Articles

Echinacea for Cold and Flu

Echinacea continues to be a popular herbal product, used primarily for treating and preventing colds and flus. Sales were estimated at $132 million in the US alone in 2009, an increase of 7% over the previous year. Reports of major negative clinical trials have had only a modest and temporary effect on the popularity and sale of this herb, contradicting claims that the utility of such research is to inform consumers.

In the current issue of the Annals of Internal Medicine there is a new study of Echinacea for the treatment of cold symptoms: Echinacea for Treating the Common Cold, A Randomized Trial. I won’t hold out the punchline – the study was completely negative. But let’s put the results of this study into the context of the history of Echinacea and the clinical evidence.

History of Echinacea

Modern proponents of Echinacea frequently cite as support the claim that this plant has been used for centuries by many Native American cultures. This much is well-documented, but what is not clear is what Echinacea was used for. For this there is no clear answer, except that Echinacea was used for 15-20 different and unrelated conditions, from fatigue to snake bites. Let us consider the value of the claim for traditional use of any treatment.

Native American cultures did not have a written language, nor a tradition of science, rigorous observation, or objective confirmation. This is not a criticism – the latter points are true of every human pre-scientific culture. This is perhaps exacerbated by the dependence upon oral tradition. Further, the concept of illness and of specific ailments was very different in ancient cultures than modern concepts of disease. Add to that the challenge of proper translation, and it is extremely difficult to correlate what Echinacea was actually used for with any modern application. In addition, there is no clear record that Echinacea was used to treat upper respiratory viral infections.

So the claim for traditional use is dubious, but even if it were demonstrably true that is still a very weak justification for any specific treatment. The traditional use claim is based upon the demonstrably false assumption that centuries of use of a treatment (without rigorous testing) is compelling evidence for its efficacy. This assumption is contradicted by history, one dramatic example of which is the use of blood-letting and purging as part of Galenic medicine (the balancing of the four humors) for not just centuries but over two millennia in Western culture. Antiquity is no guarantee of efficacy. Human nature can apparently perpetuate worthless treatments indefinitely.

The journey of Echinacea from traditional use in the New World to its modern popularity as a cold remedy is a tortuous one. It was first popularized by a dubious physician and snake-oil salesman called H.C.F. Meyer. In the late 1800s he sold Echinacea as a panacea, claiming it cured everything, including cancer. He also heavily marketed the Native American connection, which was popular at the time. Meyer was unable to convince mainstream doctors of the time of his product’s value, but he was able to convince an eclectic physician by the name of Dr. King. (Eclectic physicians followed a tradition that included heavy use of herbal remedies.) King was instrumental in the popularity of Echinacea at this time and until about the 1930s.

The story of Echinacea then resurfaces in Germany in the 1930s. A German doctor by the name of Gerhard Madaus became interested in Echinacea. He was a proponent of unconventional medical treatments and a promoter of herbal medicine in Germany. He visited the US with the intention of bringing back seeds for Echinacea augustifolia (the variety used, apparently, by Native Americans) but instead brought back seeds for Echinacea purpurea – and for that reason the latter variety of Echinacea became popular in Europe.

While there were and continue to be many studies of the basic science of Echinacea, there were no rigorous clinical trials of its safety or efficacy. Modern claims that it “boosts” the immune system, or is useful for infections, are not based upon any solid clinical evidence. Given the history, there is also no particular reason to believe that Echinacea might be useful for colds, versus any other medical ailment.

Clinical evidence

This latest study add to the prior clinical research for Echinacea and cold symptoms, which is basically negative. Like any such clinical research, there are lots of small and poorly controlled studies, with mixed results. Even the larger and somewhat controlled studies have mixed results, but the largest and best controlled studies are all negative.

A 2006 Cochrane review found:

Main results
Sixteen trials including a total of 22 comparisons of Echinacea preparations and a control group (19 placebo, 2 no treatment, 1 another herbal preparation) met the inclusion criteria. All trials except one were double-blinded. The majority had reasonable to good methodological quality. Three comparisons investigated prevention; 19 comparisons investigated treatment of colds. A variety of different Echinacea preparations were used. None of the prevention trials showed an effect over placebo. Comparing an Echinacea preparation with placebo as treatment, a significant effect was reported in nine comparisons, a trend in one, and no difference in six. Evidence from more than one trial was available only for preparations based on the aerial parts of Echinacea purpurea (E. purpurea).

