Do herbal sleep remedies like valerian or chamomile work?

By the time you're reading about valerian or chamomile, you've probably tried the obvious things already: less coffee, the phone put away earlier, a bath before bed. Perhaps you bought a bottle of valerian or a magnesium powder, and the first few nights went better before the lift seemed to fade. Perhaps it's the chamomile you make every evening, and you've started to wonder whether it's doing anything. Either way, you'd like to know whether what you reach for is helping, and whether it's safe alongside whatever else you take.

In one trial, adults who signed up online with anxiety or insomniainsomnia Trouble falling asleep, staying asleep or waking too early, that keeps happening and affects how you feel or get on during the day. took valerian at bedtime or a dummy pill for twenty-eight days. At the end, 89 percent of those on valerian said their sleep had improved, and so had 86 percent of those on the dummy pill⁵5 Bent et al., 2006. Other valerian trials found wider gaps, so this one trial can't speak for the herb in general. When researchers gathered trials of prescription sleeping pills, they found the same pattern there too, with much of the improvement on the drug coming on the dummy pill as well²2 Winkler & Rief, 2015.

Those better nights are real, and they showed on overnight sleep recordings² as well as in how people felt. People with insomnia who took a dummy pill they believed in said they slept better than people given nothing at all⁶6 Yeung et al., 2018, so it isn't only time passing. Part of it is a bad patch mending, as bad patches tend to. Much of the rest seems to come from the people themselves, from expecting to sleep better, the small act of taking something and the attention of being in a study². That part is yours to keep, with or without a capsule in your hand. So the question to ask of any remedy is how much it adds on top.

If you're awake now with the bottle beside you, wondering about another capsule, it's unlikely to add much to what the first one did. Valerian may make you drowsy, though that isn't proven³3 NCCIH, 2025, so please keep it away from alcohol, a sleeping pill or any other sedativesedative Something that slows activity in the brain and makes you feel drowsy. tonight.

The clearest reason to think more won't help comes from a large trial in Norway. There, 405 adults with insomnia took valerian or a dummy pill every night for two weeks, and the dose was 3,600 milligrams, far larger than the usual one. By the end, 28.7 percent on valerian and 21.2 percent on the dummy pill⁷7 Sateia et al., 2017 had met the study's mark for better sleep, a gap too small to be sure it wasn't chance.

That trial and one other led the American Academy of Sleep Medicine to suggest that doctors not use valerian⁷ for trouble falling or staying asleep. Three other reviews went back over the trials and found no evidence that it treats insomnia⁹¹⁰¹¹9 Valente et al., 202410 Leach & Page, 201511 Ell et al., 2023, at most a possible lift in how people rate their sleep.

Not everyone reads the trials the same way. An earlier team pooled six trials and found people on valerian 80 percent more likely to report better sleep⁵5 Bent et al., 2006. Then they doubted their own number. They spotted signs that small negative studies had gone unpublished⁵, and the trials that recorded sleep in a laboratory showed no consistent change. A later team, more hopeful about the herb, blames extracts of uneven quality¹²12 Shinjyo et al., 2020 and thinks the whole dried root might work more reliably. They offer that as a hypothesis, still to be tested¹².

So what to make of it? On one side are a guideline and three reviews. On the other are a pooled figure its own authors doubted and an idea that's still a hypothesis. On that balance, valerian as it's sold now probably adds little to your sleep, though the argument isn't over.

For most adults, a few weeks of valerian is within what's been tested. People have taken 300 to 600 milligrams a day for up to six weeks³ without apparent harm, and in trials it left them no groggier the next morning than the dummy pill did⁵. Nobody knows yet how it fares over years, so months of nightly capsules go past the testing. Some people get headache, an upset stomach, mental dullness, excitability or vivid dreams³.

If you've taken it nightly for a long time, stopping suddenly can bring some people anxiety, irritability, heart disturbances and sleeplessness³, so easing off gradually may suit you better than stopping at once. Liver injury has been reported only very rarely, most often when valerian was taken alongside other herbs³.

Chamomile has been tested far less than its place in so many evenings would suggest. Only one placeboplacebo A dummy treatment, such as a pill with nothing active in it. Studies compare a real treatment against one to see how much of the effect comes from the treatment itself.-controlled trial has looked at people with long-term insomnia. Thirty-four adults who had slept poorly for at least six months took chamomile capsules or a dummy pill for twenty-eight days, and their sleep diaries showed no clear difference¹¹³1 NCCIH, 2024b13 Zick et al., 2011. It was a small pilot, too small to rule out a modest effect, and it tested capsules rather than tea. In a wider mix of studies, mostly in people without insomnia, chamomile was linked with somewhat better self-rated sleep¹⁴¹⁵14 Kazemi et al., 202415 Hieu et al., 2019. The results varied widely from one study to the next, and it didn't lengthen sleep.

What the tea gives you may come mostly from the few warm minutes themselves, a cup held in both hands while the day winds down. The studies so far don't separate the cup from the herb¹⁴¹⁵, but the dummy-pill research²⁶2 Winkler & Rief, 20156 Yeung et al., 2018 suggests minutes like those can do a good share of the work.

