involves bodywork (a more complex procedure than the simple taking of a tablet) the larger the trial, the greater the variations between the procedures and therapeutic abilities of the practitioners. To return to the question of size of study and expense of funding, the current situation in the West means that most future research into Shiatsu will have to be done by the Shiatsu community itself, and we will have to educate ourselves in research methodology accordingly. Research can only prove the effectiveness of Shiatsu if it is done according to recognised procedures which allow comparisons with studies of other therapies already approved by the NHS or a similar body. The energy in the Shiatsu community is positive and powerful, and this Review gives us all much information about carrying out research. Nonetheless, there are certain challenges which we will be called upon to encounter along the way, and which can only increase our understanding of our therapy. The Placebo Effect While any double-blind controlled trial of pharmaceuticals involves the administration of a dummy drug in such a way that neither researcher nor subject knows it is a dummy, practitioners of any form of ‘energy medicine’ know that it cannot be applied in a ‘sham’ way as if administering a tablet, so that we can see that in some of the studies ‘sham’ acupressure works nearly as well as the “real” acupressure. The aim of a double-blind controlled trial is to measure the effectiveness of the drug or treatment method against placebo, but to some extent the practice of any form of contact therapy is arguably placebo. The simplest of touches can have measurable effects. A study showed that when a nurse laid a hand on patients awaiting surgery while they read a pre-operative pamphlet it had measurable effects on their blood pressure (Whitcher and Fisher, 1979). Although the study was designed to show the difference between the ways in which men and women reacted to the touch, we can see the powerful effect of simple bodily contact, and the ideal of ‘objectivity’ is immediately compromised. Given that in a recent trial to evaluate the effectiveness of the herb, St. John's wort, as compared with an anti-depressant and placebo it was the placebo that produced the best results, it is clear that the placebo effect is a powerful healing agent (Shelton et al., 2001). I am reminded of a cartoon I saw many years ago of two old ladies in a doctor's waiting room, discussing their ailments – one is saying to the other, ‘Have you tried placebos? I hear they're marvellous’. In fact, as we know that we work with the body's self-healing power, we should perhaps be proud to acknowledge the placebo component of Shiatsu rather than struggling to disassociate ourselves from it. However, it is clear that the much-vaunted ‘double-blind controlled clinical trial’ results can not be achieved for Shiatsu, no matter how rigorous the research procedures. Differences between Shiatsu and Acupressure The majority of the research studies examined in this review have been conducted on the use of acupressure rather than Shiatsu – in other words, the treatment under consideration has been the use of particular points selected for a certain health condition and applied according to certain guidelines. In many of these studies the health condition for which the points were used was a category defined by orthodox Western medicine, such as asthma, bronchiectasis or hemiplegia, rather than the patterns of disharmony for which the points might have been prescribed in context (e.g. Damp Phlegm in the Lungs or Kidneys Failing to receive Qi). This is poor patient sampling procedure from the TCM point of view, although it might make any results more convincing to western eyes scrutinising the results. It is obvious, however, that this is a much easier form of treatment to study than the more fluid form of, for example, Zen Shiatsu, in which the session is tailored to a Hara diagnosis which varies from receiver to receiver, and can vary from session to session, although some studies use ‘individualised acupressure’ as a category of treatment, which is likely to be more variable. Acupressure is more likely to yield quantitative data, as its methods can be more strictly controlled. Acupressure, and other forms of Shiatsu which use more long-term forms of diagnosis, is therefore better adapted for use in 1 specific conditions, although the receiver may be less likely to experience it as qualitatively lifeenhancing than the more fluid, less results-focused Shiatsu styles. We are fortunate that our therapeutic discipline contains both possibilities. So, while the life-enhancing, non-prescriptive styles of shiatsu can lead us to engage further with questions such as ‘what is health?’, it seems that if we want quantitative research results we would do well to incorporate a few symptomatic points in our sessions, and also teach them to our receivers to use in between treatments (a valuable extra research resource). We do not need a profound knowledge of point functions, as few points have been exhaustively researched and many have multiple functions. This would be a positive way of utilising Shiatsu's connection with the rich tradition of East Asian medicine, and of producing quantifiable symptomatic effectiveness within a nurturing and intuitive session. Research Criteria The