Authors’ conclusions
Echinacea preparations tested in clinical trials differ greatly. There is some evidence that preparations based on the aerial parts of E. purpurea might be effective for the early treatment of colds in adults but the results are not fully consistent. Beneficial effects of other Echinacea preparations, and Echinacea used for preventative purposes might exist but have not been shown in independently replicated, rigorous RCTs.

Basically – this is the random scatter of evidence that we typically find for an ineffective treatment. But I also think the reviewers are being too generous. There does seem to be a relationship between the quality of the study and the likelihood of negative results. In addition, there are further well-controlled clinical trials since this review that are also negative.

A 2008 study added to the evidence that Echinacea is not useful for prevention of colds or flus. They found no difference in incidence of cold symptoms in 90 volunteers taking Echinacea vs. placebo.

Echinacea studies for colds are divided into prevention and symptomatic treatment. Treatment trials generally fall into two categories – testing a cold virus challenge (giving subjects the cold), and treating colds acquired out in the world. Well-designed studies for cold prevention are all negative, including the additional 2008 study. Now we have another study of Echinacea for the treatment of symptoms. Patients were enrolled in the study within 24 hours of getting cold symptoms. They were treated either with Echinacea blinded, placebo blinded, Echinacea open-label, or nothing. Here are the results:

Of the 719 patients enrolled, 713 completed the protocol. Mean age was 33.7 years, 64% were female, and 88% were white. Mean global severity was 236 and 258 for the blinded and unblinded Echinacea groups, respectively; 264 for the blinded placebo group; and 286 for the no-pill group. A comparison of the 2 blinded groups showed a 28-point trend (95% CI, −69 to 13 points) toward benefit for Echinacea (P = 0.089). Mean illness duration in the blinded and unblinded Echinacea groups was 6.34 and 6.76 days, respectively, compared with 6.87 days in the blinded placebo group and 7.03 days in the no-pill group. A comparison of the blinded groups showed a nonsignificant 0.53-day (CI, −1.25 to 0.19 days) benefit (P = 0.075). Median change in interleukin-8 levels and neutrophil counts were also not statistically significant (30 ng/L and 1 cell/high-power field [hpf] in the no-pill group, 39 ng/L and 1 cell/hpf in the blinded placebo group, 58 ng/L and 2 cells/hpf in the blinded Echinacea group, and 70 ng/L and 1 cell/hpf in the open-label Echinacea group).

Bottom line – no benefit from Echinacea. In the discussion the authors desperately try to put a positive spin on the data, essentially making two points. One is that there was a non-statistical trend in the data towards a positive effect. Non-statistical trends are useless, however. There already is a fairly low bar for statistical significance in clinical trials, biasing them toward false positives. Lowering the bar further to talk up “trends” is disingenuous spin, in my opinion. Further, a close look at the randomization in this trial shows that 6 of the 10 pre-study conditions also deviated by an amount equal to or greater than the deviation in outcome. In other words, there was a random scatter in the randomization just as in the outcomes – but all of which fell below statistical significant and is therefore just noise.

Their second point is that the data do not rule out a clinically significant benefit. Again – this is just spin. Studies never rule out a benefit too small to be measured by the study. So you can say this about any negative study. This study was reasonably powered, but it was not a huge study with thousands of subjects, so if you take the entire range of possible statistical results, it’s within the realm of possibility that this study missed a clinically relevant effect. But that is not the same thing as there being evidence for such an effect, and it in no way changes the fact that this study is nothing but negative.

The lead author, Bruce Barrett, is quoted as saying:

Adults who have found Echinacea to be beneficial should not discontinue use based on the results of this trial, as there are no proven effective treatments and no side-effects were seen

This seems like an attempt to downplay negative results. In the discussion the authors argue that this one study, given the positive results in the past, should not change our thinking on Echinacea. But I disagree with their assessment of prior research. They reference mainly two meta-analyses, which I found to be very low quality. They also cite the 2006 Cochrane review, making it seem as if this review is positive, when in fact it in generally negative. Further there is an implied endorsement of anecdotal experience in that statement – a naive position for a researcher to take.

Conclusion

The prior plausibility for Echinacea as a cold remedy is very low, but not zero. As an herbal product it can feasibly have biological activity. The claim that it “boosts the immune system” is not credible, as such a phenomenon is not generally accepted. Basic science research essentially finds that the immune system is activated by Echinacea, but this has not been distinguished from a non-specific immune response to a foreign substance. In other words, stimulating the immune system non-specifically (I can do this by punching you in the arm) and “boosting” the immune system so that it functions more effectively against an infection, are not the same things. Leaping to the latter conclusion is not justified by the evidence.

