Panax Ginseng for Fatigue
For Panax ginseng (Asian/Korean ginseng) taken on its own, the evidence does not show a reliable benefit for fatigue. The positive findings often cited come from a different species (American ginseng) or from multi-herb formulas, not from Panax ginseng alone.
Why this grade7-layer evidence engine
The grade is low because the strongest "positive" fatigue data do not actually test Panax ginseng. The landmark Wisconsin N07C2 double-blind RCT (PMID 23853057, n=364) used American ginseng, Panax quinquefolius, at 2000 mg/day and improved cancer-related fatigue at 8 weeks (P=.003) -- a different species. Pooled and network meta-analyses reporting benefit (PMID 35776997, SMD 0.33; PMID 37180654, SMD 0.32) drew heavily on American-ginseng trials or mixed both species together, so they cannot be read as evidence for Panax ginseng itself.
When the studies are examined more carefully, the signal weakens. The largest and most recent synthesis (PMID 36730693, 19 RCTs) found no significant overall effect (SMD -0.36, 95% CI -0.82 to 0.11, p=0.13); only herbal-formula and chronic-fatigue subgroups were positive. A cancer-related-fatigue meta-analysis (PMID 35184068) was borderline (SMD -0.21, 95% CI -0.42 to 0.00, the upper bound touching zero). Effect sizes throughout are small.
Authorities reflect this caution. Mayo Clinic explicitly notes the fatigue benefit came from American ginseng, not Panax ginseng; the UK NHS warns that herbal products are not tested like standard medicines and can interact with other drugs; the US FDA stopped evaluating the relevant GRAS notice and endorses no claim. WHO's monograph describes ginseng as a traditional restorative for tiredness and weakness, but that reflects long-standing use rather than proof of efficacy.
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 = 0.605
- tier_from_score — 依分數區間映射至 tier letter
- apply_hec_rules — 高品質 SR/MA 顯示 positive (2 篇 > 1 negative)
- tier_strict_requirement_check — | B→D 因 scope.conflation_risk=true 且 L11 獨評較低 (B7-2 tier cap)
- detect_disputes — 偵測到 1 個 hard + 0 個 soft dispute
- decide_status — 依 tier + dispute 結果決定 status