Listen to this article:

Authorised Register wording is lawful only with its conditions of use — it is not proof of broad disease prevention in replete adults. The observational HRS cohort links routine supplementation to accelerated decline in cognitive test scores specifically in people with normal baseline 25(OH)D: a cautionary harm signal (association, not proven causation). Educational material — not medical advice.

Cultural hook (not a legal-claim source): An X post by Roger Seheult (status 2098081191236931938) and a Washington Post angle put the HRS warning front and centre: in older adults who already had normal vitamin D levels, routine supplementation was associated with slightly faster cognitive decline — the “supplements could harm the aging brain” frame. Seheult himself stresses: observational ≠ proven causation. Whether a pill equals cutaneous UVB synthesis (it does not) is background; the primary point here is the cognition harm signal — developed in section 5 from primary sources.
Two columns: authorised vitamin D claims and what they do not open
Fig. 1. Left: authorised physiological functions and conditions. Right: the boundary — no licence for broad disease prevention beyond Article 14 and the Register.

1. Authorised Article 13(1) claims — exact Register wordings

According to the Union Register and Commission Regulation (EU) No 432/2012, the following Article 13(1) claims are authorised for vitamin D (exact English Register text):

  • Vitamin D contributes to normal absorption/utilisation of calcium and phosphorus
  • Vitamin D contributes to normal blood calcium levels
  • Vitamin D contributes to the maintenance of normal bones
  • Vitamin D contributes to the maintenance of normal muscle function
  • Vitamin D contributes to the maintenance of normal teeth
  • Vitamin D contributes to the normal function of the immune system
  • Vitamin D has a role in the process of cell division

Condition of use (all of the above): the claim may be used only for food which is at least a source of vitamin D as referred to in the claim SOURCE OF [NAME OF VITAMIN/S]… in the Annex to Regulation (EC) No 1924/2006. “Source of” a vitamin/mineral means at least 15% of the nutrient reference value (NRV). For vitamin D the NRV is 5 µg (Annex XIII to Regulation 1169/2011), so the “source” threshold is 0.75 µg on the applicable basis (per 100 g/ml or per daily portion for supplements, under the claims rules).

2. Article 14 — children and disease-risk reduction

  • Calcium and vitamin D are needed for normal growth and development of bone in children (Art. 14(1)(b); Reg. 983/2009) — at least a source of calcium and vitamin D.
  • Vitamin D contributes to the normal function of the immune system in children (Art. 14(1)(b); Reg. (EU) 2016/1389) — at least a source of vitamin D.
  • Vitamin D helps to reduce the risk of falling associated with postural instability and muscle weakness. Falling is a risk factor for bone fractures among men and women 60 years of age and older (Art. 14(1)(a); Reg. (EU) 1228/2014) — food supplements providing at least 15 µg vitamin D per daily portion; consumer information that the beneficial effect is obtained with a daily intake of 20 µg from all sources; target group: men and women 60+.
  • Calcium and vitamin D help to reduce the loss of bone mineral in post-menopausal women. Low bone mineral density is a risk factor for osteoporotic bone fractures (Art. 14(1)(a); 1228/2014) — supplements with at least 400 mg calcium and 15 µg vitamin D per daily portion; information for women 50+ and that the beneficial effect requires at least 1 200 mg calcium and 20 µg vitamin D daily from all sources.
Important: an Article 14(1)(a) claim describes a risk factor within the authorised wording — it does not open free-form messages about cancer, COVID, dementia or “extending lifespan”.
Four layers: Register, conditions, label, science
Fig. 2. Four separate layers: Register text → quantity/target conditions → FIC/1924/2006 limits → caution from RCTs and guidelines.

3. What a claim allows on the label — and what it does not

Regulation (EC) No 1924/2006 permits only authorised claims, in the approved meaning and with conditions met. Article 7(3) of Regulation (EU) No 1169/2011 (FIC) prohibits attributing to foods the property of preventing, treating or curing a human disease, or referring to such properties — except where claims law provides otherwise. Marketing that turns “immune system” into a promise against a named infectious disease, or “bones” into osteoporosis prevention outside Article 14(1)(a), is high-risk (national authorities, civil liability).

