Chinese medical acupuncture and medical research

A good overview of the research available to date on a wide range of health conditions can be found in these pages on the British Acupuncture Council website:

The information below is a detailed examination of the medical research landscape over the last 100 years, with some reflections on how this may have impacted the body of evidence available related to acupuncture treatment.

Introduction: Randomised controlled trials and Chinese medicine

Evidence based medicine (EBM) is a well-known, common parlance term in our world today. As a concept it has no-brainer quality to it.  There is a widespread understanding within society in general, not just within medical professions, that there should be some testing or verification of any medical intervention, be it within conventional biomedicine or the traditional and holistic medicines globally. The key issues are that they are in accordance with the core medical ethics tenet of “Do no harm” - be safe, as well as there being some guide as to efficacy in treating the health issue in hand. This is all indisputable.

However, in relation to medical interventions and medical research, the phrases ‘there is evidence for..’ or ‘there is no evidence for..’ are misleadingly simple, black and white statements, which completely obscure the complexity of the issues involved.

Randomised controlled trials (RCTs) became established, in the 2nd half of the 20th century, as the ‘gold standard’ of medical research and therefore EBM for all types of medicine (see below for the history of this). Where there is a gold standard, inevitably everything else becomes sub-standard - lesser - rather than just different, potentially offering different use and value.

It has long since been suggested that assessing the validity or effectiveness of Chinese medical acupuncture through randomised controlled trials and systematic reviews of those trials has a raft of inherent problems. They are born out of 2 fundamentally different paradigms or cultures, that do not naturally align.

To put it another way:

It’s like testing the skill, quality and efficiency of a fish’s movement by assessing how well it can run.

We are now, just emerging, into an era where the implications of RCTs as gold standard, and whether the resulting hierarchy of evidence is rational, prudent, or serving patients, is being re-evaluated.

The information below attempts to pick apart some of these issues, looks at the history of EBM, and aims to give some thoughts as to where this leaves patients.

The history of evidence-based medicine

Randomised Controlled trials (RCTs), as a form of testing medicine, began in the 1940s and took off with the post-war pharmaceutical revolution (Jones and Podolsky, 2015). It was made compulsory for pharmaceutical drug testing by the US Food and Drug Administration (FDA) in 1962. Prior to this, there was concern that drugs were entering the marketplace based on marketing drives, without any FDA control over safety or effectiveness (Fultinaviciute and Maragkou, 2023). The ‘Gold Standard’ designation was later used by Alvan Feinstein and Ralph Horwitz in a paper published in The New England Journal of Medicine (NEJM) in December 1982 (Jones and Podolsky, 2015).

In the UK, Archibald Cochrane (1909-1988) was a 20th century pioneer in the fields of epidemiology, public health and, what came to be known as, evidence-based medicine (Stavrou, Challoumas and Dimitrikakis, 2014). Conceivably he had a personal attachment to the model of randomised controlled trials. Whilst he was a prisoner of war in 1941, he had, out of duress and through his own logic and determination, self-designed a test which had the fundamental shape of an RCT (see the fascinating story: Stavrou, Challoumas and Dimitrikakis, 2014, p.2). This was at a time when such trials were virtually unheard of, or certainly in their infancy.

Much later in his successful and remarkable career and born out of various life experiences which led him to be absorbed by the issue of medical evidence, Cochrane wrote ‘Effectiveness and Efficiency: Random Reflections on Health Services’ in 1972.

This book became a seminal work, enthusiastically received by the medical community internationally. Cochrane advocated that all types of treatment should be evaluated via RCTs. The impact of this text on the medical community of clinicians and epidemiologists at the time greatly contributed to the acceptance of RCTs as the gold standard of EBM (Stavrou, Challoumas and Dimitrikakis, 2014). Essentially, RCTs became broadly synonymous with the concept of EBM itself and it could be said that this thinking has seeped into the wider public consciousness.

Cochrane went on to advocate for a summary of all RCTs. Over time this resulted in the development of systematic reviews providing meta-analyses (combining and analysing the data from multiple comparable trials) in specific disease + treatment areas (Stavrou, Challoumas and Dimitrikakis, 2014). Eventually this work was formalised further within the Cochrane Collaboration – an organisation which has a unique place of authority in medical research to this day (see below re. NICE).

Thus, Cochrane’s pioneering work in the UK, together with the FDA’s commitment to such research models for the pharmaceutical industry in the US, conceivably have led us to the place we are in now, whereby sufficient favourable evidence from RCTs and associated systematic reviews becomes equivalent to ‘There is evidence for….’, and vice versa.

