L-Theanine's Best-Replicated Effect Is Focus — Not the Calm It's Sold For
Across 31 randomised trials in 1,168 people, a single 200 mg dose improved choice reaction time. The calm everyone markets it for was modest for stress and non-significant for anxiety — and fully null at 450-900 mg in patients with generalised anxiety.
Sterling beside a capsule and a glass of bright green tea against pure black
We planned a different article. The intended angle was that "calm" supplements underdose L-theanine relative to the 200 mg used in trials. Then we checked actual product labels — Nature Made Wellblends and Vitafusion both deliver 200 mg per serving. The mainstream category has converged on the studied dose. The claim was false, so we cut it. What the evidence actually shows turned out to be more useful.
The effect that replicates is focus
A 2026 meta-analysis pooled 31 randomised controlled trials covering 1,168 people. Its clearest finding: a single 200 mg dose, taken 30 to 60 minutes before cognitive testing, significantly improved choice reaction time, with a standardised mean difference of 0.51 (95% CI 0.25 to 0.77). That is a moderate, well-replicated effect on attention — the strongest thing L-theanine has going for it.
The calm is the weaker half
In that same analysis, acute stress reduction came in at an SMD of just 0.31, and the authors note it was largely influenced by studies at high risk of bias. Anxiety effects were inconsistent and non-significant except in one study on psychotic anxiety.
A separate trial went further in exactly the population the anxiety marketing implies. Sarris 2019 gave 46 patients with diagnosed generalised anxiety disorder adjunctive L-theanine at 450 to 900 mg — more than double the usual dose — for eight weeks. It did not outperform placebo for anxiety (p = 0.73) or insomnia severity (p = 0.35). Fully null, at a higher dose, in the people it is sold to.
Where it does help sleep
There is a real sleep signal, and it is narrower than the marketing suggests. Hidese 2019, a randomised placebo-controlled crossover trial at 200 mg/day for four weeks, found improvements versus placebo on three PSQI subscales: sleep latency, sleep disturbance, and use of sleep medication. Useful — but a long way from a sedative.
The real buyer trap
It is not milligrams. It is proprietary blends that hide how much is actually inside, and price spread: the identical amino acid ranges from roughly $0.07 to $0.53 per dose across the products we track. Buy plain L-theanine at 200 mg, take it 30 to 60 minutes before focused work, and pair it with your coffee if you like. Buy the dose you can see.
SOURCES
- Gerolymos C et al. — Mol Psychiatry 2026: systematic review and meta-analysis of 31 randomised controlled trials (n = 1,168); a single 200 mg dose 30-60 min before testing improved choice reaction time (SMD = 0.51, 95% CI 0.25-0.77); acute stress reduction modest (SMD 0.31) and largely influenced by high-risk-of-bias studies; anxiety effects inconsistent and non-significant (PMID 42410082)
- Sarris J et al. — J Psychiatr Res 2019: 46 patients with DSM-5 generalised anxiety disorder, adjunctive L-theanine 450-900 mg for 8 weeks; no benefit over placebo on the HAMA (p = 0.73) or Insomnia Severity Index (p = 0.35) (PMID 30580081)
- Hidese S et al. — Nutrients 2019: randomised, placebo-controlled, double-blind crossover trial, 200 mg/day for 4 weeks in 30 healthy adults; PSQI subscales for sleep latency, sleep disturbance and use of sleep medication reduced versus placebo (all p < 0.05) (PMID 31623400)