4. Scientific “warning” layer: not an anti-claim

Low 25(OH)D often co-occurs with disease (association). That does not authorise claiming that a pill prevents disease in already-replete people. On cognition, lead with the harm signal — not with “no claim”.

  • HRS / BMJ Open 2025 (DOI 10.1136/bmjopen-2025-106050) — lead warning: prospective Health and Retirement Study cohort, n ≈ 5,065 (mean age 67.5; ~40% supplement users), ~6-year follow-up. Versus non-users, vitamin D users had accelerated decline in global cognition (−0.052 points/year; 95% CI −0.092 to −0.013) and executive function (−0.021 points/year; 95% CI −0.037 to −0.005). Sensitivity analysis: the signal was significant mainly with normal baseline 25(OH)D (p=0.004), not in the insufficient/deficient group (p=0.826). Authors do not support supplementation to prevent or slow cognitive decline in people with adequate vitamin D status. This is association, not proven causation; the endpoint is cognitive test scores — not clinical dementia incidence. It is also not a basis for stopping clinically indicated therapy on one’s own. Details: section 5.
  • VITAL (Manson et al., N Engl J Med 2019; DOI 10.1056/NEJMoa1809944): 2,000 IU/day vitamin D3 did not significantly reduce primary invasive-cancer or major cardiovascular endpoints versus placebo.
  • VITAL fractures (LeBoff et al., NEJM 2022; DOI 10.1056/NEJMoa2202106) and editorial Cummings & Rosen (NEJM 2022; DOI 10.1056/NEJMe2205993): cautious verdict on routine supplementation for fracture prevention in the studied population.
  • D-Health (Neale et al., Lancet Diabetes Endocrinol 2022; DOI 10.1016/S2213-8587(21)00345-4): monthly 60,000 IU in older Australians did not reduce all-cause mortality; authors urge caution with bolus regimens in replete people.
  • BMJ reviews: Theodoratou et al. (umbrella review, BMJ 2014; DOI 10.1136/bmj.g2035) — many observational links, weaker causal support from RCTs; Zhang et al. (BMJ 2019; DOI 10.1136/bmj.l4673) — supplementation and mortality: limited/uncertain effect in light of newer trials.
  • Endocrine Society 2024 (Demay et al., J Clin Endocrinol Metab; DOI 10.1210/clinem/dgae290): shift — empiric supplementation suggested for selected groups (including ages 1–18, ≥75, pregnancy, high-risk prediabetes) and a suggestion against routine 25(OH)D testing without established indications in generally healthy adults under 75.
Manufacturer communication boundary: correcting deficiency and authorised physiological functions ≠ “disease insurance” or “brain protection” for replete adults. Do not present RCTs as a ban on Register claims — or claims as proof of endpoints the studies did not confirm.

5. Dementia and cognition — harm signal, not a “brain claim”

“What about dementia?” is not answered by noting that Article 14 does not open a dementia claim. The primary scientific fact in this thread is an observational signal of accelerated cognitive decline among supplement users with normal baseline 25(OH)D; only alongside that sit claims law and the RCT result.

Left: no authorised EU claim for dementia. Right: HRS signal — faster cognitive-score decline in replete users
Fig. 3. Left: the Register does not authorise dementia/Alzheimer/“brain protection” claims. Right: HRS — faster decline in test scores among replete users (association; not dementia incidence).

5.1. The HRS / BMJ Open 2025 signal

The cohort study (DOI 10.1136/bmjopen-2025-106050) included about 5,065 older adults from HRS. Vitamin D supplement users had accelerated decline in global cognition (−0.052 points/year) and executive function (−0.021 points/year) versus non-users. The signal concentrated in those with normal baseline 25(OH)D. Authors state plainly that the findings do not support supplementation to prevent or slow cognitive decline in vitamin-D-adequate older adults.

What this study does not do: it does not prove causation (observational; possible confounders). It does not report clinical dementia diagnosis or Alzheimer’s disease incidence as a primary endpoint — it measures cognitive test scores. Do not rewrite the association as “the pill causes dementia”.