Evidence from other types of research or from clinical experience is generally framed either as ‘there is no scientific evidence for…’ or ‘there is unreliable data which suggests…’. This does not inspire confidence in patients, but arguably is misleading and unhelpful to the cause of better national and international health.

RCTs: what they are and implications for evaluating acupuncture

To break this down further for the general reader, here is some further explanation about randomised controlled double- or single-blinded trials, highlighting some of the issues involved:

• RANDOMISED: this refers to a sample number of people involved in the trial being arbitrarily or randomly assigned into at least 2 groups (active and control). It raises issues about equivalence between the 2 groups in terms of all kinds of issues including: the phenotype and severity of their condition; the demographics of the group sample – age, gender, ethnicity, socio-economic grouping, marital status etc. These are issues which affect RCTs related to all medical interventions. There are comprehensive guidelines related to these and many other issues of methodological quality, (CONSORT, STRICTA, CASP) so in good quality trials they should be well managed.

  There are also statistical tools for ‘adjusting’ for certain variables around the patient-participants, their demographics and their lives. However inevitably the impact of intangible/unquantifiable factors such as: the impact of systemic inequality on health; support systems, diet and other lifestyle factors; psycho-emotional health etc. can never be perfectly accounted or adjusted for. This is potentially of particular significance in relation to acupuncture as it is a subtle form of medicine that both diagnoses and treats a person according to their health on every level. This is one of the smaller issues though.

• CONTROLLED: this refers to the fact that RCTs always have at least 1 active group receiving an active medical intervention, and 1 control group, receiving a medical intervention that is, or is meant to be, entirely inactive.

  This is one of the biggest issues in relation to evaluating acupuncture through RCTs. This model fits perfectly in the context of pharmaceutical drug trials: both the active group and the control group can be given a tablet that is identical in terms of size, colour, shape etc. Thus, there is no differentiating information to skew the perceptions or expectations of the participants, giving a genuinely clear picture of the treatment effect on the active group.

  However, in a modality such as acupuncture, that involves a practitioner-administered physical, skin-piercing procedure, the question of what to use as a control group treatment has been endlessly debated and imperfectly resolved.

  Commonly acupuncture RCTs have used a form of sham acupuncture as a control group treatment of which there are several types including: inserting a needle into a non-acupuncture point, or an acupuncture point not indicated for the health condition being treated; using a retractable device, intended to give an impression of a needle being inserted but without penetrating the skin; and using a non-penetrating pressure tool on the same or different points as the treatment group protocol (Birch et al., 2022b; Lundeberg et al., 2008; Vickers, 2002).

  This is a highly complex area of academic discussion. Broadly, the discussion relates to: both the physiology and psychology of the various sham acupuncture procedures (Birch, 2006; Langevin, 2011), and the fine detail of Chinese medical acupuncture techniques (activation of ‘de qi’ through stimulation, use of non-invasive pressure tools, strong stimulation or superficial insertion as intended therapeutic techniques etc.) (Birch et al., 2022a; Birch et al., 2022b).

  However, it is straightforward to see that this is a much more complex and problematic situation of accurately assessing treatment effect vs. placebo effect than in a drug trial administering visually identical tablets.

• DOUBLE- AND SINGLE-BLINDING: this refers to either both the participants and those administering the medical intervention being unaware of active and control group assignment (double-blinding); or just the participants being unaware of whether they are receiving an active or inactive/sham treatment (single-blinding). Traditional double-blinding is of course impossible with acupuncture trials. However, both the participants and the researchers who write up the data can double-blinded, even though the treatment-deliverer can’t be – a factor that may have significance.

Other challenges presented by the RCT model:

• DISEASE CATEGORIES / criteria are specified in biomedicine through specific consensus criteria, biomarkers or other measurements. Therefore, trials can be designed with samples that have a clear cohesion in terms of patient profile and standardised treatment. From the Chinese medicine framework however, within any disease category, there will be multiple Chinese medicine patterns. Therefore, the suitable treatment that would be effective would vary according to the specific pattern.

  For example, a sample of patients diagnosed with moderate asthma in biomedical terms might be diagnosed as: lung yin deficient; liver and lung heat excess; kidney yang deficient; lung qi and yin deficient; or lung and kidney yin deficient in Chinese medical terms. The suitable, and effective treatments for each of these would have significant differences in terms of point selection.