From an historical perspective, there is no particular reason to conclude that Echinacea might be effective for the cold. Its traditional use by Native Americans was for many indications, possibly including cough, but not specifically the cold or flu. The great number of conditions for which it was used indicates that Echinacea was treated as a panacea, which further indicates there was no particular evidence for its effectiveness as a cold remedy. And further, such evidence would only amount to anecdotal evidence – the fact that belief in Echinacea became culturally embedded is not a reliable indicator of efficacy.

While herbs are drugs, and drugs can have biological effects, it is unlikely that any random drug would work for any particular clinical indication.

It is therefore no surprise that the clinical evidence shows that Echinacea is not effective for the prevention or treatment of colds or cold symptoms. This latest study adds to the evidence for lack of efficacy.

The market for Echinacea, however, endures, largely on a sea of anecdotes and marketing spin that is incongruous with the evidence.

Note: This post was written partly from material and analysis provided by Wallace Sampson.

Posted in: Herbs & Supplements

Leave a Comment (39) ↓

39 thoughts on “Echinacea for Cold and Flu

  1. windriven says:

    Cochrane: “Echinacea preparations tested in clinical trials differ greatly.”

    I cannot think of many single sentences that would better differentiate EBM from SBM. In effect the authors are admitting that they don’t even know what drug they are testing because they have not isolated the supposedly active component from the plant. Some echinacea preparations are made from one part of the plant, others from different parts. There is no effort to control the potency because there is no identified potent compound to test for.

    Running a few echinacea plants through your Robot Coupe does not science make. Claiming to analyze ‘studies’ based on this methodology is laughable.

  2. Dave McGinn says:

    Hey Steve,

    Excellent article, really clearly written and easily understood by the layman (i.e. me) !

    Quick question for you. The 2008 article that you refer to (http://www.ncbi.nlm.nih.gov/pubmed/18450126) has the following results:

    RESULTS: Fifty-eight individuals were included in the final data analysis: 28 in the echinacea group and 30 in the placebo group. Individuals in the echinacea group reported 9 sick days per person during the 8-week period, whereas the placebo group reported 14 sick days (z = -0.42; P = .67). Mild adverse effects were noted by 8% of the echinacea group and 7% of the placebo group (P = .24).

    The echinacea group reported 9 sick days, whereas the placebo group reported 14 in the 8 week period. Is this not statistically significant? Seems like a rather big difference to me!

    Don’t get me wrong, I don’t imagine that echinacea does any good, but I just want to better understand clinical studies such as this.

    Thanks, and keep up the good work!

  3. trrll says:

    I’m not sure why you state that “There already is a fairly low bar for statistical significance in clinical trials, biasing them toward false positives.” This study seems to be using the standard p < 0.05 "bar" used in most other scientific work, and it does not seem to have other problems, such as multiple comparisons, that would make that less stringent than it seems.

    Of course, p < 0.05 is fairly arbitrary, not a hard cut-off. Most people consider p < 0.1 to constitute a suggestive trend. Lack of known mechanism is not a strong objection; many pharmaceuticals were used for many years before a mechanism was established. This is very different from homeopathy, where a genuine effect would require major revisions to the laws of chemistry and thermodynamics. To me, the more meaningful finding is the very small effect size–even if the benefit happens to be real, it is marginal. Of course, if there were a large benefit, it would be easy to get a low p value.

    It is also worth noting that failing to detect statistically significant adverse effects does not mean that they don't exist, just that they aren't large or common.

    By way of disclosure, I once tried taking echinacea at the onset of a cold; the cold turned out to be unusually protracted and severe.

  4. trrll – I never mentioned lack of known mechanism as a problem. I agree this is a minor consideration at this stage (as opposed to no possible mechanism, like homeopathy).

    What I said was that there is no particular reason to think that echinacea might be useful for colds. It was a rather arbitrary claim, not based on traditional use as often claimed (and not that there is much to that either).

    And to clarify – I think the 0.05 P-value (which is arbitarty) is a low bar for significance. It is meant to give clinical studies sensitivity over specificity – to create, in essence, false positives rather than false negatives. There is nothing wrong with this, when put into perspective. But it just has to be recognized that using 0.05 as a cutoff is a low statistical bar.

    That is why citing trends that do not even meet this criterion is ridiculous. Now you are just in the noise.

