excellent article!

What a disgrace to be promoting homeopathy, for which the reliable, peer-review evidence is non-existence. Not just that, but the idea of homeopathy violates basic science. The ways in which homeopathy is said by its promoters to “work” (the memory effect, etc.) amount to scientific illiteracy. It is, as the British Medical Association called it, “nonsense on stilts”.

I note also that the question of evidence (i.e. how many peer-reviewed studies show that homeopathy is effective vs. how many show that it’s not) is not discussed here, nor on the FAQ page for homoepathy on this site, even though Kaplan happily criticises other remedies on the basis of published studies.

In fact the FAQ page is very telling: Under the “Why are there so many negative articles about homeopathy” question, there is no answer; only the admission that homeopathy has been attacked for a long time. Yes, it has been, and for good reason: there is no good data that supports its efficacy.

It’s shameful to be telling people this is a genuine treatment. Doctors really should know better than this; they should be well aware of how to assess the preponderance of evidence and its quality. Clearly not in this case.

I challenge Kaplan to list the scientific studies that show homeopathy to work AND show their context in the wider literature (how many show it works, how many show it doesn’t, what is the quality of the data, and so on). My guess is that Kaplan will not rise to this. This website already fails at listing evidence and instead opts in places to criticise the “reductionist” paradigm of science instead. (If you can’t prove what you believe, attack the accomplishments of others.) The science Kaplan attacks is the same that has doubled average life expectancy in 100 years, eradicated smallpox and many other great things. Homeopathy has nothing of the sort to show for it beyond the placebo effect.

Those interested in understanding what is wrong with homeopathy and why evidence for it is lacking should read Ben Goldacre’s book Bad Science. It explains to non-scientific audiences how data is assessed for medical treatments (and why) and why homeopathy doesn’t meet the standard.

A good article to begin with is here:
http://www.guardian.co.uk/science/2007/nov/16/sciencenews.g2

You are so right Brian. Antidepressants are pretty scary – but I have found them ok for acute sitautions once or twice.
Imagine the phone rings – its the Police to say your loved one has just been killed. Aconite 10M or diazepam? I just dont know.
Lx,
Steph.

the fiction written years ago does come to reality some day ..indian mythology talks about all the adversity humans will have to see one day and this is what is happening with so many diseases and terrorism around..
for antidepressants many studies have shown they are addictive and yes come with adverse side effects ..it is time that awareness is brought about for such side effects and for what homeopathy can offer these people ..

Hi Stephanie,
Diazepam is a tranquliser not an anti-depressant. It has many uses in medicine. Antidepressants are much more contraversial and would be of little use in the situation
you describe whereas Aconite 10M may well be of use – in my opinion.

Dear Luke,
I apologise for the lateness in approving your comments and those of the other commentators. I am not an ‘agit prop’ blogger who sits in front of monitor all day – but a practising physician.
Thanks for commenting – even though I much prefer to talk to people who have the courage to use their true name on the internet. I’ve heard your position enunciated many times and just as angrily. May I respectfully point out to you that this post was about antidepressives not about homeopathy which was merely cited to show the hypocrisy of those who want homeopathy removed from the NHS for financial reasons (sic) but have nothing to say about antidepressant prescriptions soaring despite the poor evidence for their use. Your comments would be reasonable in relation to the many posts I’ve made about homeopathy and the NHS but it’s got nothing to do with this post. It is not I who isn’t addressing the issues raised, it’s you. Why don’t you address my point about antidepressants? After all the Kirsch et al metanalysis which I cite is VERY scientific, Luke Scientiae. You might also look at what I and other have written about Science, Scientism and Para-science. Then maybe change your name to Luke Scientismiae or Luke Parascientiae. Better still, have the guts to use your real name when you make a comment. We are talking about medicine here after all.

Dear Luke,

Regarding your request :
“list the scientific studies that show homeopathy to work AND show their context in the wider literature (how many show it works, how many show it doesn’t, what is the quality of the data, and so on)”.
You will find a report commissioned by the Swiss health authorities to inform decision-making on the further inclusion of homoeopathy in the list of services covered by statutory health insurance here:
http://www.springer.com/medicine/complementary+&+alternative+medicine/book/978-3-642-20637-5
The report comes to the following conclusion:
“In conclusion we have established that there is sufficient supporting evidence for the pre-clinical (experimental) as well as clinical effects of homeopathy, and that in absolute terms, as well as when compared to conventional therapies, it offers a safe and cost-effective treatment”.
The Swiss health system surely cannot be accused of being “scientifically illiterate”. Therefor I suggest you read the report as you will find plenty of evidence and valid references to accredited research articles in favour of homeopathy’s efficacy.

Thanks Alix,

However homeopathy’s detractors will not appreciate you ‘confusing them with facts’ such as these. Homeopathy is an affront to their deterministic and mechanistic view of the world and they will do everything
they can to force that viewpoint on the people whether the people want it or not. See my post on homeopathy, politics and liberty.

