Daniel Ryan’s paper1 arguing that acupuncturists are breaching the medicines act is implicitly based on a number of assumptions. In arguing that acupuncture therapy is not based on evidence he implies that medical practice is based on evidence. This is a very simplistic understanding of a complex issue. In fact very little of what I as a GP do is based on high-level evidence. For example, in an evaluation of 55 American Heart Association Guidelines on cardiovascular practice only a median of 11% of the recommendations were based on level A evidence.2 In addition, as a GP around 20% of the problems that I see do not have a clear diagnosis.3 This means that there is no chance of having evidence on how to manage the problem. Even when I make a diagnosis it is usually uncertain (until the patient is better) a viral respiratory tract infection presents the same way as an early case of meningitis. The beliefs of my patients vary considerably. Many of them believe that if they are unwell then their doctor should give them something to treat them. Some of them believe that acupuncture is effective.
For complex problems with either no diagnosis or an unclear diagnosis my most effective strategy is to provide care (listening, understanding, reassurance of absence of serious illness) and try to invoke a positive placebo response. This may be achieved by ‘doctor as placebo’; my reassurance. It may be achieved by prescribing something that I think might help, which in the current funding environment is likely to be a pharmaceutical because they are funded. A simple example is that I would prescribe paracetamol or ibuprofen for someone with back pain, even though the evidence is that they are no more effective than placebo.4 I would say that this helps some people, and ensure that they were unlikely to suffer from adverse effects. If they thought that acupuncture would help I would encourage them to try that. In both instances they would recover faster than if I declined to provide any treatment, either because of an active effect that “the evidence” has not delineated, or as a result of a placebo response.
If Daniel Ryan wanted to look at an area of healthcare that is not based on high-level evidence and costs a lot of money he should look at surgery. As Harris has documented, it is arguably “the ultimate placebo”.5 The recent banning of the use of surgical mesh for incontinence and prolapse highlights how on occasion surgeons introduce procedures (and implants) without rigorous evaluation. One of the problems of course is that like acupuncture, surgery is operator dependent, it is really hard to do “placebo” surgery (or place placebo needles) so like acupuncturists, surgeons make judgements based on their knowledge and experience as to what would be the best treatment. If the only “evidence” considered is a randomised control trial, then neither surgery nor acupuncture will fare very well. A lack of evidence means a lack of evidence; it is not proof that a treatment does not work. There is no “evidence” on the efficacy of parachutes.6 All the surgical patients will be harmed (anaesthetic risk, scarring, risk of infection). Very few of the acupuncture patients will be harmed.
Ryan suggests that there is a problem with acupuncture publication bias based on a 20-year-old reference. We know there is publication bias in pharmaceutical research.7 This problem is not limited to acupuncture.
There is no question that some of what I do is based on high-level evidence and I would not be happy if my patient wanted to treat their scurvy with acupuncture. However, an approach of presuming acupuncture is good for nothing is not justified. It is lazy to condemn treatments that are not part of your own culture by arguing that there is no “evidence”. That is one factor to consider, but we also need to consider what harm the treatment causes, the extent to which the treatment enhances a placebo response, and issues of exploitation, missing out on effective treatment and the ethical behaviour of practitioners delivering the treatment.8 It would be interesting to study how many of the procedures that ACC funds surgeons to perform are supported by high-level evidence and take that into account when interpreting Ryan’s findings.