  5. Dave – this was not statistically significant.

  6. windriven says:

    “This study seems to be using the standard p < 0.05 "bar" used in most other scientific work…"

    Or not.

    "A comparison of the 2 blinded groups showed a 28-point trend (95% CI, −69 to 13 points) toward benefit for echinacea (P = 0.089)."

    Further, consider this statement: "Mean illness duration in the blinded and unblinded echinacea groups was 6.34 and 6.76 days…"

    I can measure the transition from a '1' logic state to a '0' logic state down to nanoseconds. Recovery from the common cold isn't so clear. How does one judge precisely when a cold is over? Can one state overness accurately to the hour? Four hours? Half a day? Were potential confounding factors for assessing recovery accounted for? Recovery from a cold is a very noisy signal. Teasing meaningful data from it without very careful controls is unlikely.

  7. Good article, but sadly, I don’t think it covers my question regarding echinacea. Do we know if echinacea actually does something to activate the immune system or make it more sensitive?

    Sorry for not being more knowledgeable in this area, but is it possible that echinacea could exacerbate an auto-immune disorder, (Hashimoto’s*, SLE, SD, etc) or allergic or non-allergic asthma, rhinitis, sinusitis (are the later two even related to the immune system…I don’t know).

    There is a yummy lemon ginger drink that I love heated up when it’s chilly or I have a cold. Since it has echinacea, I have avoided it due to Hashimoto’s and asthma, etc. Maybe I’m being overly cautious. I don’t know.

    I used to drink echinacea tea as well, in the past. I suppose someone gave it to me as a gift when I had a cold and since I am a tea drinker (love most black, green and herbal teas) I found the herbal taste pleasing and I figure a cup of hot tea is good for most of what ails you. :)

    Now, I think of the public attitude toward boosting the immune system differently. I think of the immune system as a sort of hire assassin. Sure great for killing off your enemies, but you don’t want to get him too riled.

    Totally unscientific, I know.

    *Auto-immune thyroid disease.

  8. windriven says:

    Michele asked: “Do we know if echinacea actually does something to activate the immune system or make it more sensitive…”

    Cochrane suggests not: “Median change in interleukin-8 levels and neutrophil counts were also not statistically significant…”

  9. I really dislike it when a study leaps to unsupported conclusions and speculations, or tries to put spin on why negative results aren’t negative.

    Put what the data supports in your conclusion.

    Write a separate editorial or commentary if you want to spin it or speculate about the results, or at least separate that type of discussion in a post conclusion discussion section clearly delineated as such.

  10. Werdna says:

    @Dave – specifically – and I actually had to check to see if your cite wasn’t a typo – that P-value is crazy high. 0.67 implies that there is a 67% probability that this result was due to chance.

    The conclusion from the cite says pretty much the same: “Prophylactic treatment with commercially available E purpurea capsules did not significantly alter the frequency of upper respiratory tract symptoms compared with placebo use.”

  11. weing says:

    I stopped reading studies like this in the Annals back in my residency. They would always find a way to weasel out of saying that sCAM is useless. I suppose it’s on purpose to get more money for more useless studies, but it’s very frustrating to read and have them come up with gems like the following:

    “Higher-than-expected variability limited power to detect small benefits.”

  12. Wholly Father says:

    There is a randomized trial in which Echinacea treatment of colds was associated with an increased risk or recurrent otitis media in otitis prone children:

    http://www.biomedcentral.com/1472-6882/8/56

  13. @Wholly Father – Interesting! But I wonder about that study, it’s funny that they did not allow subjects on antibiotics or with ear tubes into the study, but children who got antibiotics or ear tube during the study were still included. I wonder if that would effect the results.

    Also another funny thing in the background “In children with recurrent otitis media, conventional approaches to decreasing the risk of further episodes include prophylactic antibiotic therapy and surgical insertion of tympanostomy tubes to prevent accumulation of middle ear effusion. The effectiveness of these approaches is limited and controversial [2,3].”

    This is the first I’ve heard that ear tubes are of limited effectiveness and controversial, even the sources they cite don’t seem to suggest that (in my reading.)

    But, I’m certainly not a fan of echinacea or osteopathic manipulative treatment for OME, so I guess I shouldn’t complain about the result of the study.

  14. windriven – I appreciate the effort that you took to track that down AND I did make an effort to look up “interleukin-8 levels and neutrophil counts” and how they relate to the immune response, but I’m afraid it’s over my head.

