with friends and family who you think might benefit.
Simon Parrott – Vitacuity Ltd
Vitamin D does more than support bones. It acts as a signalling molecule that switches genes on and off in your brain cells, particularly in regions linked to memory and learning. Your brain can even produce its own active vitamin D locally using enzymes in neural tissue, but it needs circulating vitamin D as the raw material to do so. Animal studies show that sufficient vitamin D levels improve memory performance and reduce brain inflammation, while deficiency increases the buildup of toxic proteins linked to Alzheimer’s disease.
The catch is that your blood test result may not tell the whole story. Vitamin D must bind to a protein receptor called the VDR to work. This receptor exists as genetic variations across the population, meaning some people have less responsive versions. You could have normal blood vitamin D levels and still have a poorly functioning receptor in your brain tissue, limiting the protection you’re actually receiving.
Human trials testing vitamin D supplementation have produced inconsistent results, partly because they don’t account for these genetic differences and partly because most volunteers weren’t severely deficient to start with. The mechanistic evidence from animal studies is compelling, but we lack large-scale human trials specifically examining people with genuine deficiency. Over 50% of people over 60 in central Europe have insufficient vitamin D, making deficiency a real public health concern rather than a theoretical one.
Verdict: Vitamin D3 supplementation at standard doses is safe and sensible for brain health, but normal blood levels don’t guarantee your brain’s vitamin D receptor system is working optimally.
What if you’ve been taking your vitamin D3 faithfully every morning, your blood levels look perfectly fine on paper, and yet your brain is still not getting the protection you believe it is? What if the problem isn’t the dose, or the form, or whether you take it with fat, but something far more fundamental: the lock that vitamin D needs to turn in your brain isn’t working properly?
This isn’t a fringe theory. It sits at the heart of some of the most important recent research into neurodegeneration, cognitive decline and healthy brain ageing. Vitacuity has analysed over 1.77 million research papers and selected the most relevant findings on this topic. What follows is an honest account of what the science actually says, the compelling, the promising, and the parts we genuinely don’t know yet.
Most of us learned at school that vitamin D is a bone and calcium thing. That’s true, but it’s barely scratching the surface. Vitamin D is actually a secosteroid hormone, think of it less like a simple vitamin and more like a master signalling molecule that reaches into the nucleus of your cells and directly influences which genes get switched on or off [2].
Here’s how the journey works. When you take a vitamin D3 supplement, or make it through sunlight on your skin, your body starts a conversion process. The liver converts D3 into 25-hydroxyvitamin D (this is what your blood test measures). Then the kidneys convert that into the fully active form: 1,25-dihydroxyvitamin D3, or calcitriol [4].
But here’s what’s changed our understanding of vitamin D and the brain dramatically in recent years: your brain doesn’t just wait for calcitriol to arrive from the bloodstream. Brain cells, specifically glial cells, have their own local machinery to convert vitamin D into its active form right there in neural tissue, using an enzyme called CYP24A1, part of the cytochrome P450 system [1]. This is essentially the brain running its own on-site vitamin D activation facility.
Once activated, vitamin D’s effects in the brain work through a protein called the Vitamin D Receptor, the VDR. This receptor has been found in some of the most critical regions of the brain: the hippocampus (your memory centre), the cerebellum, the hypothalamus, the basal ganglia, the thalamus, and the orbital and temporal regions [1]. When active vitamin D binds to the VDR, it’s like a key turning in a lock, triggering a cascade of neuroprotective effects.
The critical question, and the one this post is really about, is: what happens when that lock doesn’t work as well as it should?
The discovery that VDR is present across so many brain regions is relatively recent, and it reframes everything about how we think about vitamin D and cognitive health.
A 2025 review published in *Current Nutrition Reports* confirmed that VDR, the nuclear steroid receptor through which vitamin D exerts its effects, has been identified in the cerebellum, thalamus, hypothalamus, basal ganglia, hippocampus, olfactory system, and temporal and orbital regions [1]. These aren’t peripheral areas. They are the regions most associated with memory formation, executive function, motor control, and emotional regulation, precisely the functions that decline in conditions like Alzheimer’s disease, Parkinson’s disease, and multiple sclerosis.
