The trouble with herbals
Ben Goldacre
The Guardian
Saturday October 6 2007
The news this week that herbal remedies can be ineffective or dangerous is boring: but come with me on a journey through time (time… time…) to the origins of medicine.
Herbs contain bioactive compounds. That’s why they can be effective, but it’s also why they can have side effects (some will inactivate your contraceptive pill, others will cause renal failure). Huge numbers of bioactive compounds extracted or derived from plants are used today in medical practice, including even common stuff like aspirin. There is little difference between herbal medicine and medicine in terms of what is used, only in how.
Digoxin in foxglove is very good at treating atrial fibrillation, a common kind of irregular heart movement. Unfortunately the dose range is very narrow, so it’s really quite easy to kill your patient. From the moment of its discovery, pharmacologists worked hard to standardise the dose.
They started with standardised preparations of the plant, but this proved dangerously inconsistent, because the quantity of the active component in foxglove was so variable, so new strategies were developed: standardised preparations of each batch of plant were tested in animals first, to work out how potent they were, and so on. Eventually we worked out how to extract the digitalis, and it could be weighed out.
Now people do careful studies of large numbers of patients on digoxin to see which dose is most beneficial, to understand how it works, and careful monitoring of side effects takes place, from individual clinicians writing about their concerns in medical journals all the way up to the yellow card system, where doctors and even patients can send in their concerns to the MHRA, however trivial or unproven they may be.
This process of enquiry, and standardisation, and testing, and verification, and negative findings, and dismissal of ineffective methods, and more, is one of the great developments in medical and intellectual history. Many people were disappointed along the way, as their ideas and theories were proved to be unfounded. Digoxin pills and foxglove both contain the same ingredient, but it is how they are managed and handed over that differs.
Until a herbal practitioner can show that giving a whole plant instead of an extract really is better, they’re making stuff up when they make those claims. Similarly, until they can show that using herbs at such low doses that they have no measurable effect is somehow beneficial, beyond the placebo (and placebo is great by the way) then they’re shooting with the stars.
But there are structural problems in the way that herbalists work: they have failed – over millennia – to collectivise, so they do not work together on research, but rather as independent commercial traders. They do not move to university settings, where the culture of critical self-appraisal might infect them, and possibly extinguish them too, if their ideas don’t stand up.
And where alternative therapists do move into universities, they wall themselves off from the most valuable influences. CAM therapists don’t rub shoulders with colleagues from other disciplines, who could share ideas and insights with them, and move ideas forward, or help brush the bad ones aside. The alternative medicine university courses I have approached have simply, flatly, refused to tell me the most basic things, like what they teach and how.
It’s because of this culture, not funding, that the “research” on herbal remedies is inadequate. Huge numbers of “trials” are produced, at great expense, but the trials are inept, they are not “fair tests”, they have inadequate “blinding” and “randomisation”, positive results alone are “cherry-picked”, and worse.
An inept trial, bound by design to give a false positive result, costs just as much as a fair test of a treatment. And these problems are endemic: one study looked at the entire cannon of research on traditional Chinese medicine, and found 1100 papers: not one single trial published in China, in the entire history of research into traditional treatments, had ever found a test treatment to be ineffective. Not a single one.
Herbal pills contain bioactive compounds which can have real effects, but it is these differences of style, not content, that divide alternative therapies from medicine: and it will ever be the same.
References:
Do certain countries produce only positive results? A systematic review of controlled trials.
Vickers A, Goyal N, Harland R, Rees R.
Control Clin Trials. 1998 Apr;19(2):159-66.
A systematic review of randomised clinical trials of individualised herbal medicine in any indication
R Guo, P H Canter, and E Ernst
Postgrad. Med. J., Oct 2007; 83: 633 - 637.
doris said,
October 6, 2007 at 12:24 pm
Knowing little about the academic publication process in general,I’m wondering what the standard of peer review is in the countries which produced such positive results(Do certain countries….).
Are all academic studies reviewed internationally by experts,or do some rely on’homegrown’ expertise?
I have no idea.
What about widely differing cultural norms?
How about parallel trials,measuring the efficacy of a particular herbal compound:
one group here in the UK and one in say,China or India.
Researchers with equivalent credentials,tight control and design etc;run by relevant university departments.
Independent review of results.
Would widely differing results be found?
HarryR said,
October 6, 2007 at 12:27 pm
Random factoid: the use of digitalis to treat heart complaints was ‘discovered’ by Charles Darwin’s grandfather Erasmus, who had in fact plagiarised it from one of his friends. Not such a friend afterwards.
HarryR said,
October 6, 2007 at 12:29 pm
Oh, and this post about innumeracy struck me as your kind of thing:
http://itre.cis.upenn.edu/%7Emyl/languagelog/archives/004992.html
wilsontown said,
October 6, 2007 at 12:34 pm
Regarding peer review, it’s only as effective as the peers who do the reviewing. So the quality of review can vary greatly, even for articles submitted to the same journal.
Diotima said,
October 6, 2007 at 12:35 pm
HarryR: in fact Darwin’s grandfather’s friend borrowed the use of digitalis from folk medical workers. Digitalis was used in Ireland (the belief was that the ‘Lus Mor’ was a fairy plant) and elsewhere in Europe.
Ben is of course right as to the synthetic digoxin’s being a better bet than the old method (chew some leaves and spit them out sharpish). Digoxin still has almost no ‘cushion’ between the therapeutic dose and the fatla dose. I take 350 micrograms of Digoxin daily, so this is a subject close to my heart, as it were
Dean Morrison said,
October 6, 2007 at 5:04 pm
Up to at least the 1970’s the British Pharmacopoeia assay for Digitalin involved a LD 50 test on hamsters, complete with instructions on exactly how they should be pinned down on their backs on a board…
doris said,
October 6, 2007 at 8:05 pm
See drhadwentrust.org for an impressive array of humane research projects which are in progress at several universities(LD 50 above).
Not looking to start an argument,by the way.
Herbal medicine:herbalists say that the entire plant is essential to achieve the requisite therapeutic effect:has this been put to the test in a controlled trial?
