5 Herbal Remedies That Work

Montage of St. John's Wort, African Plum, Horse Chestnut, Red Yeast Rice, and Feverfew

Here at Skeptic North, we’ve often been a sharp critic of those herbal remedies that are unable to withstand the scrutiny of science. Yet nature does indeed house many pharmacologically active compounds, and it stands to reason that some of them will have medicinal value. So today, we’re going to turn the tables and look at 5 herbal remedies that have held up well in repeated studies and are generally regarded as effective.

1)  St. John’s Wort for Depression

If there’s one herbal medicine that consistently gets high marks for effectiveness, it’s St. John’s Wort as a treatment for mild depression. A 2008 Cochrane Review looked at 29 trials totaling over 5000 patients, including 18 comparisons with placebo and 17 comparisons with synthetic standard antidepressants, and found significant effects in both cases. The Natural Medicines Comprehensive Database indicates that St. John’s Wort is likely as effective as both first generation antidepressants (low-dose tricyclics) and the current generation of selective serotonin reuptake inhibitors like Prozac, Zoloft, and Paxil.

The National Center for Complementary & Alternative Medicine (NCCAM), the CAM research division of the US National Institute of Health, summarizes the science as follows:

There is scientific evidence that St. John’s wort may be useful for short-term treatment of mild to moderate depression. Although some studies have reported benefits for more severe depression, others have not; for example, a large study sponsored by NCCAM found that the herb was no more effective than placebo in treating major depression of moderate severity.

The American College of Physicians basically agrees, though are notably more cautious in their recommendation, pointing to the lack of standardization and the potential for adverse reactions with other drugs. Standardization is a concern with any natural medicine — preparations can vary widely, so it can be difficult to ascertain what the correct dose should be. As for adverse reactions with other drugs, the list is indeed long — in fact, Health Canada issued a special bulletin on the topic, cautioning against combined use with oral contraceptives, certain blood thinners, anti-epilepsy drugs, HIV medications, antidepressants, and a list of others. For patients not on other medications however, St. John’s Wort is reasonably well tolerated, with a list of side effects that doesn’t seem much worse than that of conventional antidepressants.

Of course the above only applies to St. John’s Wort’s use in treating depression. Like most natural medicines, it is often recommended for a much longer list of conditions, for which the evidence is not so compelling.

St. John’s Wort is probably not effective for

  • AIDS
  • Hepatitis C
  • Irritable Bowel Syndrome
  • Polyneuropathy

There’s insufficient evidence of St. John’s Wort’s effectiveness for

  • Obsessive Compulsive Disorder
  • PMS
  • Seasonal Affective Disorder
  • Smoking Cessation
  • Alcoholism
  • Wounds & Burns

2)  African Plum for Enlarged Prostate

Pygeum, an extract of the bark of a plum tree native to sub-saharan Africa, has good evidence of efficacy for reducing the symptoms of enlarged prostate. To be clear, we’re talking about benign prostatic hyperplasia (BPH) — common, age-related prostate enlargement — and not prostate cancer. If you have prostate cancer, note that conventional medical treatments have a 5 year survival rate of 96% and are by far your best bet. But if BPH is your problem, African Plum may be just the thing for you.

A 2002 Cochrane Review is positive but admits the lack of placebo controlled, larger trails. Since that time, one has been initiated and should be completed this year — the CAMUS trial, which will test both African Plum and Saw Palmetto, another BPH remedy which at this time has somewhat less good evidence. Still, the Natural Medicines Comprehensive Database lists it as “likely effective” and lists only nausea and abdominal pain as known side effects. So far, there are no known adverse reactions with other drugs.

It’s also worth noting the list of traditional uses for which pygeum has not been shown to be effective: for the treatment kidney disease, malaria, stomach ache, fever, madness, and as an aphrodisiac.

3)  Horse Chestnut for Varicose Veins

Chronic venous insufficiency (CVI) is the inability for the veins to return all of the spent blood to the heart, and often results in swelling, aches, and varicose veins. The Cochrane Review for horse chestnut seed extract has been updated twice since it was first conducted, most recently in 2010, and continues to suggest that horse chestnut is effective at reducing the symptoms of CVI. While calling for better powered studies, the meta-analysis concluded that horse chestnut is “an efficacious and safe short-term treatment for CVI.”

