Vitamin D

Verdict: Mixed Last reviewed: 2026-04-24 Triangulated against anchor: Rapamycin (Probable) — Vitamin D sits one band below

TL;DR

VITAL (n=25,871) was largely null on its primary endpoints (major CV events, invasive cancer, all-cause mortality). Updated meta-analyses including VITAL show a modest but real reduction in cancer mortality (~13%) with vitamin D supplementation, but no benefit on cancer incidence, fractures (in non-deficient adults), or CV events. The popular "vitamin D for everyone" framing significantly outruns the evidence in vitamin D-replete populations. Verdict: Mixed — meaningful in deficiency, weakly supported in supplementation of replete adults.

What it is

A fat-soluble seco-steroid hormone synthesized in skin from 7-dehydrocholesterol upon UVB exposure; obtained dietarily from fatty fish, fortified foods, supplements. Active form is 1,25-dihydroxyvitamin D (calcitriol). Status assessed by serum 25(OH)D; deficiency conventionally <20 ng/mL (50 nmol/L), insufficiency 20-30 ng/mL. Typical longevity dosing: 1000-5000 IU/day, with 2000 IU/day used in VITAL.

Proposed mechanism

• Calcium / phosphate homeostasis and bone mineralization
• Immune modulation (innate and adaptive)
• Endocrine effects on multiple tissues (vitamin D receptor expressed widely)
• Possible effects on cell proliferation, differentiation, apoptosis (the cancer-mortality hypothesis)
• Cardiovascular signaling (vascular tone, inflammation)

Confidence: Established for bone/calcium homeostasis and immune effects in deficiency; Plausible for non-deficient supplementation effects.

Evidence ladder

Animal models (T3-T4)

Vitamin D deficiency models show clear bone, immune, and metabolic dysfunction. Lifespan effects in non-deficient supplementation context are less developed.

Human (T1)

This is the best-studied supplement in modern medicine. Major RCTs:

• VITAL (Manson 2019, NEJM) — 25,871 US adults, vitamin D 2000 IU/day vs placebo, ~5.3 years. Primary endpoints (invasive cancer incidence, major CV events): null. Selected secondary endpoints showed signals.
• VITAL fracture analysis (LeBoff 2022, NEJM) — vitamin D did not reduce fracture incidence in healthy non-deficient adults; women showed an increased risk of hip fractures (counterintuitive).
• Cancer mortality meta-analysis (PMC6821324) — 13% reduction in cancer mortality across pooled trials including VITAL; effect grows with longer follow-up. Statistically significant.
• All-cause mortality umbrella review (Frontiers Nutr 2023) — ~7% reduction in total mortality across 116 RCTs; mostly driven by the cancer-mortality signal.
• Mendelian randomization studies — genetic vitamin D deficiency associates with worse outcomes; supplementing replete individuals does not show the same effect, suggesting threshold biology.
• DO-HEALTH, ViDA, D2d trials — null on primary endpoints in mostly-replete populations; some signals on falls/diabetes prevention.

Population caveats

The honest summary: vitamin D is a deficiency disease drug, not a supplement-everyone-for-longevity drug. People with low 25(OH)D benefit from correction; people with normal levels see little additional benefit from supplementation in well-powered RCTs.

Confounds

• Baseline status heterogeneity — VITAL participants had mostly normal 25(OH)D; benefit may be larger in deficient subpopulations (US Black participants in VITAL showed cancer signal).
• Dose — 2000 IU/day in VITAL may have been too low for some endpoints; higher-dose trials face safety concerns (hypercalcemia, kidney stones).
• Latitude / season / supplementation outside trial — control groups often supplement themselves, attenuating effect estimates.
• The hip fracture signal in women is unexplained and concerning; methodology demands honesty about it.

Conflict of interest scan

• Academic / NIH funded for the major trials. Minimal industry conflict.
• Some vitamin D advocates (Holick et al.) have commercial ties; supplement industry pushed the "everyone is deficient" narrative beyond what evidence supports.
• Net: minimal discount on the primary RCT evidence.

Human translation

Honest decomposition:

1. In documented vitamin D deficiency (25(OH)D <20 ng/mL): correction is meaningfully beneficial for bone, immune, and possibly mortality outcomes. This is settled.
2. In replete adults supplementing for longevity: VITAL and sister trials are largely null on hard endpoints. The 13% cancer-mortality signal is real but modest; it does not translate cleanly to "supplement and live longer."
3. In older adults with osteoporosis or fall risk: combined with calcium and as part of bone-health protocols, vitamin D plays a real role; standalone supplementation in non-deficient older adults shows null fracture effects in well-powered trials.
4. Hip-fracture signal in women is a yellow flag.

The popular framing ("everyone should take vitamin D for longevity") does not match the trial evidence in 2026.

Calibrated verdict

Mixed. Tier-appropriate evidence exists but the largest RCT was largely null on primary endpoints, with selected positive signals (cancer mortality) and a counterintuitive negative signal (hip fractures in women). Per methodology section 3, this is the textbook Mixed pattern.

Compared to rapamycin (Probable), vitamin D has more human RCT volume but lacks the broad mechanism / replicated benefit pattern that a Probable verdict requires.

Compared to omega-3 (next page, similar trial pattern), the two are roughly comparable: both null on primary endpoints in major trials, both with selected secondary signals.

Confidence interval on verdict

• Could move to Probable if longer follow-up of VITAL strengthens the cancer-mortality signal AND mechanism work in deficient subpopulations sharpens. Modest probability.
• Could move to Mostly hype if subsequent meta-analyses dilute the cancer-mortality signal and the hip-fracture signal in women is replicated.
• Most likely 2-year trajectory: stays at Mixed.

Open questions

• Q: Is there a 25(OH)D threshold above which supplementation is net harmful in women (informing the hip-fracture signal)?
• Q: Does the cancer-mortality benefit appear specifically in initially-deficient subgroups, and if so should the recommendation be "supplement only the deficient"?
• Q: What is the mechanism of the apparent cancer-specific mortality reduction in the absence of incidence reduction (improved survival post-diagnosis? earlier detection? reverse causation?)
• Q: Should vitamin D recommendations be uncoupled from calcium given the renal stone risk?

Sources

• Manson et al. 2019, NEJM — VITAL primary results: vitamin D and omega-3 for primary CV/cancer prevention
• LeBoff et al. 2022, NEJM — Supplemental vitamin D and incident fractures in midlife and older adults
• Cancer mortality meta-analysis (PMC6821324)
• Vitamin D effects on all-cause mortality umbrella review, Frontiers Nutr 2023
• VITamin D and OmegA-3 TriaL principal results and updated meta-analyses (PMC7089819)
• Vitamin D effects on falls and fractures: recent mega-trials, ScienceDirect 2024
• Health effects of vitamin D supplementation: lessons from RCTs and Mendelian randomization (PMC10592274)
• Vitamin D3 and cancer risk umbrella review 2024

Produced under methodology locked 2026-04-24. Triangulated against rapamycin anchor.