BobP said,
October 6, 2007 at 9:21 pm
Hi Doris,
In answer to your early bird posting, I once took a detailed look at a published trial done in China (acupuncture, not herbs).
a) the clinical measure was obscure
b) the control group did not match the clinical group, and were possibly much healthier (selection was not randomised into the groups)
c) there was no blinding (the clinical group got treatment, the controls didn’t)
I concluded, very uncharitably, that the trial had been rigged; this opinion lost me a few friends in the acupuncture community.
——
And in reply to you point at #7, I think this week’s news item itself (linked in Ben’s first sentence) probably clobbers that; presumably the herbalists in the trial were using the whole thing, and it didn’t work.
PalMD said,
October 6, 2007 at 9:32 pm
One of the better treatments of the subject I have seen yet. And as a bonus, it applies to most altie therapies.
People may misunderstand the statement about “no negative studies”. No one, especially not people collectively, is right all of the time. Even with publication bias, you can easily find positive and negative studies about most medical treatments, and use the info and methods to see which studies are better etc. No negative studies means that nothing was really studies–everything is foregone conclusion.
Sameful, really.
Vaughan said,
October 6, 2007 at 10:06 pm
Probably worth noting that a recent Cochrane review found Chinese herbal medicine to be effective as an adjunct to antipsychotic medication in treating schizophrenia, although noted limitations in ’sample size and study length’ of the trials.
Robert Carnegie said,
October 7, 2007 at 12:20 am
Isn’t foxglove in [Silas Marner]? Or something.
ginkgo said,
October 7, 2007 at 12:45 am
I feel there are some points missing from this debate.
1) All drugs, herbs, etc should be assessed on whether they actually work or not. I personally don’t care what I take, as long as it works, and the side effects are minimal. Isn’t this the goal of pharmacology?
2) Big pharmacetical companies (from now on known as Big Pharma) also tweak and cherry pick and all those ‘un-ethical’ practices that introduce bias into supposedly scientific studies and apparently, a whole lot of other things. Like hiding the fact that Prozac increases the risk of suicide in young people, apparently.
3) Big Pharm has no interest in developing herbal remedies - you can’t patent a herb - and so, do not research them as potential drugs. Who funds most research into drug development?
4) There is a different attidude (I know I can’t spell) in other countries. St John’s Wort is only avaiable on prescription in Germany for instance. As fars as I know, Germany treats alot of herbal products like prescription drugs.
5) Herbal remedies are often about prevention, and management of symptoms. I don’t think anyone expects them to be substitues for say, insulin, or anything like that. But if you can take ginkgo for Reynaulds disease and it helps you, then why not…. Or if you take St John’s Wort for depression (which I have personally done) and it helps you much more than any of the SSRIs (and I’ve tried alot of them) then why not? Yes St John’s wort interacts with the Pill, but then, interactions occur with countless conventional prescription medicines also.
6) What do you think to Dr Weill?
7) Alot of herbal remedies have their useage traced back thousands of years…. we evolved with the plants, we learnt how to use them for our benefit… I think its called ‘ethnobotany’.
9) Conventional drugs sometimes don’t work either as they’re supposed to. I’ve tried countless anti-depressants and none of them made me anti-depressed.
Like I said before, I don’t care what I take, as long as the side effects are minimal and it actually works. If it so happens to be a herb, then so be it….I think we really have to stop seeing herbs as ‘quackery’ and for what they are - chemical compounds, the same as conventional medicines. And if we see that conventional medicines aren’t all they’re cracked up to be either…..and if Big Pharma didn’t have such an influence in the development of drugs, then the world would be a much better place, IMHO.
ginkgo said,
October 7, 2007 at 12:57 am
“Until a herbal practitioner can show that giving a whole plant instead of an extract really is better, they’re making stuff up when they make those claims.”
Well thats just not true. I think there have been several incidences whereby an extract was actually given to a patient, only for it to prove fatal, or ineffectual. Okay, so I’m tired and I can’t be bothered to conjure up the revelent articles in pubmed. But, kava kava springs to mind (again) and St John’s wort (again). So with Pacific Islanders, who have been chewing kava kava for, presumably, centries, have had no ill effect, but when western man tries to ‘extract’ the active compound it all falls, because other compounds in the plant are working as well to prevent the bad stuff from happening.
I think there have also been studies showing that with St John’s Wort, the whole plant is more effective than just the supposed ‘extracts’.
I thought that the whole point with using the whole plant was that the plant had evolved mechancisms, compounds etc, to kinda help itself. I’m tired and I also sounds like a quack….
PalMD said,
October 7, 2007 at 4:02 am
You have expressed many misconceptions, but this not being my blog, i will only address a few:
“Big Pharm has no interest in developing herbal remedies - you can’t patent a herb - and so, do not research them as potential drugs. Who funds most research into drug development?”
Actually, they can patent herbs–if they come up with a unique formulation or delivery system. Niacin, a B vitamin, exists in patented form, and is quite profitable.
Next, St. John’s Wort works by the same chemical mechanisms as SSRIs.
Next, “the death rate from herbals is small”. I have no idea whether that is true or not (it prob is true) but the “life” rate is also quite small. For instance, the NNT (number needed to treat) to prevent a death from heart attack with a particular medication is about 11, meaning that 11 people would have to be treated with the drug to prevent one death from heart attack. Do we know of any herbs that are that effective, or that can prove they are even close?
kiplinger said,
October 7, 2007 at 6:54 am
Ben, this is startling:
“And these problems are endemic: one study looked at the entire cannon of research on traditional Chinese medicine, and found 1100 papers: not one single trial published in China, in the entire history of research into traditional treatments, had ever found a test treatment to be ineffective. Not a single one.”
Can you confirm that is what the study said? I’m asking because I haven’t got access to the whole paper; the way I read the abstract it reports that 99% of the abstracts that met the inclusion criteria were positive — and there were 405 of those (out of 1100).
doris said,
October 7, 2007 at 7:45 am
Thank you,Bob P,for your answer.
Perhaps my question about trials of whole plant remedies seems dumb,but I read Ben’s article to mean that a review of individualised herbal remedies had found them to be at best,wanting,and at worst,hazardous.
