Chondroitin for Joint Health
For general joint comfort in people without diagnosed osteoarthritis, chondroitin has essentially no direct human evidence, and regulators in the EU and UK do not consider it effective. It cannot currently be recommended for this use.
Why this grade7-layer evidence engine
This claim is rated Unverified because the evidence base does not actually address the question asked. Virtually every high-quality chondroitin trial enrolls people already diagnosed with osteoarthritis (OA), not healthy adults seeking general joint comfort. The one non-OA-adjacent randomized trial (PMID 26178634, n=50, Japan) used only 180 mg/day of chondroitin, far below the standard 1,200 mg, and found no significant difference in knee pain (VAS) versus placebo, with only a self-reported function score improving.
The supporting OA literature is itself mixed and cannot be extrapolated here. Two meta-analyses in OA patients (PMID 30859538; PMID 30879253) report modest pain reduction (SMD around -0.41 to -0.63) but with very high heterogeneity (I-squared = 94%) and suspected small-study bias, and pharmaceutical-grade benefit shrinks to SMD -0.25. A 2025 systematic review (PMID 40647198) claimed over 90% positive results, but it is low quality, OA-dominated, and likely reflects publication bias.
Regulators reinforce this caution. The EU EFSA rejected joint-health claims, ruling that OA-patient data cannot be applied to the healthy general population, and the UK NHS/NICE no longer recommends chondroitin, citing no strong evidence of effectiveness. The US FDA recognizes only its safety (GRAS) for limited food uses, not efficacy, and Mayo Clinic notes most studies show it works no better than placebo. With no valid human evidence for the non-OA population, the honest grade is Unverified.
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 = 0.403
- 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 — | C→U 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