NCCAM goes one step further, pointing to “preliminary evidence that horse chestnut seed extract may be as effective as wearing compression stockings.”

Of course, compression stockings don’t come with quite the list of side effects — from nausea & dizziness to kidney damage — that horse chestnut does. This shouldn’t be surprising considering that the raw seed is poisonous to humans — the medicinal extracts must remove the primary toxic constituent, esculin. While generally safe when prepared in this manner, it is still only recommended for usage terms of less than 12 weeks, which raises the question of its practical therapeutic usefulness, as CVI is typically chronic.  Additionally, horse chestnut is unsafe for children (who, admittedly, don’t often suffer from varicose veins) and pregnant women (who certainly do) — as well as anyone taking anticoagulants/antiplatelets, anti-diabetes drugs, or lithium, with which it can interact.

Finally, the long list of traditional uses for which horse chestnut has not been shown to be effective: hemorrhoids, phlebitis, diarrhea, fever, enlarged prostate, malaria, dysentery, eczema, menstrual spastic pain, soft tissue swelling from bone fracture and sprains, complaints after concussion, cough, arthritis, rheumatism, skin ulcers, and lupus.

4)  Red Yeast Rice for High Cholesterol

This is a pretty neat one. To quote Science Based Pharmacy:

Red yeast rice is created by fermenting rice with the Monascus purpureus yeast. What’s remarkable about this fermentation process is one of the by-products: A chemical called monacolin K is produced, which is chemically identical to the drug lovastatin (sold by Merck under the brand name Mevacor).

So there’s not really any question that it works — red yeast rice will yield the same cholesterol lowering effect as lovastatin for a given dose. Of course, the same problems with dosing plague red yeast rice as they do other natural medicines, i.e. how to determine the equivalent amount of red yeast rice required to match a standardized dose of lovastatin. This is not insignificant, as there’s simply no way for the consumer to know how much lovastatin is in a given preparation.

Of additional concern, the kidney toxin citrinin has been found in some red yeast rice products.  On the flipside, there seems to be some evidence that a known side effect of statin drugs — the elevation of certain dangerous liver enzymes — is not present for red yeast rice. Whether this advantage will withstand better quality trials remains to be seen.

5)  Feverfew for Migraines

Feverfew, a variety of chrysanthemum, is a traditional analgesic reputed to reduce both pain and fever. NCCAM summarizes the science as follows: “Some research suggests that feverfew may be helpful in preventing migraine headaches; however, results have been mixed and more evidence is needed from well-designed studies.” A 2000 Cochrane Review that looked at five trials encompassing 343 migraine sufferers came back inconclusive, though did support feverfew’s reasonable safety profile.

The Natural Medicines Comprehensive Database agrees, noting that “feverfew might be better tolerated than some conventional migraine drugs used for prophylaxis.” It calls feverfew “possibly effective,” stating that “there is some evidence that taking feverfew orally can reduce the frequency of migraine headaches and reduce symptoms of pain, nausea, vomiting, and sensitivity to light and noise.” But they also point out conflicts in the data, as well as typical herbal medicine dosing issues, such as the fact that “some feverfew tablet products can contain little or no feverfew.”

Final Thoughts

Prior to doing the research for this article, I had a vague sense that there were a relatively large number of effective herbal medicines, and was actually surprised by just how few ended up making the cut. Even coming up with a Top 5 list required some nuance (consider the weaker evidence for feverfew), and it would have been a considerable struggle to make it a Top 10 list — at least if we’re relying on science instead marketing and the wishful thinking of herbalists.

First, we need to subtract out those that work but come with an unacceptable safety profile (e.g. kava). Then those with early promise that failed to show significant effects in larger scale trials (e.g. ginko biloba). What we’re left with is a very short list of herbal remedies that are promising but need more research — Valerian for Insomnia, Hawthorn for Heart Failure, and Saw Palmettto for BPH.  There may be a few others that I’ve missed (feel free to point them out in the comments), but it’s hardly the “Nature’s Pharmacy” that proponents of phytotherapy claim.

Which says less about nature itself than it does about a basic assumption held by herbalists: that nature is better at producing medicines than science.

