St John's Wort

What it is

St John's wort (Hypericum perforatum) is a yellow-flowering plant whose extracts are sold as tablets, capsules and tinctures for low mood [1]. Trials have mostly tested standardised extracts such as LI 160, STW 3 and ZE 117, standardised to about 0.3% hypericin and 1-4% hyperforin, in daily doses of 300 to 1,200 mg a day [1][2]. It is licensed and widely prescribed for depression in several European countries [3].

How it might work

The herb contains several active compounds — hyperforin, hypericin and flavonoids — and researchers believe they act together rather than through a single ingredient [4]. Laboratory studies suggest effects on several brain signalling systems, including serotonin, noradrenaline and glutamate pathways and ion channels, but the exact contributions are not settled [4][5]. St John's wort also induces the liver enzyme CYP3A4 and the transporter P-glycoprotein, which explains many of its medicine interactions [6].

What the research says

Depression. In the largest review — 35 randomised trials with 6,993 people with major depression — St John's wort produced more treatment responders than placebo (relative risk 1.53, 95% CI 1.19 to 1.97) and lower depression scores (standardised mean difference 0.49, 95% CI 0.23 to 0.74), on moderate-quality evidence with substantial heterogeneity (I² 79% and 89%) [1]. Against antidepressants, 17 trials found nearly identical response rates (RR 1.01, 95% CI 0.90 to 1.14) and 14 trials found no difference in symptom scores (SMD 0.03) [1]. A 2017 meta-analysis of 27 trials (3,808 patients) also found response and remission similar to SSRIs, with fewer dropouts (odds ratio 0.59) [3]. A 2023 meta-analysis of 14 trials (2,270 adults) reported similar efficacy to SSRIs with fewer side effects [7]. A 2026 network meta-analysis of 163 supplement trials found the extract ZE117 reduced depression scores versus placebo (SMD 1.02 lower, 95% CI 1.90 to 0.15 lower) but rated the certainty low to very low [8]. An umbrella review of 22 meta-analyses found hypericum mono-preparations more effective than placebo [9], and a 2018 review of herbal psychiatric treatments rated the evidence for St John's wort in major depression as high-quality [10].

Tolerability. Side effects were about as common as with placebo (OR 0.83) and less common than with antidepressants (OR 0.67 across 11 trials, 1,946 people), including fewer gastrointestinal, nervous system, psychiatric and sexual side effects [1]. Fewer people stopped treatment for any reason than in SSRI groups (OR 0.59) [3]. Headache was the most common complaint in one large trial (41% versus 25% on placebo) [2]. Side-effect assessments in these trials were limited and rarely captured rare harms [1].

Severe depression. The 2016 review found no trial focused only on severe depression [1], and an earlier US trial in 200 people with baseline Hamilton scores of 20 or more found no difference from placebo on its main measure, although remission was more frequent with the herb (14.3% versus 4.9%) [2].

Anxiety. Two small placebo-controlled trials found no benefit: a 12-week trial in 40 people with social anxiety disorder found no difference from placebo [11], and a 28-person crossover trial of St John's wort plus kava found no effect on anxiety despite a short-lived antidepressant effect [12].

Doses used in studies

Most trials used standardised extracts of 300 mg taken three times a day (900 mg a day); across trials, daily doses ranged from 300 mg to 1,200 mg, for 4 to 12 weeks (median 6 weeks) [1]. The extract with the most trial evidence was standardised to 0.3% hypericin and 1-4% hyperforin [1]. A US trial started at 900 mg a day and increased to 1,200 mg for people who had not responded after 4 weeks [2]. Trials rarely compared different doses or extracts head-to-head [1].

Safety and interactions

Common side effects: in trials the herb was about as likely as placebo to cause side effects [1]. Headache was the most frequent complaint (41% versus 25% on placebo in one trial), and nervous-system side effects were somewhat more common than with placebo overall [2][1].

Upper limits: no safe upper limit or toxic dose is established; trials used up to 1,200 mg a day for up to 12 weeks, and longer-term safety is not established [1][2][3].

Medicine interactions: the herb induces CYP3A4 and P-glycoprotein, lowering blood levels and effects of many medicines [6]. With combined oral contraceptives, four studies found lower hormone exposure and more bleeding between periods, and one found signs of ovulation despite the pill [13]. It is listed among herbs and foods that can interact with warfarin [14]. Combining it with SSRIs can increase serotonin-related adverse reactions [6], and UK guidance advises against prescribing it because of interactions, product variation and uncertain dosing [1].

Pregnancy and breastfeeding: no reliable human data; a systematic review of the ten rodent studies often used for guidance found serious methodological flaws in all of them, so safety in pregnancy is unknown — ask a clinician first [15].

Bipolar disorder and other conditions: St John's wort appears among the reported triggers of manic episodes in a systematic review of bipolar disorder, mostly from case reports; people with bipolar disorder should ask a doctor before using it [16].

Caveats and open questions

The trials are heterogeneous (I² 79-89% in the placebo comparisons), most lasted only 4-12 weeks, and half were conducted in Germany [1]. There is no research on severe depression, suicidality or long-term use [1][3]. The newest network meta-analysis rated confidence in the evidence low to very low [8]. Studies rarely compared extracts or doses head-to-head, and preparations vary in their active content [1].