    Herbal remedies kind of make me crazy. If you know that some herbs can have quite strong effects, then one can suppose that others also have effects, which may not be documented or standardized. But, of course without the research, documentation and standardation, you don’t know whether that effect might be positive or negative (or either, depending upon the situation). So it might make sense to avoid them all together, to avoid potential negative effects.

    On the other hand many herbs are just, well food, one doesn’t want to get paranoid over, say a cup of peppermint tea or some ginger in your soup. And sometimes it seems that the only thing that separates herbal remedies* form food herbs is…what? mythology? a “wisdom of the ancients” reputation?

    Funny

    *This is, of course, separate from the non-food plants that have obvious effects, poison ivy, belladona, etc.

  15. tmac57 says:

    Timely article,Dr. Novella,because just yesterday, I was reading an article in Woman’s Day (I know) that was…er..interesting? titled ’10 Things You Didn’t Know About Colds’. Number 5 caught my eye:

    5. There’s a flower that may help fight cold viruses.
    You’ve probably heard of echinacea, a plant with a stunning pink flower, which is believed to help boost the immune system. University of Connecticut researchers put the theory to the test recently, and after studying more than 1,600 people, they reported that not only did echinacea cut the chances of catching a cold in half, but also those study participants who took it reduced the duration of their colds by about 1.4 days.

    Should you supplement with echinacea? It’s worth a try, says Dr. Eccles*. “As it is a natural product, it is not possible to standardize the medicine, so, like buying wine, get the best quality from [herbal supplement makers] who have been in the business for a while.”

    It’s no wonder that herbal remedies like this continue to sell,with confusing articles such as this out in the popular domain.

    *Ron Eccles, BSc, PhD, DSc, director of the Common Cold Centre at Cardiff University in the U.K.

  16. Jan Willem Nienhuys says:

    In Europe Echinacea is marketed energetically by the Swiss firm Bioforce AG. In the Netherlands their branch is called Biohorma. Bioforce/Biohorma was founded by a certain mr. Alfred Vogel (1902-1996). Vogel claimed that he obtained the seeds of Echinacea purpurea from a Sioux chief or medicine man, who taught him many things about Native American herbal lore and who revealed to him the secret of this plant. (Incidentally, the other plant is called E. angustifolia.) This must have happened somewhere in the ’50s. This wise Indian was called Black Elk. Until about 2002 nobody seemed to know who Black Elk was or had been, even though a snapshot of Vogel and Black Elk was printed in one his books (for a nice reproduction see this article. Then Gerda Sorensen tried to find out, and located a descendant of Black Elk.

    Her story is here. However, Ms. Sorensen doesn’t tell the reader that Ben Black Elk was also, from about 1949 on, the famous Indian tourist guide near the Mt. Rushmore National Monument. On summer days sometimes 5000 tourists had their picture taken with Ben Black Elk. The picture of Vogel and Ben Black Elk looks like a typical tourist pix.

    Bioforce/Biohorma sells Echinacea in the form of tablets and tincture named Echinaforce. Other companies sell it as homeopathic cure. In the Netherlands Echinaforce is the best sold phytotherapeutic/homeopathic preparation.

  17. Scott says:

    And sometimes it seems that the only thing that separates herbal remedies* form food herbs is…what? mythology? a “wisdom of the ancients” reputation?

    I’d say that food herbs are ones you WANT to eat. Drinking peppermint tea because it tastes good? Sensible. Drinking peppermint tea to, oh I don’t know, prevent gout? Not sensible.

    There are three reasons for this difference. First, it’s always a risk/benefit tradeoff. When there is an established benefit (tastes good) that changes the analysis drastically versus when there is not. Second, there’s the false advertising aspect of somebody selling you peppermint tea by claiming that it will prevent gout. Third, herbal remedy advocates often recommend doses far larger than one would use in food, which increases the risk of adverse effects.

    It is certainly true that there is some chance that (continuing with the example) peppermint tea has some adverse side effects, when drunk in typical quantities as an enjoyable beverage. But the other aspects of the situation make that possibility much less worrying than when using it to prevent gout.

  18. Joe says:

    I have long been suspicious of the value of ethnobotany as opposed to random screening. I recently encountered this: Daniel S. Fabricant and Norman R. Farnsworth Environ Health Perspect 109(suppl 1):69–75 (2001).
    http://ehpnet1.niehs.nih.gov/docs/2001/suppl-1/69-75fabricant/abstract.html which claims a good success rate for for ethnobotany; but a colleague of mine who saw it before me thinks the authors are over-generous in attributing traditional uses being related to modern use.