The same review notes that neurodegeneration is primarily associated with oxidative stress, protein aggregation, neuroinflammation, mitochondrial dysfunction, and apoptosis (programmed cell death), and that vitamin D and the VDR are believed to influence all of these processes [1]. That’s a remarkable breadth of action for a single receptor system.
A 2024 review in *Molecular Neurobiology* reinforced this picture, describing vitamin D3 as exhibiting a “pleiotropic” effect on brain-related disorders, meaning it influences multiple, distinct biological pathways simultaneously, including redox balance (managing oxidative stress), inflammation, energy production, and the synthesis of growth factors that support neuronal survival [2].
One of the most important, and underappreciated, findings in recent research is that the brain isn’t passively dependent on the liver and kidneys to supply it with active vitamin D. Neurons, astrocytes, and microglia (the brain’s immune cells) all express VDR. Some of these cells also express the enzymes CYP27B1 and CYP24A1, meaning they can both produce and break down active vitamin D locally [9].
This local vitamin D signalling appears to be paracrine or autocrine in nature, meaning brain cells may be producing active vitamin D to communicate with neighbouring cells, or even with themselves [9]. This is a fundamentally different picture from the classic story of vitamin D travelling from kidneys to bones.
But here’s the crucial implication: if your circulating levels of 25-hydroxyvitamin D are low, which is the raw material the brain’s local machinery needs, then even this sophisticated local activation system can’t do its job properly. The brain’s on-site factory still needs the starting materials [9].
The animal research here is consistent and striking, though it’s important to be clear about its limitations (more on that below).
A 2014 study published in *PNAS*, one of the most cited journals in science, used ageing Fischer 344 rats to model the range of human vitamin D blood levels from deficient to sufficient [8]. For five to six months, rats were fed diets containing low (100 IU/kg), medium (1,000 IU/kg), or high (10,000 IU/kg) vitamin D3. They were then tested in the Morris water maze, a well-validated test of hippocampal-dependent learning and memory.
Rats fed the high vitamin D diet achieved blood levels in the “sufficient” range and significantly outperformed both the low and medium groups on maze reversal, a particularly demanding task that detects subtle changes in memory [8]. The researchers also ran hippocampal gene expression analysis, finding that high vitamin D upregulated pathways involved in synaptic transmission, cell communication, and G protein function. Basal synaptic transmission in the hippocampus was also enhanced in the high vitamin D group, a finding that directly corroborates the behavioural improvements [8].
A 2024 study in *Nutrients* examined 36 adult Wistar rats who were made vitamin D deficient through a combination of paricalcitol injections and a vitamin D-free diet [3]. When their vitamin D status was restored through cholecalciferol supplementation (at either 1,000 or 10,000 IU/kg in their diet), the vitamin D-restored groups showed a significantly faster learning curve in both the Morris water maze and radial arm maze tests compared to the deficient group [3]. Crucially, supplementation also decreased the density of inflammatory microglia and reactive astrocytes in the hippocampus and amygdala, brain regions central to memory and emotional learning [3].
Evidence grade: Early stage for specific brain effects, this is compelling animal data, and the mechanisms are well-reasoned, but human trials with cognitive outcomes have been inconsistent (see below).
Some of the most mechanistically detailed research concerns vitamin D’s potential role in Alzheimer’s disease specifically. Two hallmarks of Alzheimer’s are the accumulation of amyloid-beta (Aβ) plaques between neurons and the formation of neurofibrillary tangles from hyperphosphorylated tau proteins [15].