It’s certainly true that nature is better at producing novel compounds that have therapeutic value, and a very large percentage of our pharmaceutical research is indeed naturally sourced. A recent paper by the National Cancer Institute found that:

…the utility of natural products as sources of novel structures, but not necessarily the final drug entity, is still alive and well. Thus, in the area of cancer, over the time frame from around the 1940s to date, of the 155 small molecules, 73% are other than “S” (synthetic), with 47% actually being either natural products or directly derived therefrom. In other areas, the influence of natural product structures is quite marked, with, as expected from prior information, the antiinfective area being dependent on natural products and their structures.

So the issue is not that nature doesn’t contain therapeutic remedies, it’s that they’re not often usable in their natural form — either because they’re not safe enough or not effective enough. Isolating and refining the active ingredient through the pharmaceutical development process allows us both to improve drug safety, and to better control efficacy through improved drug purity and more precise dosing.

Indeed, we’ve seen this even in our survey — in the long list of adverse drug interactions with St. John’s Wort, the poisonous nature of raw Horse Chestnut Seed, the kidney toxin in Red Yeast Rice.  And these are the best of the best — the herbal remedies that have been held up to the light and passed.  If the rest of them are less effective and less safe than these, how can anyone in good conscience make a general case for herbal medicine?

Original photos courtesy of Wikipedia (St. John’s Wort, African PlumHorse Chestnut, Red Yeast RiceFeverfew) under Creative Commons.

16 Responses to “5 Herbal Remedies That Work”

  1. Information Is Beautiful has a great online app here:

    Interestingly, when you filter for Plants/Herbs (using the Show Me tab on the top-right), it shows nine items above the “worth it” line, including green tea for cholesterol.

    Red Yeast Rice is found below the “worth it” line, in the Conflicting Evidence area of the graph.

    The data behind the visualization is also available.

  2. itzac says:


    The very existence of this field of study puts the lie to the idea that Big Pharma ignores natural cures.

    Did you ignore ASA on purpose? It’s kinda the go-to in these sorts of discussions.

  3. Loop 11 says:

    Some follow-up thoughts:

    a) Most herbs (crude, often dried, plants after all) perform poorly when applied to acute/serious/organic disease. This is a huge factor skewing overall perspectives on efficacy, especially when herbs are often matched ‘head-to-head’ with pharmaceuticals when being researched. We graduated from using plants as the main disease managing medicines 100 years ago for good reason.

    b) If more trials were applied to herbs for low-level self-resolving issues (common cold, insomnia, headache, stomachache, etc.) it would be demonstrated that they do far surpasses the misapplied overkill that pharmaceuticals often elicit in these non-emergency situations.

    c) The biggest problems the herbal medicine field faces are rampant misinformation (aka marketing hustle and the internet slurry, or ‘experts’ in name or title only spreading bad information), conventional medical professional’s knee-jerk anti-response (that which we don’t know, we usually criticize improperly), and turf-war issues (with the exception of a handful of botanicals, misapplied herbal medicine typically does nothing – leading to disappointment, whereas misapplied pharama – well no need to go into the thousands of deaths yearly by ‘oops, wrong pharmaceutical’); use herbs for inconveniences; use drugs/surgery for acute/organic disease management.

    d) Excellent book on the subject: Charles Kane’s Herbal Medicine: Trends and Traditions…I highly recommend it as resource covering the most popular herbal medicines; it neither overstates nor understates well-applied herbal medicine.

  4. Marion says:

    What about ginger for nausea? The Mythbusters seemed to think it had merit for seasickness.

  5. Herbalists have been boasting for years that St John’s Wort is as good as SSRIs. Unfortunately it now seems that SSRIs are essentially ineffective in mild to moderate depression.

    It follows that St John’s Wort is equally ineffective/ Even NCCAM is very dubious about it.

    I’d say that it was a curious choice to list among “Herbs that work”

    • Erik Davis says:

      David, I didn’t base inclusion on the boasting of herbalists, but on the results of the Cochrane analysis, which found significant effects. The comparison to SSRI’s was merely illustrative, and the fact that the tide of research may be turning on them doesn’t change the results for SJW.

      As an expert, you can diasagree with the Cochrane Collaboration, but as a layperson I’m going to trust the consensus until it changes.

  6. Bryan says:

    Cochrane reviews are generally very well done. About the only concerns I would raise with them is the age of some you’ve listed (new data may be uncovered) and the amount of data the reviewers had to work with. None-the-less, you did miss a few of the more notable examples:

    1) Ginger & ginger extracts for nausea and motion sickness.
    2) The various willow species that contain acetacylic acid (aspirin) for headache
    3) Melatonin for jet lag.