If the chamomile shares a cupboard with other medicines, a few things matter. In the amounts you'd drink as tea, it's likely safe¹. Allergic reactions are uncommon, though some can be serious, and you're more likely to have one if you're allergic to ragweed, chrysanthemums, marigolds or daisies¹. Researchers suspect it may add to the effect of sedatives¹, though nobody has shown it does. It might also make birth control pills less effective¹.

If you take warfarin or another blood thinner, please tell whoever prescribes it before making chamomile a nightly habit. Chamomile tea has been linked with a higher risk of bleeding¹⁴4 Tan & Lee, 2021 alongside warfarin, mostly in case reports.

You may have met lavender as a spray for the pillow, or as a capsule on the pharmacy shelf. Researchers have studied the two separately, and what they found for one doesn't tell you about the other. As a scent, it's unclear whether it improves sleep¹⁶¹⁷16 NCCIH, 2025b17 Cheong et al., 2021. One review of thirty-four studies of breathing in essential oils, lavender most of all, found that sleep improved. The reviewers rated those studies at high risk of bias, though, including in whether people could tell what they were smelling. That makes it hard to tell the oil from the expectation. If you like the smell on your pillow, the US government's complementary-medicine center rates it possibly safe, listing headache and coughing as side effects¹⁶.

The evidence for the capsule comes from one lavender oil, Silexan, taken at 80 milligrams a day. German researchers gave it to adults whose anxiety came with restlessness and poor sleep, and it eased their symptoms more than a dummy pill over ten weeks⁸8 Kasper et al., 2015, though the dummy pill helped a good deal too. When researchers looked at how it improved sleep, almost all of what it added ran through eased anxiety¹⁸18 Seifritz et al., 2019 rather than sedation. So it may help mainly when worry is what keeps you awake.

Its maker paid for both trials. And a capsule labeled lavender on the shelf may not be the same preparation. Taken by mouth, lavender can cause diarrhea, headache, nausea or burping¹⁶, though in its trial side effects came about as often on the dummy pill⁸. Lavender might add to the effect of sedative drugs or herbs¹⁶. That's unproven, but please keep it away from alcohol and sleeping pills all the same.

If you've been stirring a magnesium powder into a drink at night, you've probably met the names on the tub: glycinate, threonate, citrate and the rest. None of the trials so far tested glycinate, threonate or the other forms common on shelves¹⁹19 Mah & Pitre, 2021, so no one can yet say whether one works better than another. The few trials there are gave plain tablets to older adults, and those taking magnesium fell asleep about seventeen minutes sooner¹⁹ than those on a dummy pill, without sleeping longer overall. The trials were small and the certainty is low. One of them left out people who already got plenty of magnesium, so any benefit may belong mostly to people who run low, and it says nothing about younger adults.

UK health advice says 400 milligrams a day or less from supplements is unlikely to cause harm²⁰20 NHS, 2026h, with diarrhea more likely above that.

Magnesium rarely builds up to dangerous levels in the blood, and when it does, it's mostly in people whose kidneys have failed²¹21 Lewis, 2025. So if you have kidney disease, please check with a pharmacist or doctor before taking it. It's worth mentioning any other medicines you take as well, since supplements in general can interact with them²²22 NCCIH, 2024.

If you've opened a bottle of valerian, you'll know the smell, something like old socks. It comes from compounds in the root breaking down. In one comparison, a tincture kept two months at room temperature held less than five percent of them¹²12 Shinjyo et al., 2020. The valerenic acids that makers use to standardize products hadn't changed¹². Whether the lost compounds matter for sleep isn't known, but it means a standardized label can miss what changes inside the bottle.

The label can tell you less than you might expect. In the US, herbs and magnesium count as dietary supplements. The FDA doesn't approve them before they go on sale²³23 US Food and Drug Administration, 2024b, so nobody there checked your bottle before it reached the shelf. The maker answers for what's inside, and the agency mostly steps in after problems appear. A label has to give the amount of each ingredient, except those inside a proprietary blend²³. That's how a sleep blend can list five herbs without saying how much of each it holds. And since no law limits how much of an ingredient a serving can contain²³, each name could stand for a lot or a little. Those are the US rules; other countries treat some herbal products differently.

Then there's the dose. The trials that earlier team reviewed gave people 225 to 1,215 milligrams of valerian a day⁵5 Bent et al., 2006. When researchers checked thirteen common US products, the labels recommended anywhere from 75 to 3,000⁵. That survey is old, and today's shelves may differ. Still, the dose on the bottle you're holding may be far from what those trials tested.

Hidden drugs are a separate question. Between 2007 and 2016 the FDA found 776 supplements with undeclared drug ingredients²⁴24 Tucker et al., 2018. That counts only the products it caught, and most were sold for sex, weight loss or muscle building. Nothing so far shows hidden drugs are common in sleep herbs. A trial of one extract tells you about that extract, and the bottle in your hand may hold something rather different.

If you came here with a bottle in your hand, much of what made your better nights was already yours. You expected rest, and you set the evening aside for it. A capsule adds little to that, so what matters most about it is what it might meet among the other things you take. The same goes for a pillow spray or a scoop of magnesium powder.

You could try a remedy for a few weeks, keeping it away from alcohol, sleeping pills and the medicines above. You could leave the bottle on the shelf. Or, if your nights have been poor for months, you could put the money towards CBT for insomnia. Whichever you choose, the kettle will still cool on the counter. The evening you set aside for rest was doing its share all along.