Jump to follow-up

I’m bored stiff with that barmiest of all the widespread forms of alternative medicine, homeopathy. It is clearly heading back to where it was in 1960, a small lunatic fringe of medicine. Nevertheless it’s worth looking at a recent development.

A paper has appeared by that arch defender of all things alternative, George Lewith.

The paper in question is “Homeopathy has clinical benefits in rheumatoid arthritis patients that are attributable to the
consultation process but not the homeopathic remedy: a randomized controlled clinical trial”, Sarah Brien, Laurie Lachance, Phil Prescott, Clare McDermott and George Lewith. [read it here]. It was published in Rheumatology.

Conclusion. Homeopathic consultations but not homeopathic remedies are associated with clinically relevant benefits for patients with active but relatively stable RA.

So yet another case where the homeopathic pills turn out the same as placebos, Hardly surprising since the pills are the same as the placebos, but it’s always good to hear it from someone whose private practice sells homeopathy for money.

The conclusion isn’t actually very novel, because Fisher & Scott (2001) had already found nine years ago that homeopathy was ineffective in reducing the symptoms if joint inflammation in RA. That is Peter Fisher, the Queens’ homeopathic physician, and Clinical Director of the Royal Hospital for Integrated Medicine (recently renamed to remove ‘homeopathy’ from its title). That paper ends with the remarkable statement [download the paper]

"Over these years we have come to believe that conventional RCTs [randomised controlled trials] are unlikely to capture the possible benefits of homeopathy . . . . It seems more important to define if homeopathists can genuinely control patients’ symptoms and less relevant to have concerns about whether this is due to a ‘genuine’ effect or to influencing the placebo response."

That seemed to me at the time to amount to an admission that it was all ‘placebo effect’, though Fisher continues to deny that this is the case.

"Homeopathy has clinical benefits in rheumatoid arthritis patients" -the title says. But does it?

In fact this is mere spin. What the paper actually shows is that an empathetic consultation has some benefit (and even this is inconsistent). This is hardly surprising, but there is really not the slightest reason to suppose that the benefit, such as it is, has anything whatsoever to do with homeopathy.

Homeopathy, non-specific effects and good medicine is the title of an excellent editorial, in the same issue of Rheumatology, by Edzard Ernst. He points out that "The recognition of the therapeutic value of an empathetic consultation is by no means a new insight". Any therapy that provides only non-specific effects is unacceptable. Any good doctor provides that and, when it exists, real effective treatments too.

Lewith’s private clinic

The Centre for Complementary and Integrated Medicine is run by Drs Nick Avery and George Lewith. It is always a bit galling to real scientists, who often work 60 hours a week or more to get results, that people like Lewith get a professorial salary (in his case from the University of Southampton) but still have time to make more money by doing another job at the same time.

Avery is a homeopath. I wonder whether we can now look forward to the web site being changed in the near future so that there is a clear statement that the pills have no effect?

There is, at least, now no longer any mention of the Vega test on Lewith’s site. That is a test for food allergy that has been shown again and again to be fraudulent. The Environmental medicine page is brief, and avoids making any claims at all. It now contains the somewhat coy statement

“Specific food avoidance regimes are a controversial area and one in which there may be conflict between conventionally trained allergists and CAM practitioners.”

However the page about fibromyalgia still mentions homeopathy favourably. And it still fails to refer to my reanalysis of one of the positive trials which revealed a simple statistical mistake.

The front page of their web site boasts that "Dr George Lewith is now one of The Lifestyle 50!". " The Times, in an article on September 6th 2008, included George Lewith in The Lifestyle 50, this newspaper’s listing of the “top 50 people who influence the way we eat, exercise and think about ourselves”. Dr Lewith is included in the Alternatives category". It doesn’t mention that this is an honour he shares with such medical luminaries as Gillian ("I’m not a doctor") McKeith, Jennifer Ariston and the Pope,

But let’s end this on a happier note. There is one thing that I agree with wholeheartedly. Lewith says

"The use of bottled water seems to me to be a multi-billion pound industry, based on some of the cleverest marketing that I have ever encountered.  There is absolutely no evidence that bottled water is any safer, better, or more “energising” than the water you get from the tap." 

No connection of course with the multi-million pound industry of selling homeopathic water by clever marketing.

Some limitations of the paper by Brien et al.

Like any good trial, this one defined in advance a primary and secondary outcome.

The primary outcome was ACR20. which means the propertion of patients that showed an improvement of at least 20% of the number of tender and swollen joint counts and 20% improvement in 3 of the 5 remaining ACR core set measures (see Felsen 1995). Although it isn’t stressed in the paper, there was no detectable difference between consultation vs no consultation for this primary outcome.

The secondary outcome was 28-joint DAS (DAS-28), tender and swollen joint count, disease severity, pain, weekly patient
and physician GA and pain, and inflammatory markers (see, for example, Stucki. 1996). It was only on this outcome that an effect was seen between consultation and no consultation. The "effect size" (standardized mean score differences, adjusted for baseline differences) was an improvement of 0.7 in DAS-28 score, which runs on a scale from 0 – 10. Although this improvement is probably real (statistically significant), it is barely bigger than improvement of 0.6 which is said to be the smallest change that is clinically significant (Stucki. 1996).

Not only is the improvement by the consultation small in clinical terms. It is also rather inconsistent. for example Table 6 shows that the consultation seemded to result in a detectable effect on swollen joint count, but not on tender joint count. Neither was there any significant effect of the consultation on the response to “considering all the ways your arthritis affects you, please make a vertical line to show how well you are now”. There appeared to be an improvement on “negative mood score”, but not on “positive mood score”. Effects of the consultation on pain scores was marginal at best.

It seems to me that the conclusion that the consultation process helps patients, though not entirely implausible, gets marginal support from this study. It may be real, but if so it isn’t a large effect.

Like most alternative medicine advocates, the authors of this paper make the mistake of confusing caring and curing. Caring is good if there is nothing else that can be done (as is only too often the case). But what patients really want is cures and they’ll never get that from an empathetic consultation.

The problem of Human Resources

What does all this mean for alternative medicine on the NHS? Nobody denies the desirability of empathy. In fact it is now talked about so much that there is a danger that scientific medical education will be marginalised. My own experience of the NHS is that most doctors are quite good at empathy, without any need to resort to hocus pocus like homeopathy and all the myriad forms of mythical medicine.

It must be said that Drs Avery and Lewith have had proper medical training. Their views on alternative medicine seem bizarre to me, but at least they should do no great harm. Sadly, the same can’t be said for the majority of homeopaths who have no medical training and who continue to andanger the public by recommending sugar pills for anything from malaria to Dengue fever. People like that have no place in the NHS. Indeed some are in jail.

Not long ago, I was invited to tour the oncology wards at UCL hospital with their chief spiritual healer, Angie Buxton-King. Although in her private practice she offers some pretty bizarre services like healing humans and animals at a distance, I had the impression that on the wards she did a pretty good job holding hands with people who were nervous about injections and chatting to people in for their third lot of chemotherapy. I asked if she would object to being called a "supportive health care worker" rather than a spiritual healer. Somewhat reluctantly she said that she wouldn’t mind that. But it can’t be done because of the absurd box-ticking mentality of HR departments. There is no job-description for someone who holds hands with patients, and no formal qualifications. On the other hand, if you are sufficiently brainless, you can tick a box for a healer. Once again I wish that HR departments would not hinder academic integrity.

Follow-up

Steven Novella, at Science-Based medicine, has also written about this paper.