  For researchers designing a trial around moderate asthma + acupuncture treatment efficacy this presents a challenge. Either patients are given a standard acupuncture protocol which will be inherently suitable for some and less so, or not so for others, thus distorting the results on efficacy due to inherent unsuitability. Or they are each given a tailored treatment, but this is inconsistent with the RCT methodological approach of assessing a standard ‘dose’ treatment for a specific disease.

  Theoretically participants could be selected according to Chinese medical pattern and administered a standard treatment for that pattern profile e.g. lung and kidney yin deficient patients. However, given the challenges faced by all RCTs in recruiting and retaining adequate numbers of participants for a meaningful trial sample, this is probably not feasible.

  There is one solution that researchers settle on which seems to bridge this divide between the inherent treatment technique of Chinese medical acupuncture and the requirements of standardisation present by RCT methodological approach. That is to define a specific number of points which all participants receive and then add on, for example 3 additional points tailored for each patient profile. An example of this is the asthma study done by Choi et al, a trial of good methodological quality overall as assessed by STRICTA criteria (REFs).

  However, this once again is an imperfect compromise – further highlighting the inherent paradigm clash between acupuncture treatment and RCT evaluation.

• PHYSICIAN EFFECT: In a drug trial the active and control tablets can be administered in a very impersonal manner, with minimal interpersonal interaction. However, in an acupuncture trial a physically invasive procedure is administered to various parts of the patient’s body. Inevitably there is more of an impact for many trial participants as to whether they like/dislike, are comfortable/uncomfortable etc. with that person. There is an impact of this on the patient’s experience and conceivably on the treatment effect. This may go unquantified or accounted for in the trial results data (Franks et al., 2006; Prady et al., 2013).

Systematic Reviews

As referred to above, the pioneering work of Archie Cochrane led to the summarising or evaluation of RCTs within systematic reviews. These are the meta-analysis of several comparable RCTs, effectively analysing the results from a much larger data set (for further description please see University of Bath, 2025). They have become a critical part in the EBM process.

They require that enough trials have been conducted on the same condition + same therapeutic intervention of sufficient methodological quality, adhering to CONSORT and/or STRICTA frameworks, that can be compared. Where there is not sufficient trial data of good standard to be compared, this is described as ‘no evidence for…’

There is an important issue here in terms of economics, location and therefore planning of RCTs in such a way that systematic reviews become possible.

Pharmaceutical drug trials are undertaken in-house by the companies themselves. The global pharmaceutical industry had a revenue of 1.6 trillion dollars in 2023 (Mikulic, 2024). It is estimated to have spent USD 1.98 billion on research and development in 2020 (IFPMA, 2022). RCTs are lengthy and costly processes. The drug companies conduct phase after phase of trials in a coherent way, in order to build the bank of data, be it favourable or unfavourable, in relation to any particular drug which they may seek approval for.

Acupuncture research in contrast is generally undertaken by universities, or occasionally by hospitals or other research bodies. It is often funded by the universities or by charitable foundations. It may often be led by the research interests of the academics or physicians involved. Therefore, there is not the coherence or planning internationally to strategically build bodies of comparable data in specific disease category areas (+ treatment protocols + demographics etc.), to allow for systematic reviews to be done at all.

The Cochrane Collaboration does a further level of analysis. It assesses the systematic reviews on a specific condition + treatment area and produces what’s known as a Cochrane review.

To put this into context:

The National Institute of Clinical Excellence (NICE) in the UK bases its assessment of the conditions for which acupuncture is effective on favourable Cochrane reviews. Historically it has therefore concluded that acupuncture can only be deemed effective for treating chronic pain and headaches. This is because these were the only conditions which have enough RCT data, that is comparable, and is of sufficient methodological quality, allowing for systematic reviews and Cochrane reviews – all indicating favourable results.

Other types of research and evidence

There are many types of medical research models and it is beyond the scope of this page to give a comprehensive overview.

Some examples which may be applicable and useful in assessing the value and efficacy of acupuncture treatment include:

• Comparative Effectiveness Trials (CET): these studies measure the effectiveness of one type of active treatment relative to or over and above another. For example, corticosteroid medication to treat Bell’s Palsy for 1 group, and corticosteroids + acupuncture treatment for another.

• Pragmatic Clinical Trials (PCT): these are designed to test whether a treatment works in real life, maximising applicability and generalizability. In some examples they overlap with CETs in design (Patsopoulos, 2011).