What I wondered, was whether a proper parallel study had been set up to compare drug ‘A’ with herbal remedy’B’.
Ginkgo’s comments raise a number of interesting points.
To follow on from Ginkgo’s observations,I believe that herbal remedies are available,onprescription,in France and Switzerland as well.
Diotima said,
October 7, 2007 at 8:19 am
Ginko. As for the death rate being small; would you self-experiment with foxglove to demonstrate this? I can assure you that the sensation of hitting the maximum non-fatal dosage of digitalis (I experienced this in a clinical situation, being given intravenous digoxin) is something to remember all your life—if you survive it, that is.
Dave H said,
October 7, 2007 at 8:47 am
canon
BobP said,
October 7, 2007 at 9:15 am
I’ve heard the suggestion that the whole plant is a better remedy than the extract a couple of times in the past.
I’m trying to think of this in terms of plant evolution.
Generally, I think the reason why a plant contains a compound like digoxin (or nicotine, or ricin) is not because the compound is essential to the biochemistry of the plant, but because it is a biochemical defence mechanism which helps protect the plant from being eaten by predators or parasites. In which case, there’s no advantage to softening the blow by evolving mechanisms which mitigate its impact; provided that the plant’s own biochemistry isn’t affected by the compound, it can be as lethal as you like to you & me.
kiplinger said,
October 7, 2007 at 9:29 am
Is a whole orange “better” for you than vitamin C?
ginkgo said,
October 7, 2007 at 10:53 am
PalMD:
“Actually, they can patent herbs–if they come up with a unique formulation or delivery system. Niacin, a B vitamin, exists in patented form, and is quite profitable.”
OK how many patented herbs are there on the market? (I taking herbs, not vitamins.)
http://www.lichtwer.co.uk/
I think lichtwer patent some of their stuff - I’m abit hazy on the details. But you are not going to get any of the Big Pharma patenting herbs anytime soon. Why? Well, perhaps one reason is that the whole plant is often better than the extract - and you can’t patent a herb!
“Next, St. John’s Wort works by the same chemical mechanisms as SSRIs.”
No it doesn’t. Thats just another misconception. They just don’t know how it works. Please show me some reputable scientific articles that have evidence of this. If you’ve tried SSRIs (I’ve tried two common popular ones) and St John’s Wort, you will know that they are definately not the same.
I can think of a study that showed that Tea tree oil works just as well, and with less side effects, then benzoyl peroxide.
assett I, Pannowitz D, Barnetson R (1990). “A comparative study of tea-tree oil versus benzoylperoxide in the treatment of acne.”. Med J Aust 153 (8): 455-8. PMID 2145499.
But I clearly said that I don’t think anyone expects herbs to be miracle cures for, example, if you’re having a heart attack or to replace insulin.
I’m just saying that if it works, and its a herb, so what? Herbs are not quackery - they have been used for thousands and thousands of years. I bet you use herbs everyday and don’t realise it. Tea and coffee and cocoa, could both be classifed as herbs. Why don’t we invent an artifcial subsitute for coffee? ?
doris said,
October 7, 2007 at 11:02 am
Reflecting on BobP’s remarks:is there any evidence that tribes which have made extensive use of potent plants for many hundreds of years(the south american Indian tribes which have managed to resist the allure of western capitalism spring to mind),are less susceptible to plant toxicity than western europeans?
Has their body chemistry evolved/adapted to provide a more robust immunity to potentially toxic alkaloids,for example?
Does anyone know?
Secondly,I believe it is customary amongst herbalists and various agrarian societies to harvest herbs at certain times of the day,because of diurnal variations in concentrations of active compounds within the plants.
ginkgo said,
October 7, 2007 at 11:08 am
Diotima said,
Ginko. As for the death rate being small; would you self-experiment with foxglove to demonstrate this? I can assure you that the sensation of hitting the maximum non-fatal dosage of digitalis (I experienced this in a clinical situation, being given intravenous digoxin) is something to remember all your life—if you survive it, that is.
Look, you’re not getting my point.
You could say that about any number of conventional drugs. The theraputic window for taking Roaccutane is also quite small - as I expect it is with alot of other conventional drugs and yes, also with herbs. So what? Just because the theraputic window is small, doesn’t mean anything. Anything is poisonous if the doseage is wrong, even water. So. Of course there are ‘bad’ herbs that probably aren’t worth the risk of using them, the flowglove example being one of them. But then, there are also plenty of ‘bad’ drugs that are not currently used anymore either.
What I am trying to say, is that, the herbs that are available over the counter, tend to be safe. They might have some side effects or interactions, but then, so do alot of other OTC conventional drugs too. How many people have died as a result of interactions between OTC drugs? Probably quite a few (which I realise is wholly unscientific). Alot of people die of unintentional interactions between presciption drugs and OTC drugs. Sean Penn’s brother was one of them. They got the doseage wrong of that poor girl in Glasgow who had chemo. She died. Are people stopping using chemo because of that? No they’re not! Conventional drugs are just as dangerous as herbs.
ginkgo said,
October 7, 2007 at 11:19 am
I am not trying to say that herbs should replace conventional drugs. I am not saying that conventional drugs are the answer either.
But some of the most successful medical compounds are derived from the plant kingdom - aspirin and the opiates spring to mind, along with penicillin (which I suppose strictly derived from a bacterium or is it a fugus? Anyway, the natural world).
What I am saying, is that we shouldn’t disregard herbs simply because they’re not marketed by Big Pharma or that doctors don’t get any training in using them or whatever. Herbs could contain many potentially interesting compounds that we could use for our benefit. I’m not saying you should go see a herbalist when you have cancer, or that you should have chinese traditional medicine to manage your diabetes.
If we really try to view herbs as chemical compounds much like we do conventional drugs then the world would be a better place.
Andrew Clegg said,
October 7, 2007 at 11:37 am
“If we really try to view herbs as chemical compounds much like we do conventional drugs then the world would be a better place.”
Gingko, I suspect that’s already how most sensible people view herbs.
Go back and re-read Ben’s article: “There is little difference between herbal medicine and medicine in terms of what is used, only in how.”