Reply to Dr Kaplan:

MY ANONIMITY:
My blog (lukesci.com) is presently down for maintenence, but its About page explains why I use a pseudonym. The basic reason is that a number of cranks and pseudoscientists have attacked criticism by litigation (an attempt to silence scientists they don’t like), rather than providing evidence. I don’t plan to be on their list. (For an example, see here: http://www.guardian.co.uk/science/2010/apr/15/simon-singh-libel-case-dropped1)

EVIDENCE:
The veracity of truth claims made by homeopathy advocates does not rest on their authority as doctors, nor my anonymity or otherwise. It rests on the EVIDENCE.

It’s no good substituting, as you did, Dr Kaplan, the question of evidence with allegations that mainstream science is somehow determined to keep out alternative explanations; you first have to show that reliable data warranting those explanations actually exist.

The claim that mainstream science is somehow inherently biased towards mechanistic explanations is fallacious. History shows that science accepts changes of paradigm wherever the preponderance of evidence mandates it, from the Copernican revolution to the development of quantum physics (btw, quantum physics is NOT mechanistic but quantum theory is the theory that most precisely matches experimental evidence). There are many other examples.

Dr Kaplan goes on in his comment to attribute to me a “deterministic and mechanistic view of the world”, which he cannot have any information about since he knows exactly nothing about what I think about determinism, for instance. It’s a change of subject and it’s a personal attack. Very far from answering the call for evidence validating what you do and claim to be true.

I’ve been extremely disappointed to find that not only did you refuse to engage me on the evidence, Dr Kaplan, and that you resorted instead to insults and unfounded imputations, but were openly critical of Alix, who did – in your defence, I emphasise! – attempt to provide evidence. I was right in my prediction:

“I challenge Kaplan to list the scientific studies that show homeopathy to work AND show their context in the wider literature (how many show it works, how many show it doesn’t, what is the quality of the data, and so on). My guess is that Kaplan will not rise to this.”

Kaplan also refers to hypocrisy and ignores my comment about his: He fails to provide links to peer-reviewed literature (and, as we see above, is derogatory towards others’ attempts), whilst criticising anti-depressants in his post on the basis of peer-reviewed literature. You can’t have it both ways.

—————

Reply to Alix:

Thank you for standing up for your beliefs and providing the link.

However, the journal you cite is dedicated to CAM (Complimentary and Alternative Medicine). That’s hardly a neutral source, since it is dedicated to supporting the very thing that is in question, and which – were it discredited – would end the existence of the journal.

You might be interested in wider studies and meta-studies, such as this one published in J Med Aust, 2010:

http://www.ncbi.nlm.nih.gov/pubmed/20402610

It concludes:

“Collectively, the six reviews that I appraised failed to provide compelling evidence for the effectiveness of homeopathic remedies. These reviews, being Cochrane reviews, are likely to be more reliable than other sources of evidence.6 Furthermore, as most were authored by homeopaths, it seems unlikely that they were biased against homeopathy. In fact, one might argue that they were biased in favour of homeopathy.

“…In conclusion, the most reliable evidence — that produced by Cochrane reviews — fails to demonstrate that homeopathic medicines have effects beyond placebo.”

I would also ask why it is that representatives of, e.g., the British Homeopathic Association, squirm away from recommending homeopathy for malaria prevention/treatment (e.g. here: http://www.youtube.com/watch?v=s1UJ_qGZ24k) or why there have been so many disastrous consequences when people have attempted to treat their diseases homeopathically: http://www.youtube.com/watch?v=QdLOBRUwq10.

All this comes down to a desire to believe that homeopathy works, or cherry-picking the data to enable the confirmation bias. And it shows in what happens when an accurate appreciation of the evidence isn’t of concern.

I don’t attack those who have been misled by cherry-picked data, but I do attack those who refuse to provide the evidence they should be eager to show if it truly supports their claims. Dr Kaplan, as we see from his attacks on me, is a perfect example.

Okay Luke, you have brought up several points and I’ll answer them this time – not without reluctance because this post was about antidepressants, the lack of evidence behind most of their prescriptins and the hypocrisy of attacking homeopathy on the basis of ‘we can’t afford it’ ( NHS drug budge £10) when these non-evidenced SSRI antidepressants have a budget of at least £232 before the reports on their ‘soaring’ and ‘skyrocketing’ prescription rate. You fail to comment on this, central theme of the post and simply segway off into attacking homeopathy. In future I will only publish comments pertaining directly to the main theme of my posts.

INSULTS: You use the words ‘disgrace’ and’shameful’ towards my writing and repeat the ‘nonsense on stilts’ (is this EBM?) insult spewed at some pathetic meeting of junior BMJ (I think) – and I now think they rather regret saying that. Not that you shouldn’t be inspired by it of course. I think the word ‘witchcraft’ was used there too.

I admit using the words ‘mechanistic’ and ‘deterministic’ towards the sort of philosophy I frequently find lies behind comments such as yours in your note but (these are not necessarily insulting words and will aways be seen as objectively less offensive than calling someone a ‘disgrace’, ‘shameful’ and their art ‘nonsense on stilts’ My guess is that you are a lot worse at receiving any form of insult than you are at dishing insults out.