    She points out that Withering came upon digitalis (1785) as one component in a 20-herb mixture. Thus, the traditional drug was 95% inert; and the local healer could not tell the difference. It’s more like the herb mixture was a shot in the dark that accidentally hit the mark.

    There is also the case of the antimalarial drug artemisinin, it was discovered in a screening program of more than 100 antimalarial herbs (some say 200) in traditional use in China. http://www.ncbi.nlm.nih.gov/pubmed/16722826?ordinalpos=21&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_RVDocSum One in 100 (or 200) is not a very good record for traditional herbs being effective.

    One must also consider that truly effective herbs are (traditionally) only apparent as safe and effective when the results are quick to manifest themselves. Aristolochia was routinely given to women after childbirth (it has no effect), nobody knows how many women suffered kidney failure in the long run.

    I don’t think that “traditional uses” are a good guide, and it would be a shame if an actual drug in an herb were overlooked because it was not in the ethnobotanical profile. Taxol is one example of that. Perhaps David Kroll will provide a critical article on ethnobotany.

  19. Alyss says:

    I realize I spend a little too much of my reading time on vaccines, but under the heading “Echinacea for Cold and Flu,” I see an article about vaccine administration. Did my brain break?

    Oh noes! The mercury!

  20. Joe says:

    @micheleinmichigan on 22 Dec 2010 at 2:47 pm wrote “On the other hand many herbs are just, well food, one doesn’t want to get paranoid over, say a cup of peppermint tea or some ginger in your soup.

    This vexes me, as well. Has anyone tested the safety of our foods (to the same extent that was necessary to tease-out the relationship between smoking and lung cancer; which was difficult)? I mean, everybody that breathes air dies- the same can be said for everyone who eats ginger (and I do enjoy ginger). A complication is that some traditional beverage flavorings, e.g., comfrey and sassafras, have proven to be hepatotoxic and carcinogenic, respectively.

    Maybe someone can enlighten us.

  21. Jan Willem Nienhuys says:

    @tmac57

    The Woman’s Day article goes back to Shah et al., Evaluation of echinacea for the prevention and treatment of the common cold: a meta-analysis of 2007. The ‘cut in half’ refers to the 58% reduction in odds. But odds ratio is not the same as chance. If p is the chance, then p/(1-p) is the odds ratio.

    Julie J. Rehmeyer explained it nicely in:
    Hyping Echinacea with Bad Numbers

    The Shah et al. paper included the notorious Cohen paper, summarized here.

    A Dutch statistician summarized the Shah data:

    no cold, no echinacea: 210 subjects
    cold, no echinacea: 394 subjects
    no cold, echinacea: 414 subjects
    cold, echinacea: 337 subjects.

    The odds ratio becomes then (337 x 210) / (414 x 394) or:
    (337/414) : (394/210) = 0.43, or: odds of getting a cold in the Echinacea group,
    divided by the odds of getting a cold in the no-Echinacea group.

    In the Shah paper it’s done slightly differently, I think essentally by taking a weighted geometric mean of all individual odds ratios of 14 different studies.

    An odds ratio of 0.42 is indeed 58% less than 1. One might also say that in the Echinacea group the chance of getting a cold was 45%, and in the no-Echinacea group it was 65%. Ordinary people might say that 45 is only 31% less than 65.

    As the Cohen paper (with a large number of children in Israeli day care centers as subjects) weighed so heavily in the Shah statistics, I would be interested in any comments on the Cohen paper.

  22. Werdna says:

    @Joe – I just read that study. It’s not clear to me that it makes the argument the writer intends. For example…

    “Because these compounds are derived from only 94 species of plants, and a conservative estimate of the number of flowering plants occurring on the planet is 250,000, there should be an abundance of drugs remaining to be discovered in these plants”

    Why do we get to make the assumption that if a sample of effective pure plant derivatives (and it’s unclear if these plants are as effective or as useful as whatever it’s being contrasted with) which has been selected for efficacy is somehow now representative of the medicinal use of plants as a whole. Maybe that’s all there is. Maybe the rest of plants only replicate the features of existing drugs, etc…

    What would seem more accurate is if we RANDOMLY selected plants from some catalog and did the same comparison.

    Personally given what we know of the history of medical science – what we had to go through to get to a place where we can actually quantify or confidence in a medicine.

    Is there any reason to believe that anything other than the coarsest most base correlations are going to be evident from historic use of plants?