However, there’s at least two prerequisites for using chemical compounds as drugs which herbal medicines don’t (generally) have: dose standardization for the purposes of safety and efficacy, and well-executed trials to supply robust evidence of that same safety and efficacy.
If herbal remedies are to get on a ‘level playing field’ with chemical compounds, as you so clearly wish, they need both those things.
Andrew.
MickB said,
October 7, 2007 at 12:27 pm
Gingko, I understand what you’re saying but I can’t help feeling that you’re missing the point of the article. It’s true that conventional drugs are sometimes dangerous, but because they are also beneficial we still use them while doing our best to undestand the dangers and to provide accurate information to users. Herbal remedies certainly contain active ingredients so they possess the same potential for both benefit and harm, but the problem here appears to be that they are not subject to proper clinical trials to verify and quantify their effects.
As with so many alternative remedies, the practitioners make claims that are not backed by evidence, and they refuse to run the proper blinded clinical trials that would give answers one way or the other. In fact it’s worse than that, they pretend the trials have been done properly and claim that everything’s perfect. It’s possible that using the whole plant could confer some benefit over using the isolated active ingredient alone, and that in some cases that might outweigh the dangers of taking relatively uncontrolled doses of that ingredient, but unless proper tials are run we have no way of knowing whether it’s true. It’s a shame because we all lose out by not having accurate information about these medicines. The companies that sell these remedies make enough profit to be able to run proper trials so one has to ask why they’re not being done.
Diotima said,
October 7, 2007 at 12:30 pm
Ginko: I don’t think that you grasp my point (or Ben’s) which is that synthetic medications are very carefully developed so that the exact therapeutic dosage can be delivered. This is not so with herbal products. If I reverted to the C18 method of taking digitalis there would be a high level of risk. Typically those who used it then had to guess how many drops of foxglove tincture to take. And they were probably unaware of the fact that the clear indication of digitalis poisoning is the visual field turning yellow.
Anything in excess can kill, sure. But folk medication which is difficult to measure will kill you quicker than Digoxin. And just because there are tragic errors in conventional medicine this does not give a licence for uncritical support for folk medicine. You might remember the brave young woman who told the Guardian her story of using traditional Chinese medicine for an early stage cervical cancer, for which simple laser treatment was indicated. This flirtation with alternative medicine ended with an emergency hysterectomy at the age of 30.
Ben Goldacre said,
October 7, 2007 at 12:58 pm
okay, so rather than simply repeat marketing platitudes about “natural is good” and “pharma is evil” for the weakminded, if any herbalists want to engage with what i said:
1. extracting the active ingredients from plants is hugely beneficial. or show me otherwise?
2. give me some evidence that giving a whole plant is frequently a better idea (not “natural is good”, alkaloids often really arent good, plants are very frequently filled with things that aren’t good: give me evidence).
3. herbalists come out with tosh because they do not work and communicate in environments where their ideas are subjected to critical appraisal. because of this, when their ideas are challenged, for example here, instead of engaging with those criticisms, they dodge them. if evidence doesn’t exist “yet”, then don’t pretend it does. if the trials don’t exist, don’t do inept false positive ones, come to me and i’ll help you design a fair test, there are tens of thousands of people in this country alone who would happily advise you. you don’t need a lot of money to do a trial. if you claim you do, that only demonstrates your ignorance, and i’ll say it again: herbalists do trials, they have done heaps of them, at expense, but rubbish trials, that gave false positive results. it’s not anyone else’s fault but your own if people ignore those. you wasted your money.
4. a useful crystallisation of the issue might for you be this: why are critical reviews of herbs, summarising the negative findings, performed and publicised by someone like ernst, and not by herbalists? every single one of the top 3 articles in the BMJ last year was hugely, corruscatingly critical of a key feature in the practice of medicine. that is why medicine moves ahead, and herbalism has degenerated into tosh.
projektleiterin said,
October 7, 2007 at 1:05 pm
Sorry, that’s not correct. You can get it in drugstores, the concentration is probably much lower though. And I don’t believe the second claim that in Germany herbal products get treated like prescription drugs. What kind of herbal products are you referring to?
By the way, if you you know how to write “ginkgo” your spelling can’t be that bad.
Andrew Clegg said,
October 7, 2007 at 2:10 pm
I can’t help wondering, when discussions like these come up…
If ‘traditional medicine’ really is better than modern pharmacology, why are life expectencies so much higher now, and incidences of nasty things like epidemics so much lower, than they were in ‘traditional’ times? (Whenever they’re supposed to be, anyway…)
In fact it seems to me (just a theory) that those parts of the world with the least penetration of modern medicine, and thus the most reliance on ‘traditional’ therapies, are often the sickest.
Andrew.
PalMD said,
October 7, 2007 at 4:44 pm
doris asked,”whether a proper parallel study had been set up to compare drug ‘A’ with herbal remedy’B’.”
This (www.ncbi.nlm.nih.gov/sites/entrez?cmd=Retrieve&db=PubMed&list_uids=10755847&dopt=Citation)
is, unfortunately a meta analysis, but tries to look at the question.
This was a bit more vigorous (jama.ama-assn.org/cgi/content/abstract/278/16/1327).
Here’s more (archneur.ama-assn.org/cgi/content/full/55/11/1409?ijkey=Yi8.4P6swGajA)
So you see, studies can be done, but herbalists choose not to, but rather rely on emotional appeal.
ginkgo said,
October 7, 2007 at 6:25 pm
Okay so people have problems with herbalists per se, not academic researchers researching herbs? Well, I don’t know about that. From my personal experience (I’ve seen a herbalist for depression) what she prescribed for me was pretty tame. Nothing outrageous. And it did work I have to say. Better than any of the conventional medicines that I’ve tried anyway.
