Supplement Studies  /  evidence, scored
Evidence File 037 · St John’s Wort · Method v0.14 of 4 claims researched

St John’s WortHypericum perforatum · 4 claims

It works. That is exactly what makes it dangerous — because people take it instead of telling their doctor, and it disables a long list of prescription medicines.

4Claims scored
7Sources
Varies by extractStudied dose
Jul 26Reviewed

Claim 01


Treats mild to moderate depression
7.6/10
Moderate·Cochrane: 29 trials · 5,489 patients

A Cochrane review of 29 trials covering 5,489 patients found St John’s Wort more effective than placebo and comparable to standard antidepressants for mild-to-moderate depression, with fewer side effects than SSRIs.

Its own adverse event profile is genuinely mild: gastrointestinal irritation 0.6%, allergic reactions 0.5%, fatigue 0.4%, restlessness 0.3% — an incidence roughly ten times lower than synthetic antidepressants.

A more sceptical review characterises the efficacy as slight at best. Both readings can be true: a real effect, modest in size, in a condition with a large placebo response.

The working

8Quantity
7Quality
7Consistency
9Directness

(8 × 0.25) + (7 × 0.35) + (7 × 0.25) + (9 × 0.15) = 7.6
One of the better-supported efficacy claims on this site. Held below 8 for the extract variability problem described in claim 04.

Claim 02


Treats severe depression
3.6/10
Weak·Insufficient evidence · and a mania signal

There is insufficient evidence to determine efficacy in more severe depression. The Cochrane finding applies to mild-to-moderate presentations and does not extend upward.

Separately, use has been associated with the induction of mania or hypomania in people with bipolar disorder. That is a meaningful signal for anyone self-treating low mood without a diagnosis.

The working

2Quantity
4Quality
2Consistency
8Directness

(2 × 0.25) + (4 × 0.35) + (2 × 0.25) + (8 × 0.15) = 3.6
Scored as support for the claim. Reviews state directly that evidence is insufficient in more severe depression.

Claim 03


Reduces the effectiveness of prescription medicines
8.4/10
Supported·Mechanism established · documented clinical failures

Hyperforin activates the pregnane X receptor, which induces CYP3A4 and P-glycoprotein. The result is that susceptible drugs are metabolised and cleared faster, so less medicine reaches or stays in the bloodstream.

Human studies and case reports document reduced blood concentrations of cyclosporine, tacrolimus, warfarin, digoxin, indinavir, nevirapine, simvastatin, methadone, midazolam, amitriptyline, theophylline, phenprocoumon and fexofenadine. It reduced the active metabolite of irinotecan in cancer patients. It did not alter carbamazepine, dextromethorphan, mycophenolic acid or pravastatin.

These are not theoretical. In 2000 a St John’s Wort–cyclosporine interaction caused acute rejection in two heart transplant patients. There are published cases of breakthrough bleeding and unintended pregnancies in women taking oral contraceptives. Combined with SSRIs, SNRIs, MAOIs, tramadol or triptans it carries serotonin syndrome risk.

The working

8Quantity
8Quality
9Consistency
9Directness

(8 × 0.25) + (8 × 0.35) + (9 × 0.25) + (9 × 0.15) = 8.4
The highest score on this site, and it is for a harm. Mechanism, pharmacokinetic studies and real clinical consequences all line up.

Claim 04


The product on the shelf matches the extract that was trialled
3.1/10
Evidence against·Composition varies · and it determines the risk

Cochrane warns that hypericum preparations can differ considerably, because composition depends on the raw material, the extraction process and the solvent used. A trial-tested extract is not automatically interchangeable with a capsule or tea bought off a shelf.

For most supplements that is a quality annoyance. Here it is a safety problem, because the degree of CYP3A4 induction correlates significantly with hyperforin content — which varies between products and is not on the label. Two products at the same milligram dose can carry materially different interaction risk.

The working

1Quantity
4Quality
1Consistency
8Directness

(1 × 0.25) + (4 × 0.35) + (1 × 0.25) + (8 × 0.15) = 3.1
Scored as support for the claim. Cochrane raises the variability directly, and the pharmacology explains why it matters more here than for most supplements.

Dosage as studied


Most-studied regimen
Varies by extract hyperforin content is what matters

The degree of CYP3A4 induction correlates with hyperforin content, which differs between products. Dose alone does not tell you the interaction risk.

Trial extracts
Cochrane notes preparations differ considerably because composition depends on raw material, extraction process and solvent.
What drives interactions
Hyperforin activates the pregnane X receptor, inducing CYP3A4 and P-glycoprotein. Induction correlates with hyperforin content, not with milligrams of herb.
Do not combine with
SSRIs, SNRIs, MAOIs, tramadol and triptans — serotonin syndrome risk.
Tell your doctor
If you take any prescription medicine at all. The interaction list includes contraceptives, anticoagulants, immunosuppressants, antiretrovirals and some chemotherapy.