Vitamin A for Immune Function
Vitamin A's effect on immunity depends entirely on who takes it: high-dose supplementation sharply cuts death and measles severity in vitamin-A-deficient children, but it provides no proven immune benefit in well-nourished adults and raises lung-cancer risk in smokers. There is no good evidence it "boosts immunity" in healthy, replete people.
Why this grade7-layer evidence engine
The grade is Weak/Disputed because the evidence is high-quality but splits sharply by population rather than pointing one way. In deficient children the signal is strong: an Imdad 2017 Cochrane review (PMID 28282701, ~1.2 million children) found all-cause mortality RR 0.88 and measles incidence RR 0.50, the foundational Sommer 1986 Indonesian trial (PMID 2871418) cut child deaths ~34%, and a measles-treatment Cochrane review (PMID 15495002) lowered measles mortality RR 0.39. WHO accordingly gives a rare strong recommendation, but only where deficiency is endemic.
The counter-evidence is what holds the grade down. In smokers and asbestos-exposed workers, the CARET trial (PMID 8602180) raised lung-cancer risk (RR 1.28) and was stopped early for harm, and the ATBC trial (PMID 8127329) found an 18% increase (RR 1.18). Benefit also fades where deficiency is uncommon: the ~1-million-child DEVTA trial (PMID 18502287) showed a non-significant RR 0.96, and in HIV-positive pregnant women vitamin A gave no benefit and possibly increased transmission (PMID 9605804).
Regulators and clinics mirror this nuance. EFSA permits only a modest 'contributes to normal immune function' claim, while NHS warns against retinol supplements in pregnancy, and Cleveland Clinic and Harvard stress that most people get enough from food and that 'more is not better' — explicitly flagging cancer risk in smokers. For a typical well-nourished adult seeking an immune boost, the evidence does not support supplementing, and for smokers it warns against it.
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 = 0.492
- tier_from_score — 依分數區間映射至 tier letter
- apply_hec_rules — 無高階證據可裁決
- tier_strict_requirement_check — Tier 條件達標，未降階
- detect_disputes — 偵測到 1 個 hard + 0 個 soft dispute
- decide_status — 依 tier + dispute 結果決定 status