5.2. VitaMIND RCT — no protective cognitive benefit

VitaMIND (Corbett et al., J Am Med Dir Assoc 2025; DOI 10.1016/j.jamda.2025.105711): 24-month double-blind RCT; n=620 adults ≥50 years with mild/moderate vitamin D deficiency and early cognitive impairment. Primary outcome (executive function, Trail Making Test Part B): no significant benefit of supplementation versus placebo; secondary cognition, function and well-being likewise without significant effect. Contrast with HRS: the RCT shows no cognitive protection; the replete-elderly cohort shows a signal of possible faster decline.

5.3. Meta-analyses and certainty

Some RCT meta-analyses report mixed effects across cognitive domains; GRADE/certainty is often limited. We do not invent Alzheimer’s disease trial results or a “therapeutic proof” here.

5.4. Label law — no claim ≠ free rein for “brain” messaging

The Union Register contains no authorised health claim for preventing or treating dementia, Alzheimer’s disease or “brain aging”. Immune, bone and muscle claims are not cognitive-disease claims. Article 7(3) of Regulation (EU) No 1169/2011 (FIC) and Regulation (EC) No 1924/2006 close medicinal and unauthorised claims. Marketing such as “for dementia”, “protects the brain”, “against Alzheimer” — without authorisation — is high-risk / unlawful as a health claim. In parallel: overselling “brain benefits” while omitting the cautionary HRS signal in replete older adults is communication that sits poorly with the available evidence.

6. Practical line for manufacturers

ChannelLower riskHigher risk
LabelExact / equivalent Register wording + “source” / Art. 14 doses“Protects against cancer”, “for COVID”, “for dementia”, “protects the brain”, “against Alzheimer”, “extends life”
Web / socialEducation on deficiency + link to authorised claimMedicinal suggestion, drug comparison, cherry-picked associations
InfluencersScript with authorised text“I take it and never get sick” as a product claim
DocumentationTrack Register, EFSA UL, national maximaIgnoring quantitative conditions and target group

7. EFSA UL and Polish GIS context

The EFSA NDA Panel (2023; DOI 10.2903/j.efsa.2023.8145) established a tolerable upper intake level (UL) of 100 µg VDE/day for adults (including pregnant and lactating women) and adolescents 11–17 years; for children 1–10 years the UL is 50 µg VDE/day. In Poland, the Dietary Supplements Team (resolutions published by GIS) indicates maximum amounts in the recommended daily portion of, among others, 2,000 IU (50 µg) for the healthy adult population up to 75 years and 4,000 IU (100 µg) only for products exclusively intended for healthy adults over 75 — with a clear target-group statement on the pack. Those are composition/safety frames, not extra health claims.

Conclusion

The EU Register gives vitamin D lawful label language — bones, muscle, immunity, calcium — under the “source” condition and Article 14 doses. The science layer warns more sharply: HRS links routine supplementation to faster decline in cognitive test scores in replete older adults (association; not a dementia diagnosis), and VitaMIND shows no cognitive benefit in an RCT. Association of deficiency with disease does not open “brain protection” marketing or broad prevention in the replete. A manufacturer within the Register is on the right side of the law; one who promises dementia, Alzheimer, cancer, COVID or longevity is not.

Educational material, not legal or medical advice. For deficiency, disease or pregnancy, clinical decisions belong to a physician.

Primary sources

  1. Commission Regulation (EU) No 432/2012
  2. EU Register on nutrition and health claims (Commission PDF)
  3. Commission Regulation (EU) No 1228/2014
  4. Commission Regulation (EU) 2016/1389
  5. Regulation (EC) No 1924/2006
  6. Regulation (EU) No 1169/2011 — Art. 7(3); vitamin D NRV = 5 µg
  7. Manson JE et al. VITAL. N Engl J Med 2019
  8. LeBoff MS et al. VITAL fractures. N Engl J Med 2022
  9. Cummings SR, Rosen C. Editorial. N Engl J Med 2022
  10. Neale RE et al. D-Health. Lancet Diabetes Endocrinol 2022
  11. Demay MB et al. Endocrine Society CPG 2024
  12. Theodoratou E et al. BMJ 2014
  13. Zhang Y et al. BMJ 2019
  14. HRS cognitive cohort. BMJ Open 2025
  15. Corbett A et al. VitaMIND RCT. J Am Med Dir Assoc 2025
  16. EFSA NDA Panel. UL vitamin D 2023
  17. GIS — Dietary Supplements Team resolutions
  18. Roger Seheult — X status (conversation trigger only)