The herbalist belonged to National Institute of Medical Herbalists:
http://www.nimh.org.uk/
There wasn’t anything quack like about her approach. And, I would like to point out, a research scientist (geology as it happens) so I’m not some tree hugging hippy either. I will take whatever works. My experiences with conventional medicine to manage my symptoms of depression where awful. A GP who didn’t even realise to connect my 12 hour sleeping patterns to the drug cipramil which I was taking at the time. I still can’t believe that. I was the one who found out that cipramil caused excessive sleeping - not him! ‘Oh gee, Dr, I’m sleeping like, 12 hours, could it be a side effect of the (only) drug I’m taking (and I’m an otherwise healthy, fit, young woman). Anyway I’m digressing.
Has anyone heard of the Commission E monographs?
Here is a sample:
http://content.herbalgram.org/abc/herbalmedicine/default.asp?m=40
http://www.amazon.com/Complete-German-Commission-Monographs-Therapeutic/dp/096555550X
http://en.wikipedia.org/wiki/Commission_E
Actually the wiki article enforced my hunch that herbal system is completely different in Germany.
As for standarised extracts, well to the best of my knowledge the St John’s Wort I took was, in fact, standarised….
“1 tablet contains 300mg standardised dry extract of St. John’s wort (Hypericum perforatum), providing up to 900µg of hypericin.”
Its from Lichtwer, a herbal company that also does alot of research into herbal remedies.
http://www.lichtwer.co.uk/stJohnsWort/
In fact, most of their products are standarised.
http://www.lichtwer.co.uk/
PalMD said,
October 7, 2007 at 6:44 pm
Just to remind you, St Johns Wort really is thought to work in the same way as conventional SSRIs. See, for example:
http://jpet.aspetjournals.org/cgi/content/abstract/290/3/1363
Although the exact mechanisms and the efficacy remain unclear.
http://www.theannals.com/cgi/content/abstract/32/11/1201
However, it has been found to have significant effects on drug metabolism in the liver. Unfortunately, most of the time this is hard to quantify, given the wide range of variations.
If you feel standard extracts are widely avail outside Germany, then they should be tested against standard therapy (SSRIs). It is unclear to me what the advantages or disadvantages of using a purified herbal extract with bioactive properties (i.e. a medicine) are vs other medicine.
ginkgo said,
October 7, 2007 at 6:44 pm
I was under the impression that herbalists only use the herbs that are ‘deemed’ safe. Like I said before, the herbalist gave me something completely tame, and not in any way ‘dangerous’. And it did work.
But perhaps asking herbalists to conduct trials is like asking GPs to conduct trials too? They simply don’t, do they? Its the pharamcologists who do the trials. Perhaps its the same with herbals - its the phytomedicine people who do this. I don’t know! I’m not a herbalist - just someone how has had thoroughly awful management of her symtoms on the NHS (I’ve seen countless GPs and one psychiatrist) but lovely care from a herblist. And most, importantly, took St John’s Wort, which did work better than any of the conventional drugs that I have tried.
Perhaps there are more than one type of herbalist. There’s the medicinal herbalists (which I saw) and the Chinese Traditional Medicine people, and then….umm, quacks?
Like I said before, I don’t think people see a herbalist when they’re having a heart attack. Its more management of chronic exisiting symptoms….
PalMD said,
October 7, 2007 at 6:55 pm
Im not suggesting that herbalists conduct studies, only that if companies want to sell things and make claims about their efficacy it should be tested, just like any other medicine.
I certainly wouldn’t argue with your personal experience–it is subjective, and irrefutable, as it is yours.
Over here in the U.S., companies can sell any substance with any claim, as long as they put a little “Quack Miranda Warning” at the bottom of the page.
Ben Goldacre said,
October 7, 2007 at 7:00 pm
i really don’t want to get into enforcing rules, but let me say again, i simply dont want people discussing their personal medical problems here. it only leads to upset and it never informs a discussion about the evidence.
i wouldn’t say that pharmacologists are particularly the people who do trials, they might do, but it’s not necessarily their bread and butter, you’ll find many of them clamping axons, ucking about with rabbits, squirting stuff on dishes, and the like.
And GPs routinely do do trials, thousands of GPs and professors of primary care would be startled to hear you say otherwise.
in fact as i’ve said before, all the boohoo quacks who moan about not being able to get the money to do research should bear in mind that:
(a) for the millioonth time, they do do research, they just squander their resources on bad research that gives false positives, which is a <i>grotesque</i> betrayal of the people who consent to be in their studies, just as it is when big pharma companies run bogus rigged trials.
and
(b) a huge amount of research is in fact performed by junior doctors trying desperately to advance their careers, on top of 60 hour a week medical jobs. it might not be the best quality stuff, but they do it, on top of busy jobs, and they don’t get paid for it. CAM therapists seriosuly need to grow out of their fantasies about how everyone should pay them to do flawed research. the world does not owe you a living or a research grant. get on your bike and do research.
Robert Carnegie said,
October 7, 2007 at 11:34 pm
Oh, btw, where does cannabis fit in the herbal remedy spectrum? It’s natural, right?
kiplinger said,
October 8, 2007 at 12:41 am
“And these problems are endemic: one study looked at the entire cannon of research on traditional Chinese medicine, and found 1100 papers: not one single trial published in China, in the entire history of research into traditional treatments, had ever found a test treatment to be ineffective. Not a single one.”
Ben , can you confirm that is what the study said? I’m asking because I haven’t got access to the whole paper; the way I read the abstract it reports that 99% of the abstracts that met the inclusion criteria were positive — and there were 405 of those (out of 1100).
Diotima said,
October 8, 2007 at 7:10 am
Ginkgo: your argument is of the ‘it worked for me’ school, which I frequently encounter from friends who are enthusiasts for homeopathy. By the way St John’s Wort reacts dangerously with anticoagulants, so the ‘natural’ is not always free of side-effects.
Ben: apologies for mentioning my heart condition; I only did so to give a good example of the superiority of the synthetic over the ‘natural’. My garden is full of magnificent foxgloves and I’m leaving them well alone.
outeast said,
October 8, 2007 at 9:11 am
Ginko,
As a scientist, you should be able to look at a sentence like ‘[I had] thoroughly awful management of [my] symtoms on the NHS … but lovely care from a herblist’ and see at least the possibility that the pharmacological efects of StJW might have been reinforced by the placebo effect? That statement looks to me like a texbook description of exactly the circumstances in which a placebo impact would be highly likely. (And, of course, this impact would be reinforcing a pharmacologically active compound, not a placebo.)
outeast said,
October 8, 2007 at 9:12 am
Ben
Only just noticed the rubber duck in the address bar. Nice touch!!
muscleman said,
October 8, 2007 at 9:18 am
In answer to the first question about peer review, it depends on the journal the research gets published in. If you for eg publish in the Journal of Pharmacology it is likely the article will get reviewed by people in a number of countries. However the article will have to be in English.