ANONYMITY: I don’t accept your explanation. Untrue damaging ad hominem attacks on people are still illegal in the UK. Personally I favour the USA style 1st Amendment influenced law. ie Freedom of Speech. However unless you attack individuals unfairly, you should not fear libel at all – as British justice showed in the Simon Singh case. Where the campaign to ‘Keep Libel Laws out of Science’ was wrong is they would have like to have continued attacking (in many vile ways) individuals whose views with which they disagreed – without fearing libel for their unprovable ad hominem slights. Get rid of the libel laws altogether and I for one would not object. But to campaign to get libel laws ‘out of science’ is utterly disingenuous because it seeks to allow people who call themselves ‘scientists’ to break the libel laws and make untrue ad hominem attacks people whose views on science annoy them. What nonsense.

SCIENCE & SCIENTIFIC STUDIES: I do NOT attack science. I am a doctor, have practised surgery and treated a syphilitic primary chancre, meniningitis, pneumonia etc with Penicillin. But I do attack SCIENTISM and PARASCIENCE. I also draw attention to the fact that huge swathes of conventional medicine are far from evidence based as this graph in the BMJ’S CLINICAL EVIDENCE clearly shows. Let the ‘authorities’ create a level playing field and say what ‘level of evidence’ is needed for use on the NHS. Reasonable suggestion, Luke? Or am I asking too much? Or is the fact an overambitious young doctor at a junior BMJ meeting called homeopathy ‘nonsense on stilts’ good enough ‘evidence’ for you. As far as I am concerned outcome studies all over the world, in particular Bristol are very convincing that homeopathy has powerful therapeutic effects – however it works.. Any decent human being or honest doctor would want to investigate what is happening at these institutions – rather than insult, trash call for the NHS banning on homepathy. And any decent doctor would also look at that BMJ pie chart and be honest about it. I’ve written about this here and here. So feel free to comment on those posts Luke. Comments on this one will now be restricted to those that address the subject of the post.

ANTIDEPRESSANTS: This is the subject of this post I have written extensively about homeopathy, studies and the nasty, hypocritical and disinginuous attack on NHS homeopathy (citing financial (sic!) reasons for the reason to abolish NHS funding of it) when so much of conventional medicine lacks an evidence base yet is MUCH more burdensome to the taxpayer than homeopathy. Antidepressants are just one rather good example of this. If you want comments published in this blog, then please write about the the subject of any given post. Comments on other issues on my site can be made in relation to those posts and I’ll certainly publish them – even though you continue to hide behind a mask.

Dear Brian,

Firstly apologies for entering the fray rather late, but I have only just picked up this debate via twitter.

For the record, I am a livestock farmer who uses homoeopathy on a regular basis with good and sometimes surprising results.

The point of your post is well made (i.e contrasting the hypocrisy of the NHS spending on antidepressants vs homoeopathy and the subsequent flak that the latter gets whilst not an eyelid is batted over the former). For my part, I get even more angry when I consider the naked waste made by the NHS in huge requirements for IT and administration which serves no real clinical benefit. This phenomenon is never pointed out by the Skeptics [sic].

However (and at the risk of incurring your ire in going off topic) I feel I have to respond to Luke’s reference to one of Edzard’s little ruminations, to which Luke links above.

Firstly, it is published in the Medical Journal of Australia. I am unable to say if this is a first rate medical journal or not, but it seems odd that Ernst felt compelled to send what would otherwise be an item of interest (albeit a short one) so far afield. His observations could just as easily have been made in a UK or US journal and have been given a wider audience if they had been.

Ernst’s inclusion protocol from within the Cochrane database are fair enough and the numbers of reviews accord with my own searches.

His assessment of individual reviews are arguable in two or three cases, but principally it is his use of “non-significance” in a review to mean “No effect” by the time he gets to his conclusions that I take issue with, because he does it habitually with many of his so-called systematic reviews. In this case his conclusion states: “….the most reliable evidence…..fails to demonstrate that homeopathic medicines have effects beyond placebo.” Given that he does this with a set of Cochrane reviews; and given that the Cochrane guidelines repeatedly state “No evidence of effect is not the same as evidence of no effect” I find it astonishing that this paper got past peer review.

Thank you very much for your considered and erudite comments. I am delighted to hear from a farmer who values homeopathic treatment of his animals.

All I can say after reading some of these comments is that the proof is in the pudding. If you haven’t tried it then you can’t really comment. Also how many so called scientifically tested drugs are later removed due to adverse affects on patients. The saying “the side-effects are worse than the complaint” is becoming more common with medical drugs and also Herbal remedies.

I only wish I found Homeopathy before I left my well paid job. It would have helped me cope with the stress both physical & mental rather than the anti-depressants that caused Anxiety attacks far worse than I had ever experienced before being prescribed them for Anxiety.

If you ever get unwell enough you’ll soon come to “give it a try” and like so many other sceptics and open minds alike wish you’d known about it years ago.

Ten years ago science put down Alkaline water saying it was a rip off, now they say it been proven to have positive health effects. Google “Unique water” I think that will bring up the website.

We may be living longer but prevention is needed to avoid the pain of treatment for most. I’d rather die than be treated for 5-10 years with poor quality of life and then die a horrible death anyway, as I have seen so many others suffer. If modern medicine & treatment is required for life saving then thank goodness we have Homeopathy to help us survive the “treatment”.

thanks Sue, I’m sure you’ll have something to say about the today’s post (11/5/2012)

Sue

Come, come, now. How can you say that?

New EU legislative bodies have quite clearly refused to permit any claim that water is good for you.