A 2014 study, using both mice and cell cultures, found that even mild vitamin D deficiency (20–30% below normal levels) significantly increased amyloid-beta peptide levels [12]. The mechanism appeared to work through two separate pathways: first, vitamin D deficiency increased the activity of an enzyme called BACE1, which promotes the cleavage of amyloid precursor protein into harmful Aβ fragments; and second, it downregulated neprilysin, one of the brain’s primary amyloid-clearing enzymes [12]. In other words, low vitamin D simultaneously turned up amyloid production and turned down amyloid clearance. The finding was replicated in cultured N2A cells supplemented with 25(OH) vitamin D3, suggesting the effect is direct rather than mediated through some other pathway [12].
A 2023 review in *Nutrients* expanded on this picture, noting that vitamin D and its analogues have been examined for their potential to modulate amyloid precursor protein processing, tau phosphorylation, neuroinflammation, and oxidative stress in the context of Alzheimer’s disease [15]. The review also highlighted that in central Europe, more than 50% of people over 60 are not sufficiently supplied with vitamin D, a statistic that puts the potential public health relevance of this research into sharp relief [15].
Evidence grade: Promising to early stage, the mechanistic evidence is strong in animal models and cell studies. Human intervention trial data remains limited and inconsistent.
Here’s where intellectual honesty is essential, and where the “VDR mutation” angle becomes critically important for understanding why human trials have often disappointed.
A 2020 randomised, double-blinded, placebo-controlled trial published in *Nutrients*, one of the more rigorous human studies in this space, tested the effects of vitamin D3 supplementation, vitamin D2-enriched mushrooms, standard mushrooms, and placebo on cognition and mood in healthy elderly adults over six months [11]. Despite raising circulating vitamin D levels, the trial did not find significant cognitive benefits from vitamin D3 supplementation in this population [11].
A 2021 review in *Nutrients* specifically addressed this disconnect between compelling animal data and inconsistent human findings [9]. It concluded that while 1,25(OH)2D and 25-hydroxyvitamin D have both been detected in different brain regions, and while VDR is expressed throughout the CNS, “the influence of supplementation has not yet been demonstrated to have a direct impact on neuronal functions” in human trials [9].
Why? Several explanations are proposed. First, the populations studied often already had relatively adequate vitamin D levels, making benefit harder to detect. Second, trials have varied enormously in dose, duration, baseline vitamin D status, and cognitive outcome measures [4]. Third, and this is the key insight for this post, VDR itself is a genetic variable. The VDR gene has known polymorphisms (variants) that affect how well the receptor binds and responds to vitamin D [2]. Someone with a less-responsive VDR variant may have textbook-normal blood levels but still experience significantly blunted downstream signalling in brain tissue.
This is why the headline “standard vitamin D3 fails to protect the ageing brain” deserves nuance rather than dismissal. It’s not that vitamin D doesn’t matter, it’s that blood levels are an imperfect proxy for whether vitamin D is actually working inside brain cells. For a subset of people, the bottleneck isn’t circulating vitamin D at all, it’s the receptor [1][2].
A 2023 review in *Nutrients* raised an intriguing angle: vitamin D analogues, synthetic modifications of the vitamin D molecule, differ from standard vitamin D not just in pharmacokinetics but in their binding affinity to the VDR itself [15]. Analogues including alfacalcidol, paricalcitol, and eldecalcitol have been studied for their effects on mechanisms relevant to Alzheimer’s disease, and some show different receptor-binding profiles that may be relevant when the VDR is functioning suboptimally [15].
This is early-stage science in the context of brain health, but it points toward a future where personalised vitamin D therapy, based on genetic VDR profiling, might look quite different from a standard D3 supplement [15].
Let’s be direct about the gaps, because there are significant ones.
The human intervention data is thin and conflicted. Almost all the mechanistic evidence comes from animal studies or cell cultures. The human trials that do exist are relatively short (six months is the longest well-designed one we found [11]), use inconsistent doses, and often study populations who aren’t severely deficient to begin with. We don’t yet have long-term, large-scale RCTs that specifically examine vitamin D supplementation’s effect on cognitive decline in people who are genuinely deficient at baseline.