If you are Chinese in China trying to get a paper on Acupuncture or herbals published in Chinese then your options are limited to Chinese journals and their international options will be limited to Chinese working overseas.
As for examples of when extracts are worse than using the whole plant I am surprised nobody has mentioned beta-carotene and THAT cancer trial. Yet there is so much evidence that eating beta-carotene rich plants (carrots anyone?) is good for you*. After all that was why they thought testing it in extract pill form was a good idea.
*though not apparently for prostate cancer
though at least it doesn’t make it worse…
superburger said,
October 8, 2007 at 11:31 am
“If we really try to view herbs as chemical compounds much like we do conventional drugs then the world would be a better place.”
Er, ginkgo, science does and scientists do. (in fact, show me a physical objectthat isn’t a chemical compound….)
There’s immense, incredible, research being done all the time investigating the medicinal properties of natural produces.
You can make a good argument that most of modern chemistry is devoted to trying to extract, characterise, synthesise, study and perhaps improve products from plants.
Chemists, biochemists, pharmacists etc are all very much alive to the notion that herbs are complex mixtures of compounds with massive scope for curing the spectrum of human illness.
it’s not easy to see how it *can* be possible that the mixture of compounds in plants could have a better effect than a single biomolecule. It just seems that very few good trials have shown this to be true, compared to the vast evidence base for the use of single plant derived compounds.
muscleman said,
October 8, 2007 at 12:56 pm
Superburger: “it’s not easy to see how it *can* be possible that the mixture of compounds in plants could have a better effect than a single biomolecule. It just seems that very few good trials have shown this to be true, compared to the vast evidence base for the use of single plant derived compounds.”
The thing here is we must be careful not to make a logical error. Just because we have had a history of success at extracting and using single bioactive compounds from natural sources does not mean all such will be effective. For one thing, unsuccessful compounds don’t get noticed. For every synthesised aspirin trumping willowbark tea numerous other things fail in trials or at the first hurdle. There are high throughput screening facilities working 24/7 the world over at this testing millions of compounds, they won’t all work.
Also as the beta-carotene/cancer thing showed our success with things like aspirin or artemisin has given us a false expectation. I bet the hunt is now on for what other compounds beta-carotene needs, the question is just such a no-brainer. It might simply be to do with dietary fibre slowing gut release and the extract pills were absorbed too quickly. Or it may be that other compounds stop it being broken down too fast or ensure it gets processed into a bioactive form in the liver. There will be a lot of people asking what compounds are common between carrots and pumpkin or (insert orange vegetable of choice here).
Wonko said,
October 8, 2007 at 4:20 pm
“If ‘traditional medicine’ really is better than modern pharmacology, why are life expectencies so much higher now, and incidences of nasty things like epidemics so much lower, than they were in ‘traditional’ times?”
Probably nothing to do with modern pharmaceuticals at all, but rather social deprevation, poverty, poor housing, lack of clean water, poor air quality, poor diet. Similarly, life expectancy in the UK owes much more to getting doctors to wash their hands between patients than to modern drugs.
That said, at least if you contract HIV in the UK you get a choice between an antiviral or sweet potatoes together with good quality research evidence on which to base your choice. In S Africa, you just get the herbal approach. May not do much to overall life expectancy, but it will do a great deal to yours!
doris said,
October 8, 2007 at 5:39 pm
Just started reading an amazing book:The Black Swan,by Nassim Nicholas Taleb.
Something for everyone;how humans get wrongfooted by the unpredictable:real mind stretching stuff,some of which relates to what has been discussed here(and in other blogs).
Thank you,PalMD for links on analyses of trials of ginkgo remedies in AD patients.
Does anyone have any comments on my question about differing susceptibilities to plant compounds in different population groups?
I haven’t found anything.
Diotima said,
October 9, 2007 at 1:13 pm
Outeast: Yes indeed– ‘lovely care from a herbalist’ is a strong indication of the power of the therapeutic encounter. A friend who consulted a medically qualified homeopath at the Royal London Homeopathic Hospital was given an hour’s kindly discussion of her problems and walked out feeling much better before she had popped a single pill. As a result she is a passionate endorser of homeopathy and has quarrelled with a family member who has refused to go down the homeopathic path.
projektleiterin said,
October 12, 2007 at 10:27 am
What cancer trial? What was the result of it?
Gramlin said,
October 13, 2007 at 12:08 pm
Right, I was in a conversation with an old friend last night, who has obtained a medical herbalism degree (I can’t remember where, it was somewhere in London, but it was fairly heavy on the accademic side – and appeared scientific, from the outside at least).
He is of the opinion that much of the research done that goes against herbal remedies – which leads to treatments being taken off the market – is of the order of taking a mega-dose of some extract, bombarding a rat’s liver (or some such) and then extrapolating from this that it *might* be bad for humans.
This sounded rather like the same criticism now being levelled at Patrick Holford, and the like, about some of the “evidence” he uses, only in reverse.
Can anyone comment on this?
Robert Carnegie said,
October 13, 2007 at 3:36 pm
I suppose you start with a list of “herbal remedies” taken off the market. But do we mean banned, or just dropped by the manufacturer because people don’t think they’re any good, possibly because of negative reporting? If we stick to banned then we can get somewhere. If we include marketing decisions then it’s very difficult to prove. There must be hundreds of people selling St. John’s Wort and some of them must have gone out of business, and while that may not be due to scientific evidence, it probably didn’t help them.
Gramlin said,
October 13, 2007 at 7:15 pm
No, we mean banned.