According to committee, it doesn’t even cure dehydration!

And if it has trace boron, like the Italian spring water that was used by arthritis sufferers, it’s sale is banned!

How can you possibly question the Authority of EU Science?

I refer to the notorious “allowed health claims list”, the EU’s controversial Nutrition and Health Claims Regulation Community list of 222 health claims – as approved by the European Food Safety Authority (EFSA), the Nutrition and Health Claims Regulation (NHCR; No 1924/2006) under the auspices of the Committee on the Environment, Public Health and Food Safety (ENVI).
And the banning in secret of borax sales under REACH, of course.

Such acromyns are not to be argued with!

But wait!
now we have this: bit.ly/LX4VMC
EFSA budget approval postponed over conflicts of interest
and this (9th May): http://www.epha.org/a/5150
EFSA chair forced to resign over conflicts of interest

Which all rather points to F.R.A.U.D. and C.O.R.R.U.P.T.I.O.N.

Perhaps the junk Science isn’t so good after all.

(With apologies for late post & sarcastic tone..)

Hi Dr Kaplan.

Have just stumbled across your interesting blog from elsewhere, and was particularly taken by this article. I’m always intrigued by CAM proponents who cite the damning meta-analyses into anti-depressants, but are resistant to similar studies exposing their own particular field. If the evidence is sufficient to pull the rug from under anti-depressants, why not, say, homeopathy? Could committed biopsychiatrists, for instance, argue that the first author, Prof. Kirsch (a psychologist) has never trained in psychiatric prescribing and therefore is unqualified to report on their efficacy/action? Or that the combination of psychiatric consultation and prescribing doesn’t lend itself to RCT? Or demand that skeptics explain why they witness so many patients getting better and returning for repeat-prescriptions? Or that anti-depressant prescribing is rising year on year….

You also mention this:

“[Anti-depressants and addictive tranquilising drugs] have potentially dangerous side effects and patients really should be monitored carefully on them – if their doctors can find the time of course. But it seems nobody cares that much about these horrific statistics. After all should we really be knocking such a profitable industry during a recession? Much better to start a new organisation protesting against all that money wasted on treating naive patients who want NHS homeopathy.”

Firstly, a small corner of the psychiatric and related mental health professions are trying to do something about it. I point you in the direction of the Critical Psychiatry network. Notice, here is a professional group that subjects psychiatry and the pharmaceutical industry to rigorous scientific and ethical scrutiny, but at no point have I witnessed a member relax their scientific/ethical standards and make an appeal to CAM. (Whether or not these groups are having anywhere near the desired effect is another debate, but it does challenge your suggestion that the profession is doing nothing about unethical and unevidence-based treatments.)

I’m also interested in your point here and elsewhere about the improbable naivety of homeopathy consumers? So what do you think’s going on with the mild-moderately depressed who take anti-depressants? Are homeopathy patients inherently wiser than the medicated depressed?

Thanks for your time.

Adzcliff

Dear Adzcliff,
Thanks for you thoughtful remarks. In answer I would say that I don’t think the rug should be pulled under anti-depressants or homeopathic remedies. I think that GPs are educated enough about both to make responsible decisions for their patients. That illustrates the disingenuous side of the campaign against NHS homeopathy. What the detractors really want is for GPs who want to send patients for NHS homeopathy to be thwarted. You cannot get NHS homeopathy without your GP referring you to another medical doctor who uses homeopathy? Is that not enough protection for patients?

Re: Concerned psychiatrists: Yes, I know there are psychiatrists who take these things seriously. However I attended a huge meeging for GPs on ‘How to treat depression in the 10 minute consultation’ – sponsored by makers of anti-depressants. Mild and moderate cases continue to have SSRIs thrown at them – by GPs and psychiatrists – despit the evidence and the views of the concerned and ethical.

Re: Re: naivety etc: Yes of course those patients (mild and moderate) are getting better due to placebo/self-induced healing/suggestion/hypnosis and I say: ‘For the moment let it be. It’s the outcome that counts’ And IF its placebo that’s doing the job I’d rather take a homeopathic remedy than an SSRI. But the main point is that the witch hunt against homeopathy is disingenuous because it singles out homeopathy for special attack under the guise that it’s a waste of taxpayers’ money. SSRIs are a MUCH bigger waste (if you look at things that way and I don’t) and that surely illustrates the hypocrisy I am talking about.

Dear Dr Kaplan.

Thanks for this. I suppose your position is a little clearer to me now and isn’t as hypocritical as I first supposed. If you’re advocating homeopathy as a safer placebo than anti-depressants for the mild-moderate depressed, then I suppose that’s an argument. Had you been ignoring the evidence to insist homeopathy is medicinal, and using evidence to show anti-depressants probably aren’t, then that, I suggest, would’ve been hypocritical.

It’s disappointing but unsurprising that your drug-sponsored meeting advocated a narrow biochemical approach to depression, but disagree that this goes anyway towards an argument for any other unevidence-based treatment, regardless of how safe or comparatively little is spent on it.

Anyway, thanks for your time.