VDR polymorphisms haven’t been systematically studied in supplementation trials. This is a glaring gap. If VDR function varies significantly between individuals, and we know from genetic research that it does [2], then lumping everyone together in a supplementation trial and looking for average effects is a flawed design. Future trials need to stratify by VDR genotype to understand who actually benefits.
We don’t know the optimal dose for brain health. The doses studied in animal models use weight-adjusted amounts that don’t translate simply to human dosing. The 2014 PNAS rat study used 10,000 IU/kg of diet [8], that’s per kilogram of food, not body weight, and scaling that to human equivalents is not straightforward.
The blood-brain conversion question remains open. While we know the brain has local vitamin D activation machinery [1][9], we don’t fully understand how efficiently supplementation raises active vitamin D levels within specific brain regions in living humans. Blood levels of 25(OH)D are a reasonable proxy for overall vitamin D status, but they may not reflect what’s actually happening inside neural tissue.
The difference between prevention and treatment is unclear. Can raising vitamin D levels slow or reverse cognitive decline once it’s underway, or does it primarily matter for prevention earlier in life? We don’t have good human data to answer this [4].
So what does a sensible, informed person actually do with all of this?
First, the foundational point: vitamin D deficiency is genuinely common, particularly in the UK, particularly in people over 60, and particularly in anyone who spends most of their time indoors or covers their skin. More than 50% of people over 60 in central Europe are insufficiently supplied [15]. In the UK, the figures are comparable, and deficiency is near-universal in winter months. This is not a theoretical risk, it’s a lived reality for a large proportion of exactly the people reading this post.
Supplement D3 daily, this is the sensible default. Vitamin D is fat-soluble, but genuine toxicity requires sustained mega-doses far above normal supplementation levels. At the doses most commonly used, 1,000 to 4,000 IU daily, the risk of supplementation is vanishingly small compared to the well-documented risks of deficiency. The research is clear that low vitamin D is associated with worse cognitive outcomes, compromised neuroprotection, and accelerated neurodegeneration [1][2][4][5]. The risk of staying deficient is real and documented. The risk of supplementing at normal doses is not.
Take D3 with vitamin K2. The research in this set of papers focuses on D3 and VDR, but vitamin D influences calcium metabolism, and K2 is needed to direct that calcium appropriately. This isn’t elaborated in the papers above, but it’s standard practice in evidence-based supplementation.
Understand the VDR piece, and don’t assume your blood test tells the whole story. If you’ve been supplementing consistently and your 25(OH)D levels look fine on paper, that’s good, but it doesn’t guarantee optimal VDR signalling in brain tissue. The receptor itself is a variable [2][1]. If you have a family history of neurodegenerative disease, or if you’re particularly motivated about this, genetic testing for VDR polymorphisms is available and becoming more accessible, though its clinical implications are still being worked out.
Focus on sufficiency, not just adequacy. The 2014 PNAS study found that rats needed blood levels in the genuinely sufficient range, not just borderline adequate, to show cognitive benefit [8]. Most UK guidelines set “adequate” at 50 nmol/L; many researchers in the brain health space suggest that optimal neuroprotective levels may be higher, perhaps 75–100 nmol/L. This isn’t firmly established in human trials, but it’s a reasonable inference from the available mechanistic data.
Don’t wait for perfect evidence. The human trial data on vitamin D and cognition is genuinely limited and sometimes disappointing, but absence of conclusive trial evidence for cognitive benefit is not the same as evidence of no effect. The mechanistic case is compelling, the deficiency rates are real, the supplement is safe, and the downside of supplementing appropriately is essentially nil. A brilliant, practical friend with access to this research would say: supplement D3 daily, aim for genuinely sufficient blood levels, and don’t be fooled into thinking normal blood levels mean your brain’s VDR system is firing on all cylinders.
The lock matters as much as the key. Make sure you’re giving your brain enough raw material to work with, and watch this space, because the science of personalised vitamin D therapy is genuinely moving.
