Acupuncture and Science

Your acupuncturist is not your primary healthcare provider.  Nevertheless, anyone who provides treatment to patients for symptoms such as pain, nausea, sinus congestion, etc, needs to be confident that the treatments they are offering rest on a sound base. From the patient’s point of view, the most important test will be the personal one.  Does it work FOR THEM. However, a scientist will work towards building up an evidence base, from which [known] personal bias has been removed, to provide a base of evidence to indicate that a treatment is of general efficacy and general safety.  Such an evidence basis can never be 100% reliable in predicting how an individual person will respond to an individual medicine or therapy, therefore the statistical approach used in many scientific studies are most useful as a guide for policy makers helping to expertly allocate public budgets to achieve the greatest possible gains in the overall health of the population.

In medicine, the most important consideration is that the treatment is safe – bearing in mind the medical ethical dictum, “first, do no harm.” On this score, acupuncture has proven to be very safe. Adverse events are extremely rare, and those that do occur are usually mild and self-limiting, for example bruising. A second line of evidence attesting to the safety of acupuncture treatments comes from prices set by insurance actuaries (people who specialise in the study of successful financial claims and payouts), who rate the overall insurable practice risk for an acupuncture clinic as being many orders of magnitude smaller than that of a standard medical practice.

Even so, proper training for acupuncturists is essential, as the absolute prevention of the most serious risks, such as cross-infection or pneumothorax (a puncture of the chest wall which can lead to a collapsed lung), is an essential part of every properly trained practitioner’s skillset.

The second important consideration concerns the efficacy of a treatment. No one deserves to have their time wasted, especially if they are suffering.  Evidence based medicine specialises in formulating study designs which are intended both to isolate the effects of a treatment from other confounding factors and to remove bias from the outcomes, bearing in mind that in the real world a treatment is never received in research-level “isolation”.  Still, the application of research has been successful in identifying the kind of bias that can confound claims of efficacy, which can come from many sources. One of the most important is that the expectations and attitudes of both the practitioner/doctor and of the patient are known to have effects on the outcomes of treatments, even when such treatments contain no ingredients or involve no procedures that in themselves have been proven to be effective.

This is called the “placebo effect” (if positive) or the “nocebo effect” (if negative), and may in itself be one of evidence-based medicine’s strongest, if persistently unrecognised, discoveries.  The evidence based medicine framework therefore requires, and has striven to put in place procedures to ensure, that all drugs and other treatments show that their positive effect is greater than that of a placebo to a significant degree larger than chance in order to be considered to be effective.

Some minority voices among clinical researchers, as you might imagine, suggest that the placebo effect is worthy of study for its own sake, as every double-blinded random controlled clinical trial attests to its baseline effectiveness, but this is not a suggestion that seems to receive wide support.

There is a body of clinical research into the effectiveness of acupuncture.  It’s results can be confusing. Many of the early studies (and some current ones) were not well designed, and suffered from well recognised problems such as small or non-random samples, which means that nothing useful can be concluded from them, one way or another.

A further difficulty that persists, even as research design improved, concerns the question of determining what would constitute a proper placebo in acupuncture.  For research purposes, a placebo should be indistinguishable to both the patient and to the practitioner, from the “effective” treatment being tested.  This allows for “blinding”, which means that both patient and practitioner are kept “blind” during the research trial in order to prevent their personal expectations affecting outcomes.  However, acupuncture is more like surgery than medicine, in that it is difficult to see how you could blind someone who is practicing a skill rather than administering a medicine. The practitioner, in any case, surely knows whether they have performed acupuncture (or surgery) in accordance with their training, or whether they’ve faked it in some way. And, as we’ve seen above, the expectations of the practitioner/doctor can introduce bias into the results every bit as much as the expectations of the patient.

Nevertheless, some interesting attempts have been made. One method is to compare acupuncture at traditionally chosen points with acupuncture at randomly chosen non-points. Another is to compare deep needling with shallow needling. One rather ingenious attempt to come up with an acupuncture placebo required the invention of a device which hides the needle (or a dummy) inside a casing. The patient feels a puncture at the skin level, and sees the casing depressed, but cannot perceive whether the needle penetrates the skin or not.

The flaw all of these acupuncture “placebo” techniques have in common is that it has not been possible to show that any of them are inert. In fact a common complaint of practitioners trained in difference acupuncture styles, is that one person’s placebo acupuncture may be another person’s effective treatment.  And so, it is perhaps not surprising to find that while both the “effective acupuncture” arm and the “placebo acupuncture” arm frequently perform significantly better than no treatment, in many studies there is little significant difference between them.

Some people interpret this to mean that acupuncture itself is effective, but that the style or location is far less important than the fact of getting the treatment. Others interpret this to mean that all of the treatment effect of acupuncture can be explained by the placebo effect, (because any improvements are seen against a background of no treatment) and that therefore it is not an effective treatment. The debate over interpretation will rage on, but hopefully researchers with a better understanding of how to design a proper study will soon add to the body of research.  Also, perhaps, eventually, the nature of the placebo effect itself will be characterised in a useful way, and not disregarded if it turns out to be clinically useful in its own right.

It would not be inconsistent with the current body of evidence, such as it is, to make the following small claims: that acupuncture, in any style, is a significant treatment option for the relief of pain, especially in situations where painkillers are contraindicated, such as in pregnancy, or when they cause adverse gastric symptoms, which is distressingly common. Many research projects are suggestive of positive results with other symptoms and situations, including improving the effectiveness of IVF treatment, reducing the need for interventions in labour, reducing addictive cravings, reducing the number and severity of heartburn events, to mention some examples, but further research is needed.

Acupuncture is not effective for everything, or for everyone.  But for some people, in some situations, particularly pregnancy or gastric problems, which preclude the intake of certain medications, it can provide an especially welcome source of relief of pain. Many people also find it useful to clear sinus pain and congestion, for chronic functional diarrhea or constipation, for period pain, for some types of insomnia, and to reduce the effects of stress.

For the individual person, however, the question of efficacy will be answered by their own experience.  Good results will be the proof of the pudding, that can only be tested, by the eating.