In this series we’ve found doses hidden in blends, inflated on front labels, diluted inside formulas, and borrowed from other extracts’ trials. This is the fifth mechanism, and it’s the strangest one: a dose with no trial behind it at all — born, as far as we can tell, from a podcast.
Apigenin is a flavonoid found in chamomile. The NIH label database mentions it 3,029 times — almost always as a standardization marker inside chamomile extracts. But 84 products list apigenin as an ingredient in its own right, and because that’s a countable number, we read every one. A census of a newborn niche.
The wave has a date on it
The database records when each label entered it. For apigenin, the timeline is blunt:
- Through 2021: a trickle — 33 products in a decade, mostly chamomile-based lines listing their apigenin content. Median apigenin per serving: 5 mg.
- 2022 onward: 51 products — 61% of everything that exists. Twenty-three arrived in 2025 alone. Median: 35 mg, with a hard cluster at exactly 50.
What happened between the two eras wasn’t a clinical trial. In late 2021 a hugely popular neuroscience podcast described a three-part sleep stack — magnesium, L-theanine, and 50 mg of apigenin — and the shelf reorganized itself around it. Among dedicated apigenin bottles with a stated dose, eight of nine sell exactly 50 mg. Six products now combine apigenin, theanine and magnesium in one capsule; a couple are literally named after the stack. One product sells 500 mg — ten times the podcast number.
The apigenin census · all 84 products · NIH database
When apigenin products entered the database
Every product listing apigenin as its own ingredient, by year of entry.
Before the line: median 5 mg, mostly chamomile-based lines. After: a hard cluster at 50 mg — eight of nine dedicated bottles sell exactly that. PubMed trials of isolated apigenin for sleep, at any dose: zero.
What the evidence says: we looked, and there isn’t any
We searched PubMed on 30 August 2026, and we’ll show our queries so you can repeat them:
- “apigenin AND sleep AND randomized controlled trial” — zero results.
- “apigenin AND insomnia AND humans” — six results: a mouse study, a fruit-fly screen, two network-pharmacology papers on herbal decoctions, a chemistry paper about passionflower, and a general review. No human trial of isolated apigenin for sleep. Not one, at any dose.
The human evidence that does exist is for chamomile — the whole extract. A randomized pilot in chronic insomnia used 270 mg of chamomile extract twice daily (PMID 21939549). Here’s the part worth sitting with: the pre-2022 chamomile-based products in our census deliver a median of 5 mg of apigenin. The new pills sell ten times that — a dose that appears in no study of the whole herb and no study of the isolated compound, because the latter doesn’t exist.
So the 50 mg number on those labels traces to exactly one source, and it isn’t a journal.
To be fair to apigenin
Untested is not disproven. There’s animal work — including a 2026 mouse study of magnesium-plus-apigenin combinations — and a plausible mechanism story. Apigenin at 50 mg might work; nobody has checked. Our objection isn’t to the molecule. It’s that an entire product category standardized on a dose that no human has ever been given in a controlled setting — and the labels give no hint of that. A buyer comparing a 50 mg apigenin bottle to a 600 mg ashwagandha bottle sees two numbers that look equally scientific. One has a decade of trials behind it. The other has an episode.
Where we stand
Full disclosure, sharper than usual: apigenin is on our published list of rejected ingredients — we considered it and said no, for exactly the reason above, before we ran this count. And our product competes with the very stack-in-a-capsule products this article describes. We are not neutral here; the data is public so you don’t have to trust us.
Limits: the census covers the NIH database, which is voluntary — marketplace-only brands are underrepresented, and this niche lives on marketplaces, so the real wave is likely larger than our count. Entry dates record when a label was added to the database, which can lag the product’s launch. And absence of trials was checked by the searches above — if a registered trial published after our search date, we’d genuinely like to hear about it.
How to read an apigenin label in ten seconds
- Ask what the 50 is. It’s not from a trial. If the label or the brand’s site claims otherwise, ask them for the citation — we couldn’t find one.
- Chamomile is the studied form. If you want the evidence-backed version of this flavonoid, it comes as 270 mg of chamomile extract twice a day — carrying a few milligrams of apigenin, not fifty.
- A stack in a capsule inherits the weakest link. Magnesium and theanine have trials. Bundling them with an untested third ingredient doesn’t lend it their evidence.
How we counted, so you can check us
We pulled all 3,029 “apigenin” matches from the DSLD on 30 August 2026 and kept the 84 on-market products listing apigenin as an ingredient line (not merely a standardization note); all 84 labels were read in full. Doses are per serving; 11 products hide the dose in a blend. Era analysis uses the database’s label entry dates. PubMed queries are quoted verbatim above with their result counts as of the same date.
The database is free and needs no account. If you repeat this and get different numbers, we’d like to know.
†These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease. This article is educational and is not medical advice; if you have a medical condition or take medication, talk to your healthcare provider.
GABA is the older cousin of this phenomenon: two dosing camps an order of magnitude apart, one descending from manufacturer-run studies, the other from bodybuilding folklore.