st john’s wort: what it actually does, and what it interacts with

what the trials found

last verified: august 18, 2026. the five studies below are the whole basis for what this page claims. each citation was fetched live from PubMed and Europe PMC, and funding is recorded as what the funding statement actually said — or UNKNOWN, never inferred. numbered references link to the primary sources at the bottom of the page.

does st john’s wort actually work?

for mild-to-moderate depression, the honest answer is yes — and it is worth saying plainly, because the usual framing on this substance is either uncritical enthusiasm or blanket dismissal. the RAND review is the strongest single piece of evidence: 35 trials, 6,993 patients, more responders than placebo, roughly equal to prescription antidepressants, with fewer adverse events [1]. the head-to-head meta-analysis against SSRIs reaches the same place from a different angle — indistinguishable response and remission rates, and fewer people stopping st john’s wort than stopping the SSRI [3].

three limits belong next to that. first, the effect shrinks as the trials get better: in Cochrane’s analysis the nine larger trials produced a response rate ratio of 1.28 while the nine smaller ones produced 1.87, which is the standard signature of small studies inflating an effect [2]. second, the literature is geographically lopsided — trials from German-speaking countries, where hypericum is a licensed prescription medicine, reported systematically more favourable results, and the Cochrane authors say in their own words that this "complicates the interpretation" [2]. third, and most practically: every trial in the head-to-head meta-analysis ran between four and twelve weeks, in mild-to-moderate depression [3]. there is essentially no research in severe depression, in people at high suicide risk, or over the long term [1][3].

and the reviews carry their own conflict problem. all three Cochrane authors disclosed financial relationships with Schwabe, a hypericum manufacturer [2]. the RAND review’s evidence tables record that a substantial share of the trials it pooled were industry-funded, had a manufacturer author, or were supplied the product by the manufacturer — and that funding was simply unreported for several others [1].

what does st john’s wort interact with?

this is the part that changes decisions, and it is the part most "natural remedy" coverage skips. st john’s wort activates the pregnane-X receptor, which induces CYP3A4 and P-glycoprotein — the enzyme and transporter that clear a large share of prescription medicines out of your body. inducing them means lowering the blood level of whatever they clear. the drug does not feel like it is doing anything; the other drug simply stops working.

does it make hormonal birth control stop working?

the CDC and FDA reviewers found only four studies that compared combined oral contraceptives alone against the same pills taken alongside st john’s wort — a thin evidence base, and worth saying so. but three of those four showed breakthrough bleeding, three showed reduced hormone exposure on pharmacokinetic measures, and one showed increased follicular growth and probable ovulation [4]. none of the four was powered to measure pregnancy rates, so what the literature establishes is a mechanism and a warning signal, not a failure rate you can put a number on.

one detail matters for anyone tempted to rely on a single reassuring study: the one product that showed no significant pharmacokinetic difference was a low-hypericin preparation [4]. different st john’s wort products produce different degrees of interaction, so a negative result for one brand does not clear the others.

can you take it alongside an antidepressant?

not as a solo decision. two separate risks stack here. the first is serotonergic: st john’s wort taken with an SSRI, SNRI, triptan, tramadol or MAOI adds serotonergic load and can precipitate serotonin syndrome. the second is the induction problem running underneath — it can lower the levels of other drugs you take at the same time.

there is also a stopping problem that almost nobody mentions. because the effect is enzyme induction, quitting st john’s wort abruptly while you are on an interacting drug swings that drug’s level back the other way — into toxicity. the safe version of both starting and stopping runs through a prescriber or pharmacist who can look at your whole list.

why the same herb behaves differently between brands

in the US, st john’s wort is a dietary supplement under DSHEA. it reaches the shelf without FDA review of efficacy, safety or potency, and the agency can act only after a problem has already emerged. the consequence is that hypericin and hyperforin content varies widely between brands and batches — and hyperforin content is what drives the interactions: the degree of CYP3A4 induction correlates significantly with how much hyperforin the particular preparation contains [5]. two bottles labelled identically can behave very differently, in both directions.

this is also why the european picture looks so different. in Germany and several other countries standardised hypericum extracts are licensed medicines that doctors prescribe for depression, which is both why most trial evidence comes from there and why "the studies say it works" does not automatically transfer to a US supplement aisle.

what the evidence cannot say

it cannot say whether st john’s wort helps severe depression — that population is essentially unstudied [1][3]. it cannot say what happens past twelve weeks, because the trials stop there [3]. it cannot give you a contraceptive failure rate, because no study was powered for one [4]. and it cannot tell you how much of the pooled benefit is real versus an artefact of small, manufacturer-adjacent trials from a single research culture — the Cochrane authors themselves flagged that trial size and country of origin both predicted the result [2]. the interaction findings, by contrast, are the sturdiest thing on this page: they are documented across independent regulatory reviews and confirmed even by authors employed by an st john’s wort manufacturer [4][5].

related evidence on resolv

st john’s wort sits alongside sixteen other non-prescription substances in our supplements and non-prescription substances hub, which grades each one on regulatory status, evidence and risk. for the prescription side of the same question — what psychiatric drugs are actually approved for, versus what they are prescribed for — see FDA-approved vs off-label.