Galphimia glauca for Anxiety
Two double-blind RCTs suggest standardized Galphimia glauca may ease generalized anxiety about as well as lorazepam, but the evidence is weak: both trials come from a single Mexican research group with no independent replication, so it should not be treated as a proven anxiety treatment.
Why this grade7-layer evidence engine
The grade rests almost entirely on two moderate-quality, double-blind randomized trials of a standardized galphimine-B extract in generalized anxiety disorder. The first (PMID 17562493; n=152, 4 weeks) found the herb anxiolytically non-inferior to lorazepam from week one with better tolerability. The second (PMID 22828921; n=192, 15 weeks) reported a small effectiveness edge over lorazepam (0.686 vs 0.588, p=0.0003) without dependence or withdrawal.
Those are genuinely encouraging signals, but the evidence base is thin and fragile. Both trials were run by the same government-funded Mexican investigators, with no independent or multi-center replication, and the 15-week study lost nearly half its participants (191 enrolled, 104 completed), which weakens its superiority claim. There is no Cochrane review or independent meta-analysis, so the findings cannot yet be generalized.
No major regulator (FDA, EFSA, NHS, WHO) or clinical authority (NIH ODS, Mayo, Cleveland Clinic, Harvard, APA/ADAA) has evaluated or commented on it, which is typical for an obscure Mexican herb rather than a red flag. Lorazepam comparisons describe trial design only and do not make this a substitute for a prescription benzodiazepine; theoretical additive sedation with benzodiazepines, other CNS depressants, or serotonergic drugs is plausible, and there are no safety data in pregnancy, breastfeeding, or children.
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 = 0.436
- 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 0 個 hard + 0 個 soft dispute
- decide_status — 依 tier + dispute 結果決定 status