L-Theanine for Stress
L-theanine shows only weak, inconsistent evidence for relieving stress: people often report feeling calmer, but the most rigorous trials find that benefit is no larger than placebo and is not matched by changes in objective stress hormones.
Why this grade7-layer evidence engine
The grade stays at weak (Tier C) because of a persistent gap between how people feel and what their biology shows. The best recent trial (PMID 38758503, a 4-week double-blind RCT at 400 mg/day, n=30) found perceived-stress scores fell about 18% on L-theanine, but they fell almost exactly the same amount on placebo, with no significant difference between groups and no change in cortisol. A 2019 RCT (PMID 31623400, 200 mg/day for 4 weeks) similarly improved self-rated scales such as STAI while salivary cortisol and IgA stayed flat, a disconnect the authors flagged themselves.
Objective signals exist but do not line up with the subjective ones. A single 200 mg dose (PMID 34562208, crossover, n=16) lowered salivary cortisol more than placebo (p<0.001) and raised alpha brain-wave activity, yet did not significantly reduce subjective anxiety during the stressor. So acute dosing moves the biology but not the feeling, while chronic dosing moves the feeling but not the biology, leaving the causal story incomplete. A 2024 systematic review (PMID 39633316, 11 RCTs) reported symptom relief in psychiatric populations, but stress-specific data in healthy adults were limited.
Regulators and clinics reinforce the caution. The US FDA treats L-theanine as safe ('FDA has no questions') but endorses no stress claim, and the EU EFSA concluded that 'a cause and effect relationship has not been established.' Cleveland Clinic says only that it 'may help promote relaxation and reduce stress' while noting limited human research. Three of the four trials were small and industry-funded, adding a conflict-of-interest caveat, so the verdict is published-with-warning rather than a recommendation.
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 = 0.475
- tier_from_score — 依分數區間映射至 tier letter
- apply_hec_rules — 高品質 SR/MA 顯示 positive (1 篇 > 0 negative)
- tier_strict_requirement_check — Tier 條件達標，未降階
- detect_disputes — 偵測到 0 個 hard + 1 個 soft dispute
- decide_status — 依 tier + dispute 結果決定 status