Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: St. John's wort is more effective than placebo for treating depressive disorders

ACP J Club. 1999 May-June;130:60. doi:10.7326/ACPJC-1999-130-3-060

Source Citation

Linde K, Mulrow CD. St. John's wort for depression. Cochrane Review, latest version 09 Jul 1998. In: The Cochrane Library. Oxford: Update Software.



Are extracts of St. John's wort (Hypericum perforatum) as effective and safe as placebo and standard antidepressants in patients with depressive disorders?

Data sources

Studies were identified by searching MEDLINE (from 1983), PsycLIT (1987 to 1997), EMBASE/Excerpta Medica (from 1989), the clinical trial registry of the Cochrane Depression, Anxiety, and Neurosis Group, the Cochrane Complementary Medicine Field, and the private database Phytodok (Munich); by scanning bibliographies; and by contacting authors and pharmaceutical companies.

Study selection

Studies were selected if they were randomized controlled trials (RCTs) that compared St. John's wort with placebo or other antidepressants for depressive disorders. Studies with physiologic outcome measures only were excluded.

Data extraction

2 reviewers extracted study characteristics and results and assessed the quality of each study.

Main results

27 studies (2291 patients) met the inclusion criteria. 17 trials were placebo controlled. Follow-up ranged from 2 to 12 weeks in 26 studies {mean 5.5 wk}*; 1 study had an unknown length of follow-up. More patients responded to St. John's wort than to placebo (14 studies, {P < 0.001}*) (Table). No differences in responder rates occurred for St. John's wort compared with low-dose antidepressants (5 studies, {P = 0.79}*) (Table). Fewer side effects occurred with St. John's wort than with low-dose anti-depressants {28% vs 45%, P = 0.002}*.


St. John's wort is more effective than placebo for depressive disorders. Responder rates do not differ for St. John's wort and low-dose antidepressants; side effects are less common with St. John's wort.

Sources of funding: National Institute of Arthritis and Musculoskeletal and Skin Diseases and Karl und Veronica Carstens-Stiftung.

For correspondence: Dr. K. Linde, Munchener Modell-Centre for Complementary Medicine Research, Technical University/Ludwig-Maximilians-University, Kaiserstrasse 9, Munich 80801, Germany. FAX 49-89-393484.

*Numbers calculated from data in article.

Table. Responder rates for St. John's wort (SJW) vs control in depressive disorders†

Comparison at a mean follow-up of 5.1 wk Weighted event rates RBI (95% CI) NNT (CI)
SJW Control
SJW vs placebo‡ 56% 25% 140% (70 to 239) 4 (3 to 6)
Comparison at a mean follow-up of 5.6 wk RBR (CI) NNH
SJW vs antidepressants 51% 52% 3% (-12 to 16) Not significant

†RBR = relative benefit reduction. Other abbreviations defined in Glossary; RBI, RBR, NNT, NNH, and CI calculated from data in article.
‡The trials were heterogeneous; therefore, a random-effects model was used.


Many physicians are skeptical of “alternative” remedies and respond to claims made by advocates of these remedies with calls for more evidence—specifically, evidence from RCTs. This meta-analysis by Linde and Mulrow reports the results of 27 such trials comparing St. John's wort with either placebo or active treatments. It suggests that St. John's wort is more effective than placebo and is as effective as low-dose tricyclic antidepressants (TCAs) in the treatment of depression. The mechanism of action seems to be similar to that of conventional antidepressants—inhibition of reuptake of monoamines (1).

The authors point to the small sample sizes, brief follow-up periods, and inadequate doses of conventional antidepressants used in many (but not all) of the trials reviewed. How serious are these concerns? It is instructive to contrast the St. John's wort trials with those comparing more conventional antidepressants: TCAs and selective serotonin reuptake inhibitors (SSRIs). In a previous review (2), we found that the median sample size of TCA and SSRI trials was 64 patients. For the equivalent trials comparing St. John's wort with active treatment, the median sample size was 101. This suggests that these trials may be of higher quality than those of more conventional treatments. This point should reassure readers who may be concerned that many of the trials are published in unfamiliar German-language journals. Similarly, although low doses of TCAs were often used in the St. John's wort RCTs, 25% of trials comparing SSRIs and TCAs also used low doses. Further work suggests that the necessity of prescribing high-dose TCAs may have been exaggerated (3). Hence, although bigger and better trials would be welcome, it is unfair to condemn St. John's wort on this basis.

Although this review gives some support for St. John's wort, some questions remain unanswered. Considerable uncertainty exists about dosage—these trials had big differences in prescribed doses. Herbal remedies are often poorly regulated and tend not to receive the same postmarketing surveillance as conventional treatments. Thus, relatively little may be known about possible rare adverse effects.

Matthew Hotopf, MBBS, MSc
Institute of PsychiatryLondon, England, UK

Matthew Hotopf, MBBS, MSc
Institute of Psychiatry
London, England, UK


1. Muller WE, Rossol R. Effects of hypericum extract on the expression of serotonin receptors. J Geriatr Psychiatry Neurol. 1994;7:S63-4.

2. Hotopf M, Lewis G, Normand C. Putting trials on trial—the costs and consequences of small trials in depression: a systematic review of methodology. J Epidemiol Community Health. 1997;51:354-8.

3. Trindade E, Menon D. Selective serotonin reuptake inhibitors (SSRIs) for major depression. Part I: Evaluation of the clinical literature. Ottawa: Canadian Coordinating Office for Health Technology Assessment, 1997 Aug. Report 3E.