Medical Acupuncture
A Journal For Physicians By Physicians

Spring / Summer 2000- Volume 12 / Number 1
"Aurum Nostrum Non Est Aurum Vulgi"

ARTICLE

TCM

Evidence-Based Medicine
And Traditional Chinese Medicine:
Not Mutually Exclusive
Lynnae Schwartz, MD

ABSTRACT
   It would appear that a dichotomy exists in the foundations and approaches of Western biomedicine and Traditional Chinese Medicine (TCM). The current emphasis in Western biomedicine is on evidence-based medicine, particularly data from randomized controlled trials. Western biomedicine and TCM need not be mutually exclusive in their practice. Single-case studies of acupuncture can be expanded into randomized controlled trials, the results of which may well support the practice of TCM/acupuncture.

KEY WORDS
   Evidence-Based Medicine, Traditional Chinese Medicine, Randomized Controlled Trials, Acupuncture

INTRODUCTION
   Physicians trained in Western biomedicine who also practice Traditional Chinese Medicine (TCM) and medical acupuncture confront a dilemma. In clinical practice, acupuncture treatment is typically highly-individualized and based on philosophical constructs, and subjective and intuitive impressions. Results may be based on physician and patient observations.
   The practice of Western biomedicine is quite different. It has evolved from reproducible scientific investigations into the relevant anatomy, physiology, and pathology of a disease process, combined with pharmacological data and objective statistical analysis of quantifiable data obtained under experimental conditions, showing measurable difference between otherwise minimally distinguishable groups. Physicians use the principles of evidence-based medicine (EBM) and may employ clinical pathways.
   An argument can be made to respect the different foundations, holding separate the standards used to determine the best practice for acupuncture and Western biomedicine. However, there exists strong pressure from the scientific and academic communities to apply the principles and methodology of EBM with statistical analysis of quantified data to evaluate alternative/complementary medical treatments, including acupuncture.1-3 In EBM, the most convincing data are obtained through large-scale, double-blind, randomized controlled trials (RCTs)4 designed to determine efficacy and disease management strategies.
   Discussion follows of the contrasting philosophical and scientific foundations of medical acupuncture and EBM, comparing standards of evidence for efficacy and potential investigative methods that may provide sufficient evidence to satisfy criteria for proving or disproving claims of efficacy under both paradigms.
Foundations and Outcome for Acupuncture
   The foundations of TCM/acupuncture are philosophical, based on the concept of Yin and Yang and a belief that human health and illness are related to the social and natural environment. Treatment decisions depend on the practitioner's interpretation of the patient's history and physical findings (Four Examinations), response to prior treatments, and diagnosis according to the Eight Principles Patterns.5 Patient experience is the most valued determinant of outcome.
   The medical literature on acupuncture includes the ancient texts,6 data from small animal experiments, and human clinical experience.
   Although information from all the above sources is taken into account when formulating a treatment plan, the patient's individual physical examination, pattern of illness, and response to prior treatment carry special significance; the patient's treatment is unique within the context of TCM. Individual presentation and clinician experience carry the highest value and are the most heavily-weighted evidence for determining specific treatment in TCM (Table 1).

   In addition to being individualized, treatment in TCM is also multimodal, i.e., recommendations are made with respect to diet, exercise, work, family, and personal relationships. Actual treatment may include acupuncture, massage, moxibustion, and/or herbal medicine. RCTs designed to measure acupuncture effect out of context from usual clinical practice are of limited use to clinicians treating patients comprehensively.
Foundations of EBM
   Recognition of the importance of objective data to guide clinical practice in Western biomedicine has led to academic and editorial endorsement of EBM.7 This approach is based on systematic analysis of objective and quantified data, "de-emphasizing intuition, unsystematic clinical experience, and pathophysiologic rationale as sufficient grounds for clinical decisionmaking."8
   Large RCTs have been generally accepted as the "gold standard" since 1948 for the unbiased evaluation of clinical efficacy for an intervention.9 Medical RCTs are designed to eliminate distinguishing characteristics among study participants and make other factors and variables equal; hence, the opportunity to measure effects of a uniformly applied intervention is created. Randomization is ideally combined with increasing degrees of blindness to eliminate bias. Statistical analysis of the data measures and facilitates interpretation of differences, providing a scientific basis for determining effect and efficacy.
   Current consensus in EBM includes the desirability of basing clinical decisions on large-scale RCT data. However, this is not always possible for several reasons. First, gold standard RCT data may not exist for the question posed. Substantial resources are required to design and carry out large-scale RCTs. Daunting numbers of patients and data are required to determine significant, clinically relevant differences between groups, especially for interventions with subtle or time-sensitive effects. Second, multiple RCTs looking at even a narrowly-defined question may give rise to different conclusions. Finally, publication bias may limit access to data, especially for RCTs with negative or non-significant results.
   Lacking large-scale RCT data, clinicians may rely on meta-analysis of pooled data from multiple small RCTs.10,11 Ideally, meta-analysis involves combining raw data from studies of similar design, in collaboration with original authors, in an effort to create sufficient sample size and power to determine likelihood (odds) of treatment benefit, adjusted for underlying risk in otherwise comparable groups.12 Actually, studies are typically not similar, containing highly-variable subjects, randomization, control, and treatment strategies, outcome measures, and length of follow-up. Pooling data under these circumstances may exaggerate bias and imprecision.13 Additionally, the results of meta-analyses and subsequent large RCTs on the same subject may not agree.14,15 It is also not uncommon for 2 meta-analyses of the same subject to reach opposite conclusions.16,17
   Despite these difficulties, all available published information may be considered in EBM; however, RCT data (with or without meta-analysis) carry the most weight with respect to recommended practice and disease management. Data from controlled studies without randomization, cohort or case-control analytic studies, multiple time series, uncontrolled experiments with dramatic results, respected opinions, and descriptive epidemiology are taken into account in descending order18 (Table 2).