Dear Adzcliff:

In all walks of life there is a tendency to support those whose opinion one agrees with. Homoeopaths, psychiatrists, GPs and the like are no exception. The problems emerge when we are each trying to be strictly objective and fair to those whose opinions do not concur with our own, by looking carefully at the way those opinions are framed. Publication of a meta-analysis in a peer reviewed journal does not, in itself, make that paper accurate, fair, unbiased or meaningful in a clinical context.

In terms of meta-analyses and reviews, homoeopathy has had a bit of a raw deal in recent years – although it has not all been negative. Whilst some meta-analyses are cricised as being beholden unto the pharmaceutical industry because their authors are in some way connected to that industry, other meta-analyses are equally prone to a non-financial bias because of the prejudices of the authors. Idealogical convictions can lead to both positive or negative biases.

A good example of the former is the recent meta-analysis on statins, reported here: Pulse:http://www.pulsetoday.co.uk/newsarticle-content/-/article_display_list/13960036/gps-should-prescribe-statins-to-all-over-50s

This is typical of a large meta-analysis (and its equivalent, the huge mega-trial) where a huge sample size inevitably creates tiny effect sizes because of between study and within study heterogeneity, amongst other things such as variable outcome measures, variable application of protocols between centres and so on. Read that uncritically reported link and then just look at the responses from GPs. The sheer insanity of the whole over-50 population taking a powerful drug for up to, say, 40 years in the hope that a tiny proportion will benefit, is to me just mind blowing in its stupidity. Hopefully GPs will ignore this meta-analysis.

However, you say: “…..demand that skeptics explain why they witness so many patients getting better and returning for repeat-prescriptions?” Unless you are being heavily ironical, here you are actually pointing to the same problem that so many clinicians in CAM experience. That is, the patients get better despite the findings of meta-analyses. There is a disconnect between the formal findings of EBM and clinical experience.

I suggest to you that the elegant beauty of the EBM model, and which has beguiled so many in medicine and elsewhere, is actually a rather flawed beauty with serious shortcomings – not least because of this inability to take serious cognisance of clinical experience.

Dear Adzcliff,
Let me be clear: ‘I do not advocate homeopathy as a safer placebo…’ Anybody who studies and practises homeopathy and thinks it is placebo must be quite mad considereing
a) the years of study required to learn the medicines
b) the time spent with each patient in order to find a medicine that suits them.
When I say homeop. remedies are safer than SSRIs I am trying to compare them from the point of view of the skeptic in an attempt to show how disingenuous many of the skeptics are when they attack NHS homeopathy (budget £10m annually for the meds as opposed to £232m just for SSRIs) This is especially hypocritical when they say the taxpayer’s money is being wasted on homeopathy.
If I’ve sait it once, I’ve said it 100x:

I call for a level playing field in the NHS to decide whether any intervention should be available.

Some may say a certain level of EBM. Well then let’s hear what that level is. Some will say (eg myself) that if it’s safe and if the people want it and the doctors prescribing it are responsible and accountable for their actions and the only way you can get it on the NHS is if your GP refers you – then of course it should be available on the NHS

Dear David,
I would like to say that reading your comments on my blog is an absolute pleasure. In the inarticulate, ungrammatical world that is the internet, it is a pleasure to read your erudite comments. That I agree with them is also a bonus

Hi David – thanks for that.

My reason for posting hypothetical arguments for biopsychiatrically-minded health professionals was to illustrate exactly what you identify. There’s a fair proportion of CAM users/practitioners who will (1) argue that clinical trials don’t account for clinical experience and ‘real world effects’; and (2) cite the SSRI meta-analysis as an example of how conventional medicine isn’t what it’s cracked up to be. My argument is that you can’t have it both ways. If Kirsch et al’s study is meaningful, then so are equally exposing studies into CAM; if it’s not, then it’s of no use to them as an argument. Likewise, if CAM proponents can deflect negative evidence through ‘clinical experience’, and/or by denying it’s generalisabilty to the real world or consultation room; then this is surely an equally useful argument for biopsychiatrically-minded mental health professionals. Again, this neutralises Kirsch’s meta-analysis as an argument for or against anything.

My personal view is that it’s the intellectual responsibility of the clinician to explain why their clinical experience doesn’t tally with the evidence (e.g. placebo, regression to the mean, consultation effects, selective/inaccurate recollections), rather than using clinical experience as a barometer for good research. For instance, the confident psychiatric prescriber might want to ask themselves how their patient-types are so consistently excluded from the trial data, so as to throw up so many false negatives. (Unless of course their patients are, on average, just like any other and something non-medicinal is at play – which was the shift in outlook I took in response to Kirsch et al.) Likewise, the homeopath or reiki healer (etc.) might want to give this some thought too.

Cheers for now.

Adzcliff

Brian,

Oh, come now sir, such praise……. ’tis unwarranted……modesty prevents…..etc.

Cheers,

David.

When old age shall this generation waste,
Thou shalt remain, in midst of other woe
Than ours, a friend to man, to whom thou say’st,
‘Beauty is truth, truth beauty,—that is all
Ye know on earth, and all ye need to know.’

I don’t ‘want it both ways’. I want a level playing field for all possible therapeutic interventions on the NHS. Or is that too much to ask for?

Thanks again Dr Kaplan.