[1] Vitamin D and Neurodegenerative Diseases Such as Multiple Sclerosis (MS), Parkinson’s Disease (PD), Alzheimer’s Disease (AD), and Amyotrophic Lateral Sclerosis (ALS): A Review of Current Literature (2025). *Current Nutrition Reports*. DOI: 10.1007/s13668-025-00663-y | https://pubmed.ncbi.nlm.nih.gov/40464816/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12137432/
[2] Tremendous Fidelity of Vitamin D3 in Age-related Neurological Disorders (2024). *Molecular Neurobiology*. DOI: 10.1007/s12035-024-03989-w | https://pubmed.ncbi.nlm.nih.gov/38372958/
[3] Cholecalciferol Supplementation Impacts Behavior and Hippocampal Neuroglial Reorganization in Vitamin D-Deficient Rats (2024). *Nutrients*. DOI: 10.3390/nu16142326 | https://pubmed.ncbi.nlm.nih.gov/39064769/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11279879/
[4] Vitamin D in Central Nervous System: Implications for Neurological Disorders (2024). *International Journal of Molecular Sciences*. DOI: 10.3390/ijms25147809 | https://pubmed.ncbi.nlm.nih.gov/39063051/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11277055/
[5] Vitamin D and Brain Health (2026). DOI: 10.1007/978-3-032-04357-3_5 | https://pubmed.ncbi.nlm.nih.gov/41219598/
[8] Vitamin D prevents cognitive decline and enhances hippocampal synaptic function in aging rats (2014). *Proceedings of the National Academy of Sciences (PNAS)*. DOI: 10.1073/pnas.1404477111 | https://pubmed.ncbi.nlm.nih.gov/25267625/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4205629/
[9] Role of Vitamin D in Cognitive Dysfunction: New Molecular Concepts and Discrepancies between Animal and Human Findings (2021). *Nutrients*. DOI: 10.3390/nu13113672 | https://pubmed.ncbi.nlm.nih.gov/34835929/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8620681/
[10] Vitamin D3 and Alzheimer’s Disease: Amyloid Plaques and Cognitive Dysfunction (2022). DOI: 10.1007/s11011-022-01086-2 | https://pubmed.ncbi.nlm.nih.gov/36156759/
[11] The Effects of Vitamin D-Enriched Mushrooms and Vitamin D3 on Cognitive Performance and Mood in Healthy Elderly Adults: A Randomised, Double-Blinded, Placebo-Controlled Trial (2020). *Nutrients*. DOI: 10.3390/nu12123847 | https://pubmed.ncbi.nlm.nih.gov/33339304/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7766163/
[12] Impact of Vitamin D on amyloid precursor protein processing and amyloid-β peptide degradation in Alzheimer’s disease (2014). https://pubmed.ncbi.nlm.nih.gov/24192346/
[14] Vitamin D3 as a potentially modifiable factor in mild cognitive impairment (2018). *Journal of Neurology, Neurosurgery & Psychiatry*. DOI: 10.1136/jnnp-2018-319021 | https://pubmed.ncbi.nlm.nih.gov/30279214/
[15] Vitamin D and Its Analogues: From Differences in Molecular Mechanisms to Potential Benefits of Adapted Use in the Treatment of Alzheimer’s Disease (2023). *Nutrients*. DOI: 10.3390/nu15071684 | https://pubmed.ncbi.nlm.nih.gov/37049524/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10096957/
This article is for informational purposes only and does not constitute medical advice. Food supplements should not be used as a substitute for a varied and balanced diet and healthy lifestyle. If you are pregnant, breastfeeding, taking medication or have a medical condition, consult your doctor before taking any supplement. These statements have not been evaluated by the Food and Drug Administration (FDA) or the Medicines and Healthcare products Regulatory Agency (MHRA). This product is not intended to diagnose, treat, cure, or prevent any disease.
Quick Read Multiple large studies show that people who drink coffee regularly have lower rates of dementia and cognitive decline.
Quick Read Creatine is known as a muscle-building supplement, but research suggests it may be equally important for brain health,
Free guide: The research behind brain health supplements
Plus our weekly research digest — straight to your inbox.
Just leave your email address.