Multivitamin / Multivitamin-Mineral (MVM) for All-cause Mortality
For healthy, well-nourished adults, taking a daily multivitamin does not lower the risk of dying from any cause, and the best evidence points to no benefit and a possible small increase in mortality risk from certain high-dose antioxidant ingredients.
Why this grade7-layer evidence engine
This claim earns a Counter-Evidence (D) grade because the highest-quality trials converge on no survival benefit. A Cochrane review of 78 randomized trials in roughly 296,000 people (PMID 22419320) found antioxidant supplements slightly increased all-cause mortality (RR 1.03, 95% CI 1.01-1.05), with the harm concentrated in beta-carotene and high-dose vitamin A. The 2022 USPSTF evidence report of 84 studies (PMID 35727272) likewise found no effect on all-cause or cardiovascular death, concluding evidence was insufficient for multivitamins and recommending against beta-carotene and vitamin E.
The observational signal points the same way: in the Iowa Women's Health Study of 38,772 older women (PMID 21987192), multivitamin use was tied to a modestly higher death rate (HR 1.06), though sick-user confounding limits that finding. One meta-analysis of 21 trials (PMID 23255568) hinted at a borderline benefit (RR 0.94, CI 0.89-1.00), but the result touched the null line and was not robust in sensitivity analyses, so it does not overturn the overall picture.
Health authorities echo this. The UK NHS says most people get what they need from a balanced diet and warns that taking too much or for too long can be harmful, while the Cleveland Clinic states plainly that multivitamins will not help women live longer. The US FDA permits no disease-prevention claim, and NIH ODS flags interactions (e.g., vitamin K with warfarin), reinforcing a no-longevity-benefit verdict.
Scoring transparency
All scores computed by a 7-layer evidence engine — fully auditable▸View the full decision path (audit trail)
- compute_raw_score — 加權公式: L2×0.30 + L3×0.25 + L5×0.25 + L11×0.10 + L1×0.10 = 0.443
- tier_from_score — 依分數區間映射至 tier letter
- apply_hec_rules — 高階證據未達主導 (1 positive vs 1 negative)，由 raw_score 決定
- tier_strict_requirement_check — | C→D 因 scope.conflation_risk=true 且 L11 獨評較低 (B7-2 tier cap)
- detect_disputes — 偵測到 0 個 hard + 0 個 soft dispute
- decide_status — 依 tier + dispute 結果決定 status