Firstly (a) the length of time it takes to study a subject, says nothing about it’s validity. Some people spend their lifetimes studying their chosen religion, but at best, only one can be true (e.g. either the Abrahamic monotheistic claims are true, or Hindu polytheism is; or Bhuddist’s have it when they say God isn’t ‘up there’ at all but ‘in here’). Secondly (b), we know that skeptics have exposed the unreliability of homeopathy/homeopaths – different homeopaths diagnose and prescribe differently – so this time spent with the patient doesn’t necessarily enhance reliability.

As for those skeptics who attack homeopathy (and other CAM), but nod through the ethical and scientific controversies of big pharma/’conventional medicine’, I’m not sure I know who they are? I’d be more than happy to debate with them too. For instance, you may be aware that one of homeopathy’s most vocal and impactful critics, Ben Goldacre, is soon to release a book all about the malpractices of the drug industry. My understanding is that good skeptics aren’t partisan, and go wherever the evidence leads them…

Regards.

Adzcliff

Now we differ significantly;

on (a) I mentioned the time spent with patients and the huge amount of time invested in studying as evidence that homeopaths do not believe that homeopathy is placebo. Only that. You cannot practise homeopathy if you think it’s placebo.

aside: (actually I don’t think your example of religion is valid at all and many would say that those positions are indeed compatible. Different gods are the same as different aspects of one god – that sort of thing)

on (b) No ‘we’ don’t know this at all. Homeopaths treat people with conditions, not conditions independent of people.

on ‘My understanding is that good skeptics aren’t partisan, and go wherever the evidence leads them…’ Well, Adzcliff I simply cannot agree with that. Just because Ben Goldacre (how did he rise so quickly to such heights in journalism is anybody’s guess) also criticises Big Pharma does not make what you say about the skeptics true.
1. The attack on homeopathy has been aggressive, loud, jeering and media savvy. Same thing for SSRIs? That will be the day!
2. Those that defend homeopathy are subject to savage ad hominem attacks in the media. The same for the small group of psychiatrists you mentioned who aren’t happy about overprescription of psychotropic drugs? That will be the day.
3. Campaign to ban NHS homeopathy culminating in a biased and imo deeply flawed and discredited S&T committe recommending to the Government that it stops funding for NHS homeopathy? Same for all orthodox treatments that lack the requisite evidence of efficacy demanded of homeopathy. That will be the day!
4. Level playing field for homeopathy and conventional medicine (ie one standard of EBM)? That will be the day.

REASON: Skeptics hate homeopathy because it is an AFFRONT to their belief system not because it doesn’t serve the people or any other hypocritical rubbish they sometimes spout in the name of altruism or saving money. Many skeptics are deeply disrespectful of democracy and cannot accept that NHS homeopathy exists because about 6m people in the UK value it and the SUBJECTIVE opinion of these people is very important (also at the ballot box) as both Labour and Coalition health ministers have recognised.

So three cheers for democracy and two cheers for the skeptics!

Thanks again Dr Kaplan.

As an aside, I think the God/religion analogy does work, as if ‘Different gods are the same as different aspects of one god’ then that rules out the Abrahamics for a start: their scriptures are very clear this isn’t the case (I am the one and only God, tear down the temples of those who worship other Gods etc.). If you’re suggesting a meta-perspective over and above existing religious teachings, then that’s fine, but that’s a whole new religious perspective, not an established one. Likewise, biophysics, faith healing, reiki, homeopathy, reflexology, ayurvedic medicine (etc.) are mutually exclusive in parts, and it won’t do to say they all hold medicinal potential

As for homeopathy being an affront to my belief, I like to think I’d believe any evidence-based idea, regardless of it’s counter-intuitive nature (e.g. time and space are inter-connected, I share a common ancestor with a carrot, a quantum-particle can be in two places at once, taking a little bit of a disease can give you immunity against that disease, there’s no noise in nature without hearing devices…). It’s my view that homeopathy (for instance) is perfectly to be true, and if this was shown to be the case, I’d be all over it!

As for the 6m people who use homeopathy, again, you’d be wrong to infer this says anything about efficacy. If history tells us anything, it tells us that the argument from majority is a fallacy (the earth was never flat). I agree that people should be able to subject their minds and body to whatever they choose in their pursuit of health and well-being, but disagree the NHS should fund anything of unproven efficacy.

(I wonder where you stand on Dr Trossel’s controversial and expensive bovine stem-cell treatments for MS? After all, he seemed sincere, had anecdotal evidence, and a client-base who believed in him? Perhaps it’s about cost, or the amount of people he could convince and go on to convince others?)

Cheers for now.

On religion: Yes a meta-perspective is a good idea just as a meta-analysis is also often useful.

On affronts and you: We are each entitled to our subjective beliefs and can VOTE according to them – as many people do.

On 6m homeopathic supporters: I didn’t say that that anything about efficacy can be inferrred from this. What can be inferred from this is that in a democracy, those people should be able to access homeopathic medicine and in a NHS funded by the taxpayer, the subjective views of those 6m need to be taken into account as Mike O’Brien, Ann Milton, the Head of State and anybody who doesn’t think scientism should trump democracy, believes.

On Dr Trossel: Know nothing about him.

Fair enough. I fear I’m taking up too much of your time. Thanks anyway for the dialogue.

Adzcliff

Not at all. Thanks for commenting.