• Langevin et al. advocates both a top-down approach, which aims to understand a whole system on its own terms, and also a bottom-up approach which examines the impact of specific components or actions (Langevin et al., 2011)

In addition, the NICE strategy 2021-2026, lays out a new framework for “Real-world evidence”. This is defined as a new rigorous but also achievable framework to collect evidence addressing ethical, practical and technical challenges in various health and treatments areas. They advocate for research that is transparent and has integrity, but within research designs that are more flexible than traditional approaches (NICE, 2025).

One final note on this: whilst case study results have traditionally been dismissed as incidental or anecdotal evidence, the enduring nature of a system of medicine does imply some relevance, value or effectiveness.  

2000+ years of Chinese medical clinical practice arguably implies a value in its treatment.

Where this leaves patients

Cochrane’s motivation and the core purposes of EBM related to 2 core aspects:

• Safety

• Efficacy

One approach for patients in the current status of EBM, which has many gaps in it in terms of assessing the efficacy of acupuncture, could be to separate these 2 issues out and address them differently.

The Professional Standards Authority recognises one Professional membership association as holding their members to the highest standards in terms of training and ongoing clinical practice. The best assurance for receiving safe acupuncture treatment therefore is to select a practitioner who is a British Acupuncture Council (BAcC) member.

In terms of efficacy, the BAcC research pages (link at top of this page), gives as thorough an overview of the research data to date as you are likely to find.

Asking practitioners in this field as to their clinical experience in relation to particular conditions is also valid, even though it requires some trust in their ethics and honesty.

Conclusion

• RCTs were developed because of the need to test pharmaceutical medicines and designed for this purpose.

• As a form of testing that focuses on standardisation and control, they are appropriate to test medicines of fixed substance and dosage, delivered in tablet form so that both the active and control group patients have exactly the same experience of receiving the ‘treatment’.

• This type of testing is also appropriate for medicine that carries a high degree of risk in terms of side effects.

• RCTs became established as the gold standard of evidence-based medicine in the 20th century.

• As the gold standard of EBM, other types of medical intervention have been forced to fit into the RCT model, even when there is widespread questioning of the suitability or efficacy of this.

• In relation to acupuncture, questions have been widely raised regarding suitable control treatment and the efficacy or obscuration caused by sham acupuncture issues. This also calls into question whether it is possible for acupuncture RCTs to be blinded.

• Furthermore, there are questions raised related to: standardised treatment protocols being fundamentally against the inherent approach; and the ‘physician effect’ in distorting results, amongst many other issues.

Going forward:

• We are moving into a new era of medical research.

• There are calls being made for the gold standard label of RCTs to be removed, and for medical interventions to be assessed in more sensible, tailored ways, that can truly assess the value of that therapy (Fultinaviciute and Maragkou, 2023).

• NICE has developed frameworks for ‘real-world evidence’ in order to address gaps in evidence. This represents an important culture shift.

The development of EBM was intended to ensure safety for patients and some reliable basis of information that ethically guides them in their choices about the type of treatment for their condition.

The quality of assessment does need to be high. Transparency and integrity are important and possible within different research approaches (NICE, 2022). However, it could be said that the gold standard label of RCTs has distorted the research picture. Arguably, we need to move into an era where other research models, more suited to other medical interventions, are also accepted as valid on their own terms.

Reference List

Birch, S. (2006). A Review and Analysis of Placebo Treatments, Placebo effects, and Placebo Controls in Trials of Medical procedures When Sham Is Not Inert. The Journal of Alternative and Complementary Medicine, 12 (3), 303-310. Available from DOI:10.1089/acm.2006.12.303. [Accessed 20.01.2025].

Birch, S. et al. (2022a). Historical perspectives on using sham acupuncture in acupuncture clinical trials. Integrative Medicine Research, 11 (1), 100725. Available from DOI: 10.1016/j.imr.2021.100725 [Accessed 28.03.2022].

Birch, S. et al. (2022b). On defining acupuncture and its techniques: A commentary on the problem of sham. Integrative Medicine Research, 11 (2), 100834. Available from https://doi.org/10.1016/j.imr.2022.100834 [Accessed 04.03.2022].

Critical Appraisal Skills Programme. (2021). CASP Randomised Controlled Trials Checklist. Casp-uk.net. Available from https://casp-uk.b-cdn.net/wp-content/uploads/2020/10/CASP_RCT_Checklist_PDF_Fillable_Form.pdf [Accessed 12.04.2022].

