In the paper, the authors distinguish between several meanings of "homeopathy". Does it refer to the medicines? To consultations with a homeopath? To the principles of homeopathy? Or to the homeopathic therapeutic system as a whole? To be honest, I don't really care. But then the authors look at the definitions of 'efficacy' and 'effectiveness':
In clinical epidemiology ‘efficacy’ refers to ‘the extent to which a specific intervention, procedure, drug, regimen, or service produces a beneficial result under ideal conditions’. Evidence of efficacy is usually interpreted as requiring evidence from placebo-controlled randomised controlled trials (RCTs). Whereas ‘effectiveness’ is a ‘measure of the extent to which a specific intervention, procedure, drug, regimen or service, when deployed in the field in routine circumstances does what it is intended to do for a specific population’. Evidence of effectiveness requires pragmatic RCTs and well conducted observational studies of routine clinical practice.
The authors then argue that evidence of cost effectiveness, based on the clinical effectiveness of the intervention is required by NICE for interventions that are provided by the NHS, not evidence of efficacy. Then they go on to look at the evidence.
The authors divide the evidence into experimental evidence, by which they mean randomised, double-blind, placebo-controlled trials. This is where things go off the rails a bit. The authors suggest that looking at the 'active ingredient' of homeopathy as being the medicine given is far too reductionist; perhaps the active ingredient should be thought of as including "Six putative active ingredients... : patient's openness to the mind body connection, consultational empathy, in depth enquiry into bodily complaints, disclosure, the remedy matching process, homeopathic remedies". Not only that, but "Other authors have discussed the difficulties of separating out the effects of the homeopathic medicine from the consultation effects".
Now, this is interesting. It seems that what the authors are saying here is that much of the benefit that patients get from homeopathy is unrelated to the medicine they are given; it is related to the relationship between the homeopath and the patient, and to the consultation process. In medical science, there's a term for this: it's called the placebo effect. That's why you conduct placebo-controlled trials; because those effects exist, and you therefore need to separate them from the effect of the actual medicine that is given to the patients. That placebo-controlled trials are specifically designed to address this problem seems to have passed the authors by. Placebo-controlled trials of individualised homeopathy (i.e. the whole homeopathic package, including consulation) have been carried out, and show no better results than trials that did not include the consultation. But at least there is a recognition here that non-specific effects play a very large role in homeopathy.
The authors go on to look at what they call 'observational evidence', consisting of observational studies and case reports. The authors admit that this sort of evidence is "prone to many types of bias, it is regarded as weaker than experimental evidence, and is generally disregarded in systematic reviews of evidence". But they say "However, this type of evidence does acknowledge the homeopath i.e., focuses on treatment by a homeopath in contrast to the experimental evidence which focuses only on the homeopathic medicine".
This seems to be just the usual special pleading that 'my woo can't be tested by your cold-hearted science'. Then again, the authors do seem to be suggesting that the pills containing nothing are not as important as the context in which those pills are delivered. So where next? Unfortunately, the authors miss this opportunity by simply making some weak recommendations about terminology. But if the homeopathic community could admit that their drugs don't work, and sit down to figure out how they could strip out all the pseudo-science about energy medicine, water memory, quantum entanglement and so on, and how they could harness the power of the placebo effect, then what remained could conceivably be of some use.
Of course, that the homeopaths haven't managed to figure this out for themselves yet is a little surprising, given that this is what Ben Goldacre has been arguing for some time. It's also interesting to reproduce the last paragraph of the famous Shang et al. meta-analysis which was so damning of homeopathy:
"We emphasise that our study, and the trials we examined, exclusively addressed the narrow question of whether homoeopathic remedies have specific effects. Context effects can influence the effects of interventions, and the relationship between patient and carer might be an important pathway mediating such effects. Practitioners of homoeopathy can form powerful alliances with their patients, because patients and carers commonly share strong beliefs about the treatment’s effectiveness, and other cultural beliefs, which might be both empowering and restorative. For some people, therefore, homoeopathy could be another tool that complements conventional medicine, whereas others might see it as purposeful and antiscientific deception of patients, which has no place in modern health care. Clearly, rather than doing further placebo-controlled trials of homoeopathy, future research efforts should focus on the nature of context effects and on the place of homoeopathy in health-care systems."
Could the homeopathic and conventional medicine communities be moving towards common ground?