Same Vitamin, Opposite Directions: K1 Tracked 14% Less Dementia, K2 Tracked 14% More

If you're researching what actually affects long-term dementia risk, a new study out of Denmark offers a rare, genuinely large-scale look at diet's role, and a useful lesson in reading a headline statistic carefully. Researchers tracked nearly 55,000 adults for up to 28 years and found that two forms of the same vitamin, K1 and K2, pointed in opposite directions for dementia risk. But only one of those two signals held up when the researchers tried to break it.

Vitamin K1 and K2 Show Opposite Dementia Associations

The core finding is that dietary vitamin K1 (phylloquinone, the form found in leafy vegetables) was associated with a 14% lower rate of dementia in people with the highest intake compared to the lowest (hazard ratio 0.86, 95% CI 0.78-0.93), while dietary vitamin K2 (menaquinones, forms found mostly in cheese, meat, and eggs) was associated with a 14% higher rate in the highest-intake group (hazard ratio 1.14, 95% CI 1.04-1.24). Both figures come from the study's fully adjusted models, comparing the top fifth of intake (quintile 5) against the bottom fifth (quintile 1), published in the European Journal of Nutrition using data from the Danish Diet, Cancer, and Health cohort. The two forms of vitamin K share a basic chemical backbone but come from different foods and behave differently in the body, which is exactly why the researchers analyzed them separately instead of lumping "vitamin K" together, as many earlier, smaller studies had.

Why This Matters if You're Watching Your Cognitive Health

For anyone managing an aging parent's care, or simply paying attention to modifiable dementia risk factors as they age themselves, this is one of the largest and longest-running looks yet at how a specific, common nutrient tracks with dementia over a real human lifespan. Dementia affects memory, thinking, and independence, and diet is one of the few risk factors people can actually act on. A 2024 Lancet Commission estimated that up to 14 modifiable risk factors could account for around half of dementia cases worldwide, and diet touches several of them. This study adds vitamin K1 intake, easily boosted through common vegetables like spinach, cabbage, and broccoli, to that list of things worth understanding, while also showing why the K2 side of the story deserves more skepticism than headlines might suggest.

A 25-Year Cohort of Nearly 55,000 Danes

The researchers used the Danish Diet, Cancer, and Health cohort, which enrolled 54,968 dementia-free adults aged 50 to 65 between 1993 and 1997. At enrollment, participants filled out a single, detailed 192-item food frequency questionnaire capturing their typical diet over the prior year; that one snapshot was then used to estimate vitamin K1 and K2 intake for the entire follow-up period. Over a median of 25.5 years (up to 28 years maximum), researchers tracked incident dementia through Danish national health registries, capturing both hospital diagnoses and dementia-related prescriptions. By the end of follow-up, 4,802 participants (8.7%) had developed dementia. The researchers used Cox proportional hazards models, a standard statistical method for tracking how quickly an outcome occurs over time, adjusting for demographic factors, lifestyle, other dietary habits, and existing health conditions.

What This Could Mean for Someone Watching Their Diet

This is an observational cohort study, not a controlled trial, so it cannot prove that eating more vitamin K1-rich vegetables prevents dementia, or that vitamin K2-rich foods cause it. What it can reasonably support is a modest, evidence-based nudge: vitamin K1 intake, mainly from leafy greens and cruciferous vegetables, is a plausible piece of a broader dementia-conscious diet, alongside other established factors like managing blood pressure, staying physically active, and avoiding smoking. What this study should not do is scare anyone off cheese, meat, or eggs because of the K2 numbers. As the next section explains, the K2-dementia link is far less solid than the topline hazard ratio suggests, and this study is not grounds for treating any vitamin K2 supplement or food group as something to avoid.