    The major problem with “natural” treatments is that controlling doseage and toxicity is nearly impossible. The amount of the active ingredient(s) can vary plant-to-plant and preparation-to-preparation, making consistent and effective dosing difficult. A larger concern is unintended toxicities and drug interactions; plants and plant extracts contain thousands of potentially biologically active compounds, of which one or two will have the desired medical properties. The effects of the others may range from some form of beneficial effect, to no effect, to long-term toxic effects, or even right through to acute toxicity.

  7. Well even the evidence that you cite for St John’s Wort is pretty heal-hearted. But the Cochrane review on which you rely is from 2008. That is before it was discovered that SSRIs had little or no efficacy for mild/moderate depression. The new data make a world of difference.

  8. Try again
    Well even the evidence that you cite for St John’s Wort is pretty half-hearted. But the Cochrane review on which you rely is from 2008. That is before it was discovered that SSRIs had little or no efficacy for mild/moderate depression. The new data make a world of difference.

  9. Bob Carroll says:

    In the Skeptic’s Dictionary entry on the Cochrane Collaboration I write: “While the Cochrane Collaboration is an excellent place to begin one’s research on a health-care treatment, it is not an infallible guide and should be considered within the context of all the available evidence regarding a treatment.” Within the context of all the available evidence St. John’s wort has feeble support for the treatment of depression. Tthe plum tree bark has no controlled studies done on it. I don’t think I’ll be using horse chestnut any time soon despite its making your cut. I think I’ll continue with my lovastatin, thank you; and, given what you write about feverfew, I don’t see why it’s even on the list.

    Another guide for recent research on herbal bedicine is “Trick or Treatment” by Ernst and Singh.

    Finally, you may as well include willow bark (aspirin) and foxglove (digitalis) as herbal treatments, if you’re going to include red yeast rice.

    • Erik Davis says:

      Thanks Bob. I’m well aware of the limitations of Cochrane analysis, and did not rely solely on it. For each of these, I checked the natural medicines comprehensive database first, then Cochrane, then NCCAM, and they all basically agreed. Many of these are also called out by Ernst himself here – http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2249806/.

      That said, I agree feverfew is a stretch, and say as much. And I personally wouldn’t use any of these given the risk / benefit profile. But the criteria here was raw efficacy, and I do think the article fairly captures the weight of the research.

      • Bob Carroll says:

        I was surprised that Ernst and Singh listed ginkgo (for dementia & memory improvement) and echinacea (for reducting severity of colds) as “good,” but maybe more data has come out since publication of their book. And, like you say, “raw efficacy” isn’t the only thing to consider. There may be side effects; there may be adverse reactions with other herbs or drugs; controlling dosage can be a problem with herbs; and then there is always the possibility that there is a safer, more reliable, alternative available from the drugstore.

  10. Looks like red yeast rice are not as reliable as promised – this new report finds a great variance in the actual dose of lovistatin and the presence of the poison you mentioned above


  11. Miguel says:

    Re: itzac’s comment

    “Big Pharma” certainly doesn’t ignore “natural” remedies. Most of the products on health food store shelves are manufactured by large drug companies. I shake my head sometimes when I here people shopping for herbs and vitamins complain about pharmaceutical companies and doctors trying to “poison” them or conspiring to take their money.

  12. badrescher says:

    I’d like to suggest that the NCCAM is not great source for accurate information: http://icbseverywhere.com/blog/first-years-archives/nccam-national-scam/.

  13. Hello Erik
    I too have questions about red yeast rice. The jury’s still out on this, and in the U.S. it is no longer legal to sell supplements that contain more than trace amounts of the active ingredients of red rice yeast. These ingredients have apparently been removed from Cholestin marketed in the U.S. (Hypocol, another product containing red yeast rice is no longer being sold in the United States.)

    And your accurate warning about the lovastatin comparison likely makes red yeast rice a poor fit for the Top Five list


  • Erik Davis

    Erik is a technology professional based in Toronto, focused on the intersection of the internet and the traditional media and telecommunications sectors. A reluctant blogger, he was inspired by the great work Skeptic North has done to combat misinformation and shoddy science reporting in the Canadian media, and in the public at large. Erik has a particular interest in critical reasoning, and in understanding why there’s so little of it in the public discourse. You can follow Erik's occasional 140 character musings @erikjdavis