I can’t comment on the veracity of this person’s claims, despite being an old friend I felt the opinion being touted – that evidence used to reduce the remedies available to herbalists is usually spurious – was in part a perpetuation of the myth that big pharma is out to get herbalists. I should try to find out which examples, specifically, my friend meant.
However, we discussed the case of Kava Kava, which was indeed banned for sale in Britain after it was found that extracts in high strength had lead to death in certain people. There is a twist though: it is claimed that ALL of the 18 deaths (or thereabouts) that occurred in other parts of Europe, not the UK, were among people who either a) Had combined it with Alcohol, b) Were on medications known to harm the liver, or c) Already had diseased livers.
My friend contends that Kava in its raw state contains the amino acid Glutathione, which is the principle detoxification chemical in the liver responsible for neutralisation of alcohol and even paracetamol overload, I’m told.
Standardised extracts would likely not contain this amino acid, but simply remove the active kava-lactones and concentrate them in capsule form. Thus, the herbalist I happen to know claims, the evidence used to implement the ban is spurious.
As to the broader issue of whether most evidence used to ban herbs is also invalid, I will need to ask for more specific examples, I think, but I just thought I’d post it here to see what anyone else might have heard.
Gramlin said,
October 13, 2007 at 7:21 pm
Oh yeah, one more thing, Kava *works* – it’s drunk for fun in the pacific islands where it is traditionally grown, and is not said to cure anything, though medical herbalists used to prescribe it for relaxation, anxiety, sleep etc.
Mongoose said,
October 14, 2007 at 11:15 am
Ben Goldacre writes that herbalists need to provide evidence that giving a whole plant instead of an extract or active ingredient is a better idea. Bear with me…
We know that single bioactive compounds can be used medicinally because of the experimental methods developed by medical science. These methods have been specifically developed to test the effects of single bioactive compounds on our bodies. However, it is one thing to say ‘we have an experimental method that enables us to develop medicines from single bioactive compounds’, but it is another matter entirely to say, as medical scientists do, ‘therefore, all medicines must be single bioactive compounds’. This is simply false logic. Unfortunately, this is the mind-set of the medical orthodoxy. They have reached a point half way up the mountain, planted a flag and called it the summit. Instead of viewing their method as useful but limited, they have set up camp and attempted to impose its limitations on the world. This is the approach of the fundamentalist and the anti-scientist…
If we followed Ben Goldacre’s ‘digitalis=digoxin’ argument to its logical conclusion we would be asking greengrocers why we should consume whole fruits and vegetables as opposed to pills containing the active ingredients. Fruit and vegetables also contain a number of toxic compounds that can be isolated, concentrated and used to kill rodents (for example, isothiocyanates in the brassica family; tropane
alkaloids in potatoes, tomatoes, and peppers; opiates in lettuce; prussic
acid in stone fruits etc.). Perhaps we should close down the greengrocers… Just as the human race has consumed certain fruits and vegetables for thousands of years, so it has used whole herbs medicinally. The knowledge of what works and what doesn’t has been passed down the generations - this is the knowledge of experience – and to many people this is sufficient evidence to use herbs medicinally.
The ‘digitalis=digoxin’ example implies that herbs contain a single active ingredient that can be isolated, concentrated and administered in a pill. This is simply not true. In many cases the so-called ‘active ingredient’ of a herb is unknown (for example, Vitex agnus-castus) or there are a number of contenders (for example, hypericin and hyperforin in St John’s Wort). Herbalists use the concept of ‘synergy’ to explain the action of whole herbs medicinally. The idea is that the various compounds that make up a herb work together to produce various medicinal effects. Therefore, isolating ‘active ingredients’ not only reduces a herb’s efficacy, it is also potentially harmful. An experimental method of testing the idea of ‘synergy’ is yet to be developed, but that doesn’t mean it won’t happen. Remember we’re only half way up the mountain…
Orthodox medicine’s criticism of herbalism is old hat. Requiring herbal medicine to prove efficacy using methods designed to develop medicines from single bioactive compounds is never going to work. Herbalists shouldn’t even attempt to play a game where the rules are skewed in favour of the other team. In any case, it’s not a competition.
Ben Goldacre said,
October 14, 2007 at 11:18 am
sure.
have you got any evidence to show that giving a wholeplant as a herbal medicine is better than giving an extract?
i’m not saying it doesn’t exist, i’d just be more interested to read it than this slightly dreary bombast.
Robert Carnegie said,
October 14, 2007 at 4:14 pm
Isn’t the natural behaviour of a plant to poison anyone who eats it, until we learn not to eat it? Except for those that we have tamed and learned to defuse. Which bits to eat, which not. Potato fruit for instance.
Gramlin said,
October 14, 2007 at 4:31 pm
There appears to be no reason not to conduct a study using, lets say Hypericum (St. John’s Wort) in which half the subjects receive a standardised extract and the other half receive an equivalent amount of whole herb – from the same crop as the extract was obtained. Who shows most improvement? Let’s see.
Repeat this for other herbs, blinded and controlled for, obviously.
My herbalist friend, whom I already mentioned, blamed the peer-review process for research not getting “through” to mainstream, but when I suggested the above, he went on about the lack funding instead. So which is it? Either the studies are flawed due to lack of funding (so don’t blame peer review) or they are not (in which case they will pass peer-review).
Mongoose said,
October 15, 2007 at 8:47 am
No, there is no ‘evidence’, but you’ve missed the point of my argument - there’s a surprise! - I guess it’s not for nothing your column is called bad science…
gonzo said,
October 15, 2007 at 9:48 am
Ayurvedic Medicine is the primary health care system in India and a few other countries and serves nearly 1 billion people. It uses methods developed over 5,000 years which bear little relationship to western medicine. A large part of this system is the use of herbs for healing. The herbs and how they should be prepared and used are published in “The Ayurvedic Pharmacopoeia of India” in 5 volumes. These are “legal” documents. Modern Ayurveda has identified more than 1200 useful plants. Different parts of the plants are used in complex hetrogeneous mixtures. http://www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=hstat1.section.95686
None of these concoctions has ever been tested in the West by random double blind controlled trials. Are we to conclude then this whole healthcare system is unsound? Are these so called remedies ineffective, or worse make people ill? How can a system that treats so many people be so flawed?
