Millions of adults take daily vitamins, fish oil capsules, or antioxidant blends expecting to protect their brains as they age. A growing body of research built on one of the world’s largest health databases is challenging that assumption. Multiple analyses drawn from a UK Biobank cohort of approximately 269,229 participants have linked regular use of common supplements not to lower dementia rates, but to higher ones. The findings do not prove that popping a multivitamin causes cognitive decline, but they introduce a sharp tension between what consumers believe and what the observational data actually show.
Why the UK Biobank supplement findings demand attention now
The scale of the evidence is what sets these results apart from smaller, inconclusive studies that have circulated for years. A prospective cohort study published in the American Journal of Clinical Nutrition used the UK Biobank to follow vitamin D status and supplement habits in roughly 269,000 adults, estimating how those factors related to incident Alzheimer’s disease and vascular dementia over time. In that analysis, regular use of vitamin D and multivitamin products was associated with higher dementia incidence after multivariable adjustment, a pattern that runs counter to the widespread belief that such pills are neuroprotective. The full report on vitamin D exposure underscores that these are associations, not proof of causation, but the signal is difficult to ignore.
A separate UK Biobank study, this time focusing on antioxidants such as vitamin C, vitamin E, selenium, and zinc, applied similarly rigorous methods. Researchers defined supplement users based on baseline questionnaires and then adjusted for other common pills, including calcium, glucosamine, and fish oil. Even after that additional control, the analysis still found that people reporting antioxidant use had higher rates of dementia during follow-up than non-users. This work on antioxidant intake again stops short of claiming that supplements cause disease, but it adds to a consistent pattern emerging from the same large, community-based cohort.
These are not fringe results from small clinical samples. They emerge from one of the most heavily studied population databases in modern epidemiology, and they converge on a signal that challenges conventional wisdom: the supplements people take to stay sharp may correlate with worse cognitive outcomes, not better ones. One hypothesis that could explain the pattern involves interactions between nutrients and genes that affect lipid metabolism and amyloid processing. A UK Biobank analysis of fish oil capsules, for example, reported that the association between supplement use and vascular dementia varied by APOE ε4 dosage, with risk elevation appearing in certain genotype groups. If similar gene–supplement interactions exist for vitamins and antioxidants, the aggregate data may be masking subgroup-specific harms that only surface in carriers of high-risk alleles.
Randomized trials and observational studies tell different stories
The tension between these observational findings and the randomized trial literature is real and unresolved. Randomized controlled trials are designed to minimize confounding by assigning participants to treatment or placebo groups, whereas observational cohorts rely on what people choose to do in real life. In the supplement space, those two approaches are currently pointing in different directions.
The Prevention of Alzheimer’s Disease by Vitamin E and Selenium Trial, known as PREADViSE and conducted as an ancillary study to the SELECT prostate cancer trial, tested whether long-term antioxidant supplementation could reduce dementia risk in asymptomatic men. Participants were randomly assigned to receive vitamin E, selenium, both, or matching placebos and were followed with structured cognitive screening. According to the published PREADViSE results, neither vitamin E nor selenium lowered dementia incidence, but the trial also did not detect the kind of risk increase hinted at by the UK Biobank analyses.
Other randomized trials exploring various vitamins, minerals, and nutrient blends in older adults have largely echoed this “no clear benefit, no clear harm” pattern. A systematic review and meta-analysis of such trials, published in the American Journal of Medicine, concluded that most supplement interventions failed to deliver reliable cognitive advantages when compared with placebo. The National Center for Complementary and Integrative Health has similarly advised clinicians that current evidence does not support recommending most dietary supplements for dementia prevention.
That leaves a gap. On one side, randomized data suggest that common supplements generally do not help preserve cognition. On the other, large-scale observational data from routine users in the community raise the possibility that some products may be linked to worse outcomes. Bridging the gap between “no benefit” and “possible risk” is now a central task for dementia researchers and public health agencies.
Gaps in dosage data and genetic stratification
Several limits in the existing evidence prevent researchers from drawing firm causal conclusions. One major constraint is how supplement exposure is measured. The UK Biobank relies on baseline questionnaires that ask participants whether they regularly use specific products, but these instruments do not capture exact dosage, brand, or duration of use. Nor do they reliably track changes in behavior over time, such as starting or stopping a multivitamin after a new diagnosis. Without detailed dose and adherence information, it is difficult to test whether higher intakes correspond to higher risks, a pattern that would strengthen the case for causality.
Another limitation is that no single primary cohort has yet presented fully harmonized, head-to-head hazard ratios for all major supplement categories using identical statistical models. Instead, vitamin D, antioxidants, fish oil, calcium, and glucosamine have each been evaluated in separate publications with slightly different covariate sets and analytic choices. That fragmentation makes it risky to compare risk estimates across nutrients or to rank them by apparent safety or danger. It also complicates efforts to understand whether the observed associations cluster in particular biological pathways, such as oxidative stress or vascular health.
Genetic stratification remains an underdeveloped but crucial frontier. The APOE ε4 allele is the strongest common genetic risk factor for late-onset Alzheimer’s disease, and its influence on lipid metabolism and inflammation could plausibly interact with nutrient exposures. Yet only a handful of analyses, mostly involving fish oil, have explicitly examined whether supplement–dementia associations differ by APOE status. Large-scale gene–environment interaction studies that incorporate polygenic risk scores and detailed supplement histories could clarify whether certain genotypes face heightened vulnerability when exposed to specific products.
Confounding by indication is another challenge. People who choose to take supplements often differ systematically from those who do not. They may have more chronic conditions, be more health-conscious, or respond to early, subtle symptoms by adding pills in hopes of protection. Even with extensive adjustment for lifestyle and medical history, residual confounding can remain. Reverse causation is also possible: early cognitive changes might prompt supplement use, rather than supplements causing those changes. Sophisticated analytic techniques, such as marginal structural models and negative control analyses, can help probe these issues but have not yet been widely applied across all supplement categories.
What consumers and clinicians can do now
In the absence of definitive answers, both consumers and clinicians face a practical dilemma. For individuals, the emerging data argue against assuming that over-the-counter vitamins or antioxidant blends are harmless add-ons to a healthy lifestyle. While the current evidence does not justify panic or abrupt discontinuation for everyone, it does support a more cautious, individualized approach. People without a documented deficiency or a clear medical indication may want to reconsider routine, long-term use of brain-health supplements marketed on the promise of sharper memory or dementia prevention.
Clinicians, meanwhile, can use these findings as an opportunity to revisit supplement discussions during routine visits. Rather than focusing solely on potential benefits, conversations should now include the possibility of neutral or even adverse cognitive effects, especially when patients are taking multiple products. Reviewing medication and supplement lists, clarifying the reasons for use, and aligning decisions with guideline-supported strategies-such as blood pressure control, diabetes management, physical activity, and smoking cessation-may offer more reliable paths to brain health than continuing pills with uncertain value.
For researchers and policymakers, the UK Biobank results highlight the need for better-designed trials that reflect real-world supplement patterns, including combinations of products and long durations of use. Embedding randomized experiments within large cohorts, improving exposure measurement, and integrating genetic data could finally clarify whether common supplements are innocuous, modestly harmful, or risky only for certain subgroups. Until then, the safest assumption may be that when it comes to protecting the aging brain, there are no magic capsules-and that the most powerful interventions are still the unglamorous fundamentals of cardiovascular and metabolic health.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