The Vitamin K2 Signal Is Fragile Under Scrutiny

Here is the detail that separates this from a simple "good vitamin, bad vitamin" story: the vitamin K2 finding did not hold up consistently across the researchers' own checks, and the authors are explicit about that. The K2-dementia association was null (not statistically significant) in the basic age-and-sex-adjusted model and in the lifestyle-adjusted model, only reaching significance after the full set of adjustments. When the researchers restricted follow-up to the first 10 years, removing longer-term follow-up, the K2 hazard ratio dropped to 0.99 (95% CI 0.73-1.34), a null result. When they additionally adjusted for processed meat intake, a major dietary source of K2, the association weakened to a borderline 1.10 (95% CI 1.00-1.21). Vitamin K1's association, by contrast, stayed consistent across every one of the same sensitivity checks, including the 10-year-restricted analysis, where it actually strengthened. The authors themselves caution that both associations "may be confounded by dietary patterns and food sources associated with each vitamin form," meaning K2 intake may simply be a marker for something else about the foods it comes from (processed meat, saturated fat) rather than a direct effect of the vitamin itself. It's also worth noting the study measured only dietary intake, not blood levels of vitamin K, and did not sequence participants' gut microbiomes, even though menaquinones from MK-7 through MK-10 are substantially produced by bacteria, including gut bacteria and fermentation in foods like aged cheese, making any gut-microbiome explanation for the K2 signal inferential rather than demonstrated.

What Comes Next for Vitamin K and Dementia Research

The authors note this is the first longitudinal study to examine dietary vitamin K1 intake specifically against incident dementia, as opposed to cross-sectional snapshots or cognitive test scores. They call for further investigation, ideally including studies that measure vitamin K biomarkers directly in blood rather than relying on a single dietary questionnaire, and research into whether K1's effect is due to the vitamin itself or the broader nutrient package that comes with eating more vegetables. No randomized controlled trial testing vitamin K1 supplementation for dementia prevention has been referenced in this study, so that remains a real gap between an observational signal and a proven intervention.

Common Questions

Should I take a vitamin K2 supplement to prevent dementia? No. This study does not support that; if anything, its K2 finding pointed toward higher dementia rates in an association that weakened substantially under the researchers' own sensitivity checks, and no supplementation trial was involved.

Should I avoid cheese, meat, or eggs because of the K2 finding? No. The K2-dementia link became null when follow-up was restricted to 10 years and only borderline significant after accounting for processed meat intake, so it is far too fragile to justify cutting out entire food groups.

What foods are high in vitamin K1 versus K2? Vitamin K1 (phylloquinone) comes mainly from leafy green vegetables like spinach, cabbage, and broccoli; vitamin K2 (menaquinones) comes mainly from cheese, meat, and eggs, plus fermented foods.

Does eating more vitamin K1-rich vegetables definitely lower dementia risk? Not definitively. This is an observational study showing an association, not a randomized trial proving cause and effect, though the K1 signal was consistent across every check the researchers ran.

Key Takeaways

• In a Danish cohort of nearly 55,000 adults followed for up to 28 years, higher dietary vitamin K1 (from leafy vegetables) tracked with a 14% lower dementia rate, while higher vitamin K2 (from cheese, meat, and eggs) tracked with a 14% higher rate.
• The vitamin K1 finding stayed consistent across every sensitivity analysis the researchers ran; the vitamin K2 finding was null in simpler models and became null again when follow-up was capped at 10 years, so it is far less reliable.
• This is an observational study based on a single dietary questionnaire taken decades ago; it shows association, not proof that vitamin K1 prevents dementia or that vitamin K2 causes it.
• Don't change your diet based on the K2 numbers alone. The authors themselves say the associations may reflect broader dietary patterns rather than a direct effect of either vitamin.

References

Negar Ghasemifard, Marc Sim, Pratik Pokharel, et al. (2026). Associations between dietary vitamin K1 and K2 intake and incident dementia: findings from the Danish Diet, Cancer, and Health cohort. European Journal of Nutrition. https://doi.org/10.1007/s00394-026-04081-w