Mongoose said,
October 15, 2007 at 1:55 pm
Good to see that my patellar reflex is in good working order…
Let me join the dots for you on the whole plant vs extracts issue:
Herbalists use whole plants medicinally. Medics use single bioactive compounds medicinally. Why should herbalists who use whole plant medicine provide evidence of efficacy for extracts, even if these are from plants? Single bioactive compound research is orthodox medicine’s area of speciality. If you want evidence for the efficacy of active ingredients from plants, then speak to medical scientists, don’t ask herbalists. Extracted, concentrated active ingredients are no longer ‘herbs’.
Orthodox drugs were originally synthesised from plant extracts (like digoxin), but we don’t call them ‘herbs’ and herbalists can’t prescribe them.
Kindly stop pestering herbalists for evidence that is not theirs to provide. I
Gramlin said,
October 15, 2007 at 6:31 pm
Ben’s point was that someone needs to prove that a whole plant is *better* than an extract, before they go around saying that it is.
gonzo said,
October 15, 2007 at 6:40 pm
The original article makes an interesting subtle turn of logic. There are herbs, and other things in nature, which when concentrated are dangerous. Some of these are used in medicine and are dose critical. The false logic in the orinal article is that because some substances need tight control, then all substances do. Many substances are not that dose critical and the herbal substances that can cause harm in excess are well documented.
gonzo said,
October 15, 2007 at 7:15 pm
Gramlin,
Depends how you define “better”. The main issue in my mind is side-effects. I’ve seen people crippled with the side-effects of statins for example. I believe herbals do not to have side effects (at least not to the same extent). In my book that makes them “better”, but they will not be as powerful or may take longer to become effective. I can call to mind an example. I sometimes used to get colds that went onto my chest and I could not get rid of the damn thing for ages, and eventually had to go to the docs for anti-biotics. Last year the same thing happened again, but instead of going to the docs I thought I’d try echinacea (6 caps in one day). On rising the next day, I ejected a most revolting load of green phlegm and that was an end to it. Not a single cough after that. In my mind it was “better” than the anti-biotics because it was much faster and anti-biotics can cause gut dysbiosis. Also, sometimes the antibiotics in the past had been completely ineffective and I’d had to go back for more. My doc says he has to keep changing them because I could become sensitive to the same one and eventually they would not work. Now I take Echinacea at the first sign of colds/sore throat and it always helps. The Mayo clinic says “A National Center for Complementary and Alternative Medicine study released in 2005 found that echinacea did little to prevent or shorten colds.” The clinic also says “Most studies indicate that zinc reduces the duration of cold symptoms but doesn’t prevent colds. However many of the zinc studies — both those that find the mineral beneficial and those that do not — have been flawed”. Why do we have so many flawed studies? They cost so much, you think we’d have learned to get them right by now! My point is that just because there is no study backing it up does not mean it does not work.
Gramlin said,
October 15, 2007 at 9:23 pm
I never said these things don’t work, and neither did I ask about additional bonuses, such as fewer side effects.
Don’t misunderstand me. All I asked was for evidence to show that treatment using a whole plant is more effective than treatment using an extract from the *same* plant.
It’s a claim herbalists make, and is pretty crucial, so why not prove it true?
It was Ben’s point, not mine, I believe.
Ben Goldacre said,
October 15, 2007 at 10:01 pm
i’d actually really be interested to see some evidence for whole herb vs extract, i assume there must be some somewhere? all this stuff about using stuff at low doses so there is no effects or side effects and anecdotal little stories is like the homeopathy nonsense.
seriously, herbalists, whole herb vs extract, please, we want to read about it, give us some evidence.
if there is some you dont have to deliver it in a triumphalist tone, just post the journal reference.
gonzo said,
October 16, 2007 at 12:42 am
“anecdotal little stories is like the homeopathy nonsense”
What do you mean by that? Please explain
gonzo said,
October 16, 2007 at 12:50 am
Gramlin,
>>Don’t misunderstand me. All I asked was for evidence to show that treatment using a whole plant is more effective than treatment using an extract from the *same* plant.
gonzo said,
October 16, 2007 at 12:57 am
Sorry , I did not realise this blog truncated so abruptly.
I’ll try again.. >>Don’t misunderstand me. All I asked was for evidence to show that treatment using a whole plant is more effective than treatment using an extract from the *same* plant.>someone needs to prove that a whole plant is *better* than an extract,
gonzo said,
October 16, 2007 at 12:59 am
Sorry , Arrows seem to upset this bolg, I’ll try try again.. !!Don’t misunderstand me. All I asked was for evidence to show that treatment using a whole plant is more effective than treatment using an extract from the *same* plant.!! No you did not. you said !!someone needs to prove that a whole plant is *better* than an extract, !! But you did not define *better than*. More effective is not the same thing. What do you mean by more effective? Works quicker? Cures more people? Please be precise.
Gramlin said,
October 16, 2007 at 9:20 am
OK, I’ll try again too.
(I actually said someone needs to prove it *before* they go round *saying* it – that’s what I asked for).
Lets set aside value judgements like “better” for a minute. What research has been done on the difference between administering a whole plant compared to an extract from the *same* plant? Not compared to pills or drugs, exactly, simply between whole herb vs. standardised extract?
For example, my herbalist friend asserts that using whole Kava Kava is safer than using an extract of kavalactones, because, I’m told, the natural glutathione present in Kava supports the liver against the otherwise toxic kavalactones. Whether this is an evidence based belief or a conjecture I do not know.
gonzo said,
October 16, 2007 at 12:12 pm
I’m not sure that anyone would ever undertake that research per se on account of ethics. Kava has been used as an infusion in the South Pacific for a long time without reports of negative effects. The kava extract on the other hand has given rise to reports of liver disease. Users are warned about excessive dose. This paper asserts the same http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T8D-4GK1GJN-2&_user=10&_coverDate=08%2F22%2F2005&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=7413d0b60a377516b680f0bf6ca638d6
This report looks into kava induced liver disease http://www.cdc.gov/MMWR/preview/mmwrhtml/mm5147a1.htm
It asserts