  23. windriven says:

    @Joe

    Interesting thought but who is going to pay for the research? Absent an epidemiological or other clue (i.e. Japanese consume a lot of ginger and, say, have a high incidence of cirrhosis) what exactly does one test for?

  24. Scott says:

    Not to mention the fact that there are a lot more foods than drugs.

  25. Mark P says:

    Adults who have found echinacea to be beneficial should not discontinue use based on the results of this trial, as there are no proven effective treatments and no side-effects were seen

    How did this pass any sort of review at all? Would our author also run with:

    “Adults who have found crossing their toes to be beneficial should not discontinue use based on the results of this trial, as there are no proven effective treatments and no side-effects were seen”.

    As others have pointed out, the “no side-effects were seen” is a weak line in itself. The unseen side-effects of herbs are much more likely to be bad than good.

    Imagine if the pharmaceutical companies sold goods on the basis that any effect not noticed must be good!

  26. windriven

    “@Joe – Interesting thought but who is going to pay for the research?”

    Particularly considering that “they”* keep showing us that all the good stuff is harmful. Sugar, Alcohol, cheese, butter, beef**, except for ‘confounding factors’***.

    *’a team of qualified scientists’

    **Not actually sure if beef is still bad for you.

    ***Like being French, which according to news outlets mean that your can’t be killed by anything.

  27. Scott – “It is certainly true that there is some chance that (continuing with the example) peppermint tea has some adverse side effects, when drunk in typical quantities as an enjoyable beverage. But the other aspects of the situation make that possibility much less worrying than when using it to prevent gout.”

    This seems like a sensible approach to take. Putting it in my context, I doubt a occasional cup of lemon, ginger, echinacea drink would likely cause any flare-up of my mild symptoms, While a three times a day mega-dose herbal supplement has a higher risk of being problematic.

  28. Let me reword that for them:

    Based on the results of this trial, this study finds no reason to recommend that adults who have found echinacea to be beneficial should continue its use any longer, as there is no demonstrated benefit to justify the practice or expense.

    This study finds that the benefits perceived by current users of echinacea are consistent with that of placebo, and as such, cannot recommend its continued use at this time.

  29. ConspicuousCarl says:

    Dave McGinn said
    Quick question for you. The 2008 article that you refer to (http://www.ncbi.nlm.nih.gov/pubmed/18450126) has the following results:

    “RESULTS: Fifty-eight individuals were included in the final data analysis: [...]“

    All of the p-value math is over my head, but I noticed when reading the summary that it also says this:

    “METHODS: In a randomized, double-blind clinical trial, 90 volunteers recruited from hospital personnel were randomly assigned [...]”

    Can someone with access to the full paper tell me if they gave a reason for only telling us the results for 58 out of 90 subjects?

  30. Harriet Hall says:

    I was interviewed on the Rob Breakenridge show on Canadian radio tonight about this Echinacea study and about alternative and natural medicines in general. I got to mention Dr. Novella’s article. The host had already read it and he said nice things about this blog.

    http://www.am770chqr.com/Shows/RobBreakenridge/Story.aspx?id=1332818

  31. Jan Willem Nienhuys says:

    @ ConspicuousCarl

    ‘Mild adverse effects were noted by 8% of the echinacea group and 7% of the placebo group (P = .24).’

    2 out 28 (= echinacea group) means 7.1% and 3 out of 28 means 10.7%, so how did they get 8%?

    Moreover, something like 2 or 3 in both groups of about equal size is just about as close as one can get to ‘no difference at all’. But the authors say that the chance of accidentally (when there is actually no real difference) getting such a large difference (either way) is about one quarter (24%). That’s incredible. Now if there had been 5 (=17.8%) with ‘mild adverse effects’ in the echinacea group and 2 (=6.7% ) in the other group, then one might obtain something near p=0.42.

    With these small numbers it is hard to get statistically significant results.

  32. Joe says:

    Mark P on 22 Dec 2010 at 4:49 pm quoted “Adults who have found echinacea to be beneficial should not discontinue use based on the results of this trial, as there are no proven effective treatments and no side-effects were seen

    Thanks, I missed that. In Varro Tyler’s “Herbs of Choice” (Haworth, 1999, p. 256) he cautions against use of echinacea by those who have severe systemic illness such as “tuberculosis, leukosis, collagen diseases, multiple sclerosis and the like” because of potential immune stimulation. He also warns that allergies may occur (elsewhere I had read that that could include anaphylaxis).