Franks et al. (2006). Studying physician effects on patient outcomes: Physician interactional style and performance on quality of care indicators. Social Science and Medicine, 62 (2), 422-432. Available fromhttps://doi.org/10.1016/j.socscimed.2005.05.027 [Accessed 20.01.2025].

Fultinaviciute, U. and Maragkou, I. (2023). Should RCTs remain the gold standard? Clinical Trials Arena. Available from: https://www.clinicaltrialsarena.com/features/rct-gold-standard/#?cf-view&cf-closed [Accessed 20.01.2025].

IFPMA coordinated by Deplano, L. and Cintra, G. (2022). The Pharmaceutical Industry and Global Health: Facts and Figures. The International Federation of Pharmaceutical Manufacturers and Associations. Available from:

https://www.ifpma.org/wp-content/uploads/2023/01/i2023_IFPMA-Facts-And-Figures-2022.pdf [Accessed 20.01.2025].

Jones, D. and Podolsky, S. (2015). The history and fate of the gold standard. The Lancet. Vol. 385 (9977), 1502-1503. Available from: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(15)60742-5/fulltext#au20  [Accessed 15/01/2025].

Langevin et al., (2011). Paradoxes in Acupuncture Research: Strategies for Moving Forward. Evidence-Based Complementary and Alternative Medicine, Art. ID180805, 11 pages. Available from DOI:10.1155/2011/180805 [Accessed 04.03.2022].

Lund, I. and Lundeberg, T. (2006). Are minimal, superficial or sham acupuncture procedures acceptable as inert placebo controls? Acupuncture in Medicine, 24 (1), 13-15. Available from DOI: 10.1136/aim.24.1.13[Accessed 04.03.2022].

Lundeberg et al. (2008). The Emperor’s sham - wrong assumption that sham needling is sham. Acupuncture in Medicine, 26 (4), 239-242. Available from doi:10.1136/aim.26.4.239

[Accessed 04.03.2022].

MacPherson, H. et al. (2010). Revised STandards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA): Extending the CONSORT Statement. PLoS Med, 7 (6), e1000261. Available from https://doi.org/10.1371/journal.pmed.1000261 [Accessed 04.01.2022].

MacPherson, H., Peters, D. and Zollman, C. (2009). Closing the evidence gap in integrative medicine. British Medical Journal, 339 (7724), 763-764. Available from doi:10.1136/bmj.b3335 [Accessed 28.03.2022].

Mikulic, M. (2024). Global Pharmaceutical Industry – statistics and facts. Statista.com. Available from https://www.statista.com/topics/1764/global-pharmaceutical-industry/#topicOverview [Accessed 15/01/2025].

NICE. (2022). NICE Real-world evidence framework. Nice.org.uk. Available from https://www.nice.org.uk/corporate/ecd9/chapter/overview. [Accessed 20/01/2025].

Patsopoulos, N. (2011). A pragmatic view on pragmatic trials. Dialogues in Clinical Neuroscience, 13 (2), 217-224. Available from https://www.researchgate.net/publication/51572423_A_pragmatic_view_on_pragmatic_trials [Accessed 28.03.2022].          

Prady, S. et al. (2013). Controlling practitioner – patient relationships in acupuncture trials: a systematic review and meta-regression. Acupuncture in medicine: Journal of the British Medical Acupuncture Society, 31 (2), 162-171. Available from https://doi-org.uow.idm.oclc.org/10.1136/acupmed-2012-010286 [Accessed 28.03.2022].

Schulz, K. et al. (2010). CONSORT 2010 Statement: updated guidelines for reporting parallel-group randomized trials. BMJ, 2010, 340. Available from:

doi: https://doi.org/10.1136/bmj.c332 [Accessed 15/01/2025].

Stavrou, A., Challoumas, D., and Dimitrikakis, G. (2014). Archibald Cochrane (1909-1988): the father of evidence-based medicine. Interdisciplinary Cardiovascular and Thoracic Surgery. Vol18 (1), 121-124. Available from: doi: 10.1093/icvts/ivt451 [Accessed 15/01/2025].

University of Bath Library. (2025). Systematic Reviews: Introduction. Library.bath.ac.uk. Available from: https://library.bath.ac.uk/systematic-reviews#:~:text=What%20is%20a%20systematic%20review,and%20social%20policy%20are%20covered. [Accessed 20/01/2025].