Vitamin A for Vitamin A Deficiency
In people who are actually deficient, vitamin A is a proven, evidence-backed treatment that reverses night blindness and dry eye and, in undernourished young children, lowers illness and death. It is not a remedy for the general, well-nourished population, and excess preformed vitamin A is harmful in pregnancy.
Why this grade7-layer evidence engine
The grade rests on a strong, consistent evidence base aimed at the right population. A large Cochrane review of children aged 6-59 months (PMID 28282701; 47 trials, over 1.2 million children) found periodic high-dose supplementation cut all-cause death by roughly 12% (RR 0.88, 95% CI 0.83-0.93) and about halved measles incidence, and an independent meta-analysis (PMID 21868478, ~195,000 children) reported a similar mortality benefit. A pregnancy meta-analysis (PMID 21069707) and the WHO guideline (PMID 24575452) point the same direction.
It lands at moderate rather than top-tier because the picture is not uniform. The large DEVTA trial in north India (PMID 23472988, ~1 million children) found no clear mortality benefit (about RR 0.96). It was folded into the Cochrane pooled estimate, which still favors supplementation, and the muted result is attributed to a milder baseline deficiency and rising measles-vaccine coverage.
Authorities agree on both the use and its limits. WHO strongly recommends high-dose supplementation only where deficiency is a public-health problem; NIH Office of Dietary Supplements names night blindness and xerophthalmia as the hallmark signs; and Cleveland Clinic describes high-dose then tapered repletion as standard care. The benefit applies to genuinely deficient or high-risk people, not routine use in well-fed adults, and the UK NHS warns pregnant women to avoid retinol supplements because excess preformed vitamin A is teratogenic.
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 = 0.783
- tier_from_score — 依分數區間映射至 tier letter
- apply_hec_rules — 高品質 SR/MA 顯示 positive (3 篇 > 0 negative)
- tier_strict_requirement_check — Tier 條件達標,未降階
- detect_disputes — 偵測到 0 個 hard + 0 個 soft dispute
- decide_status — 依 tier + dispute 結果決定 status