  33. micheleinmichigan asks, “Has anyone tested the safety of our foods (to the same extent that was necessary to tease-out the relationship between smoking and lung cancer; which was difficult)?”

    Basic foodstuffs are assumed to be safe even if that is not exactly true. An example they liked to give us back in dietetics class was that if prunes were a drug they would have an unacceptable risk/benefit ratio and not be approved. Another example is mold-fermented foods like blue cheese or soy sauce, because mold generally is carcinogenic. This is one of many reasons dieticians are continually harping on the need for a varied diet. If there’s maybe a small amount of something in pears or ham or stilton or cashews or wheat that is a little bit bad for us,* it’s not a big deal if we don’t eat that much of any one thing.

    Your contrast of tisane vs capsules is exactly right.

  34. TsuDhoNimh says:

    Echinacea affects phagocytosis and histamine release.

    The infection protection you would get from Echinacea is from increased phagocytosis, and because viruses are inside the body’s cells almost immediately, it’s not much good against colds. Won’t prevent, won’t cure, but it may slightly shorten the time it takes to clear out cellular debris after the infection, and therefore shorten the length of time you are coughing.

    Bauer VR Jurcic K Puhlmann J Wagner H
    Immunologische In-vivo- und In-vitro-Untersuchungen mit
    Echinacea- Extrakten.
    In: Arzneimittelforschung (1988 Feb) 38(2):276-81
    ISSN: 0004-4172

    Wildfeuer A Mayerhofer D
    Untersuchung des Einflusses von Phytopraparaten auf zellulare
    Funktionen der korpereigenen Abwehr.
    In: Arzneimittelforschung (1994 Mar) 44(3):361-6
    ISSN: 0004-4172

    If you can struggle through the German, they says it enhances
    phagocytosis. That has been known for a long time. It’s mentioned in my old pre-antibiotic medical texts.

    Tragni E Galli CL Tubaro A Del Negro P Della Loggia R
    Anti-inflammatory activity of Echinacea angustifolia fractions
    separated on the basis of molecular weight.
    In: Pharmacol Res Commun (1988 Dec) 20 Suppl 5:87-90
    ISSN: 0031-6989

    Basically says it has anti-inflammatory activity.

    ************
    Some other research – I’m still looking for the publication information – showed that it inhibits histamine release from mast cells. If you take small amounts continually you have fewer nasal allergy symptoms.

    Think of it as herbal Flonase.

    *************
    The classic pre-antibiotic herbal mix for pneumonias was something with berberine, which is a bacteriostatic, and Echinacea.

    Not exactly sulfa drugs, but better than cupping and purging.

  35. Joe says:

    @Alison Cummins on 23 Dec 2010 at 7:54 am wrote “Basic foodstuffs are assumed to be safe even if that is not exactly true.

    And there lies my problem. Case in point- an herbalist move-in next to my friend (Anna), and she offered herbal teas to Anna. I questioned whether she should try a mix that had not been tested for safety, and then realized that I cannot point to safety data for the brand she already drinks. Sassafras and comfrey were used for a long time before the dangers were noted (cancer and hepatotoxicity, respectively).

  36. Mark Crislip says:

    I cant read German, but if White cells are exposed to all sorts of antigens in the test tube, it will prime them to phagocytose better.

    It is a non specific reaction of no specific clinical relevance but often, erroneously, used to say some product or other boosts immune system.

    Candida makes wbc phagocytose listeria better, so treat listeria with yeast? I think not.

  37. Alison “Another example is mold-fermented foods like blue cheese or soy sauce, because mold generally is carcinogenic. This is one of many reasons dieticians are continually harping on the need for a varied diet. If there’s maybe a small amount of something in pears or ham or stilton or cashews or wheat that is a little bit bad for us,* it’s not a big deal if we don’t eat that much of any one thing. ”

    Thank you Alison, that is extremely disconcerting. :)

  38. Charon says:

    trrll: “This study seems to be using the standard p < 0.05 "bar" used in most other scientific work."

    This is not true. In the fields of physics, astronomy, and astrophysics, this p-value is considered completely unacceptable for claiming a result. The standard for significance is usually 3-sigma (e.g., p=0.0027), and for claiming a new particle detection in particle physics, 5-sigma (e.g., p=0.00000057).

    My understanding is that p=0.05 is used only in biology and medicine. Which isn't "most" other scientific work.

  39. TsuDhoNimh says:

    Mark – It was in vivo, with oral Echinacea. Feel better?

Comments are closed.