Ooh, hot of the press in the last hour or so:

http://www.quackometer.net/blog/2012/05/homeopaths-in-panic-over-new-medicines-rules.html?utm_source=rss&utm_medium=rss&utm_campaign=homeopaths-in-panic-over-new-medicines-rules

Just in case you wanted to contribute to the discussion…

Adzcliff,

Firstly, I am pleased that Goldacre is about to publish a book on the shortcomings of the pharmaceutical industry. I had a feeling some time ago that he might, from his regular utterances. However, I hope he will do so in rather more depth and consistency that he managed with his first book, whose major flaw was that it was so superficial.

You state that homeopaths cannot have it both ways when considering evidence from meta-analyses and reviews; but your argument is predicated upon the meta-analysis being both immaculately conceived as well as inviolable. Sadly, neither is true.

There are two principle sources of difficulty for homeopaths in EBM which are much cited by skeptics. The first is Ernst, whose reviews are described as “systematic” when in fact they are narrative. And as such, they suffer from the vagaries of author interpretation and, generally, three consistent errors in procedure. The first is that they do not assess the mathematical bases for the included trials. The second is that they use a system of progressive verbal denigration of those trials that are deemed positive in outcome; or those trials which are non-significant are over emphasised. This process of ratcheting negativity is resolved into a statement in the conclusion which amounts to a view that because there is no evidence, there is no is no effect. Finally, these “no effect” trials are counted, and if they outnumber the positive trials then the whole intervention is deemed to be ineffective. As I am sure you know, “vote counting” and conflation of “no evidence of effect” with “evidence of no effect” are all strictly verboten in systematic reviews and meta-analyses. The reason Forest Plots are deemed necessary in meta-analyses is to take the effect sizes and confidence limits of all included trials, accumulate them and use that to decide whether an intervention is useful or not. Because Ernst never uses the numerical data of trials in his reviews, the forest plot is a process he never uses.

This brings me to the Shang 2005 meta-analysis published in the Lancet. There are so many flaws in this study that it would take two or three whole articles to deal with them. Many of the biggest problems have been dealt with by others more qualified than me. But, briefly, to look at a couple of issues which have scarcely been dealt with in the literature so far:

The title of the paper (I am having to do this from memory, so forgive me if I am a little vague) suggested that the meta-analysis was to determine whether homeopathy was better than placebo. The final conclusion of Shang et al was that it was not. So, on the face of it, they answered their own question. But the body of the paper proceeds to develop a unique (to my knowledge) methodology of comparison of homeopathic treatment of a medical condition, matched against a conventional treatment for the same condition. Whilst pairing participants in an individual trial is sometimes used, depending upon the experimental design, I have not seen this methodology used anywhere in any other meta-analysis and neither was the method justified within the paper. The method is not described or mentioned in any of the textbooks on meta-analysis in my possession. Despite the effort put into pairing studies, those pairs were not analysed as pairs, but the gross data simply piled into an interesting but not very informative funnel plot.

At this point the authors declare that the average quality of the studies in homeopathy was slightly better than that of the conventional studies included in the paper. It also recorded that there was a weak positive effect in favour of homeopathy. What happened then was that the number of studies was trimmed down to twenty-odd on the grounds that the others were of poor quality. And then the studies were further trimmed to eight. This final group of eight included a big trial on the use of Arnica for marathon runners and which suggested no effect. This is a well known trial, criticised in its way for being poorly executed homeopathically, but I think my biggest objection to it is that we don’t use antibiotics to treat perfectly healthy people and expect to see an improvement in their condition. The authors had thereby reduced the number of included trials down to a minimum, and hence the one negative result influenced the overall results sufficiently to give a negative result for homeopathy. In the conclusion to the addendum submitted later, the authors suggest that the assymetry of the funnel plot shows that there was bias in the homoeopathy and that therefore homeopathy is no different from placebo.

I have difficulties with this reasoning, because assymetric funnel plots have so many causes aside from publication bias that nothing really can be determined from them one way or another. The forest plot contains more information and if constructed properly can potentially throw up far more of interest. However, Shang et al did not use a forest plot at all (or at least if they did, they did not publish it). This is surprising because the lead author of the study, Matthias Eggar, knows all about them, having contributed to at least one of the books on meta-analyses in my collection.

It all looks very iffy. And, lets face it, if a study or review is obviously flawed, then what clinician in his or her right mind mind is going to use it to guide their treatment of the individual patient sitting in front of them?

The point is that EBM is far from perfect and sometimes downright flawed. It does not always provide the perfect link between cause and effect, despite the skeptics’ vehemence. And as Sackett and others recognised in the 70s and 80s, there is a place for clinical judgement and experience when treating the individual patient.

Cheers,

David

On adzcliffe’s scoop: The day I write anything on Andy Lewis’s blog will be the day that the men in white coats come to take me away.
The post you have read simply celebrates the condescending and patronizing authoritarianism and statism supported by people whom I’ve dubbed: THE DISCIPLES OF SCIENTISM and ‘THE I KNOW BETTER THAN YOU WHAT’S GOOD FOR YOU BRIGADE’

Brian,

Just before the rest of us spring-clean the padded cell in anticipation of your arrival….

I think that one day soon, your consistent exposure of the Skeptical Supreme Soviets’ idea that patients can be herded into only one form of state-approved medical treatment, despite the wishes of those patients, will bear fruit.

Despite the huge advances made by modern medicine, the skeptics continual tip-toeing around the enormous chasm of some of EBM’s more obvious shortcomings is fascinating to watch. And your observations of this practice are both perceptive and vital to health care philosophy.

To your continued good health,

David.

David,
That means a lot more to me than you might think.
Thank you.
Brian

Thanks for that David (have just popped back here for Dr Kaplan’s Guardian reference).

You seem to be much more statistically minded than me (which may be the same as me admitting that I can’t tell if you are or not). I think you’re absolutely right to point out the limits and flaws of EBM, and you’ve definitely provided some food for thought. As for the funnel plot graph – I gather you’re not a fan – I certainly continue to be compelled by this illustrative technique for detecting publication bias in the pharmaceutical industry. It also makes me wonder how the anti-depressant data distributed before and after Kirsch’s meta-analysis: seeing as he’d unearthed about 40% of the entire evidence-base. I personally think John Ioannidis at University of Ioannina is a pioneer here – I recommend him if not already onto it.

However, unless I’m misunderstanding you (quite possibly), my argument doesn’t predicate on anything. I’m arguing that the choice is that of the apologist to respect or reject the informative potential of the meta-analysis – bearing in mind that many will conclude ‘insufficent evidence’ – and incorporate or eliminate them fully in their discourse. For the record, I’ve never heard anyone describe them as ‘immaculately conceived’ or ‘inviolable’ – this would be unscientific.

From reading between the lines, I’m wondering if your critique runs something like this: science can’t give us absolutes, therefore nothing is certain, therefore no belief/theory is any more/less valid than any other, therefore all theories are equal (i.e. CAM and EBM). I’m not immune to a bit of post-modernist thinking myself, but that’s a little too near solipsism for me.

Anyway…

Adzcliff,

Firstly my apologies for not being aware of, or having access to, Kirsch’s meta-analysis. Secondly, you will have to forgive my gentle irony when describing the structures and procedures of EBM as “immaculately conceived” and “inviolable”. This is a reference to the absolutist assertions of the skeptics that if meta-analysis A says that X is the case, then X is always the case and always will be. My point is that if meta-analysis A is flawed in its methodology, then all of us should have difficulties in accepting that X is, and always will be, the case. And that does not even begin to acknowledge that a meta-analysis published in 2005 (as was Shang et al’s) is out of date by 2012. Homeopathy moves on and more studies become available over time.

So, in short I am not being particularly post-modernist in my reasoning. I am simply stating the obvious: that flawed processes, or processes which are not flawed in themselves but which have been improperly or mistakenly manipulated, lead to flawed evidence; and that evidence must therefore be dismissed accordingly. I think that is a properly scientific approach. It is also one which applies in law – there are many high profile cases of miscarriages of justice caused by evidence which has been subsequently found to be “unsafe”.

Now to funnel plots. You are right, I am not a fan of these statistical constructs. And I do not believe that an assymetrical funnel plot is necessarily an indication of publication bias. There are all sorts of sources of assymetry in a funnel plot, including publication bias. In the literature you will find mentioned other sources of assymetry such as statistical chance, fraud, time lag bias and choice of effect measure – Risk Ratio will give a different distribution to Risk Difference for instance.

But there is another source, which is not found in the literature, and that can be understood by considering an intervention where all the studies have been published and located by the meta-analysis. Let us also assume that this is a stonking intervention and that just about all of the studies return a positive effect size. By definition, this will return an assymetrical funnel plot.

Think about what you are doing with this exercise. You are interrogating a set of data points, none of which are intrinsically dependent upon the others, about the existence of other data that you have not yet found – possibly because it may not exist.

The primary assumption of the funnel plot is that the distribution of data within it is random. That is an uncommonly big assumption.

Nevertheless, I agree that Ioannidis is good. His chapter in Rothstein, Sutton and Borenstein – Publication bias in meta-analysis (2005), Wiley, is a good summary on sources of bias and well worth reading.

Now, it is late and tomorrow I have TB testing for my cattle. And that is always more stressful for me than it is for the cattle.

Cheers,

David.

Thanks David.

You’re right, nothing ‘necessarily’ indicates publication bias, but it should raise alarm bells that the evidence isn’t as it seems. As for Kirsch et al’s study, this is the anti-depressant meta-analysis that Dr Kaplan referred to. On the back of this evidence, I agree that it’s a controversy that anti-depressant prescribing is rising year on year (they seem to be little more than an active placebo). I’m a fan of the meta-analysis.

Cheers.

Adzcliff

Adzcliff,

I too would be a fan of meta-analysis. On the face of it, the system has an elegance which intellectually is very satisfying. But (and there is always a “But”) the more you look into it, the more caveats appear out of the woodwork. And whilst there are growing concerns, which are admittedly only voiced by a tiny number of people, there is nothing yet which can be put in its place. We are therefore stuck with making the existing system better by being more honest about its shortcomings; and more rigorous about spotting the fiddles, the poor methodologies and the diabolically bleedin’ obvious insanities – like the conclusions drawn by the statin meta-analysis I mentioned earlier.

Cheers,

David.