Evidence of Efficacy for Acupuncture
   While respecting the above differences between TCM and EBM, it is also arguable that the distinction between them may be one of emphasis. EBM is "[t]he conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients. The practice of EBM means integrating individual clinical expertise with the best available external clinical evidence from systematic research,"4 with RCT data held as the gold standard. TCM practice places more emphasis on individual experience, but can and should be guided by convincing additional evidence concerning effective treatment.
   Clinicians seeking published external evidence concerning acupuncture treatment not found in textbooks can easily use PubMed (http://www.ncbi.nlm.gov) and other user-friendly databases, accessed by personal computer. The challenge has therefore shifted from acquiring access to published information to the question, "How good is this information, how reliable, and applicable to the clinical problem at hand?"
   The answer to that question depends, in part, on the context of the question. In TCM, questions concerning treatment strategies for an individual patient may be best addressed through individual case reports and case series, which tend to describe successful management of a specific problem. The search strategy might begin with acupuncture crossed with the diagnosis of the patient, then expanded to include references cited in the reports retrieved through the database. If available, the resulting external information is likely to be helpful to the practitioner, especially if the patients and practice setting described are similar to the practitioner's situation. RCT data looking at homogeneous groups of participants, under study conditions unlike those in the actual practice of TCM (i.e., sham acupuncture or restricted use of prescribed points), would likely not be helpful in clinical management.
   "Is acupuncture treatment effective in... ?" is a different question. It seeks information concerning the experience of a group of similar individuals treated with acupuncture, compared with an alternative modality. Queries of this nature might well be addressed through review of RCT data or meta-analysis. The difficulty arises, however, in that despite the advocacy and research training initiatives of the National Institutes of Health's National Center for Complementary and Alternative Medicine, the escalating funding for research, and the academic interest in acupuncture, quantity and quality of such data are generally marginal.19 More importantly, in meta-analysis especially, even for conditions such as asthma and pain for which acupuncture is generally believed to be helpful, the published data fail to conclusively answer the question "Does it work in...?" The following example illustrates this point. Recently (October 21, 1999), a simple PubMed search without restrictions provided the following results:

RCTs for any medical topic: 15,125
Meta-analyses, any topic: 6,048
Acupuncture references: 7,123
Acupuncture x RCT: 34
Acupuncture x meta-analysis: 13
Acupuncture x RCT x meta-analysis 6

   Convincing published data for acupuncture efficacy are minimal or lacking for most conditions, especially under the standards for EBM. This has not been ignored by either strong critics of acupuncture or by the acupuncture community itself. Despite their differences of opinion concerning efficacy, critics and advocates alike are calling for more rigorous clinical investigations with quantified analysis of objective data ideally gathered under RCT conditions.
What Is The Most Appropriate Study Design To Determine Acupuncture Efficacy?
   Accepting the importance of rigorous clinical investigation to inform and guide clinical practice, the question remains as to the most appropriate study design for acupuncture. Is there an investigative method available, consistent with the philosophical foundations and actual practice of clinical acupuncture, that would also satisfy EBM standards of randomization, control, and quantitative analysis? Is there an example of a study that would meet these requirements in the acupuncture literature?
   Many authors have addressed the first question.1,20 While acknowledging the importance of hypothesis to generate qualitative data, most have emphasized the need for quantitative data2 gathered under randomized and controlled conditions. There is no obvious consensus regarding how to address the problems of blinding, placebo control, uniform treatment, and crossover. Most call for large-scale RCTs3 consistent with EBM standards, despite difficulties, logistical and funding challenges, and the likelihood that even large studies may be inconclusive. Few have specifically urged consideration of alternative methods, specifically single-case reports and/or single-case series and their experimental designs, despite the potential to satisfy standards of evidence for both TCM and EBM.21,22
   "N-of-1" experimental designs were first proposed to measure treatment effect in behavior modification studies.23,24 Their applicability to problems in clinical medicine was soon appreciated and advocated.25 N-of-1 studies (i.e., single-case studies, single-subject designs, single-system strategies, time-series experimentation) are randomized trials in individual subjects with measurement of change from baseline for a single or multiple predetermined outcome indicator. Multiple variations of this model have been described, but all are hypothesis-driven and designed to include a baseline period of observation and measurement, randomized application of a single intervention, or series of discrete interventions applied in a structured manner with some degree of blindness, and followed by measured assessment of outcome.26 Both qualitative and quantitative data are obtained, followed by statistical analysis. Statistical methods specific to this tool have been described in detail.27
   N-of-1 studies are not without problems, including high risk of type II errors (erroneously failing to reject a null hypothesis which is, in fact, false). In addition, the results apply only to those participating in the trial; generalization of results to a population of similar patients requires a series of identically designed, single-case studies sufficient in number to become an RCT. Generally, N-of-1 studies can provide highly useful information, acceptable under the standards of individualized alternative therapies including acupuncture and EBM. An excellent example of this investigative strategy has recently been published on chiropractic manipulation in childhood asthma.28

CONCLUSION
   Despite their different foundations and weighing of information, TCM and EBM are not mutually exclusive. Both share a commitment to providing the best possible treatment for patients based on individual experience and the best available external evidence. Single-case experimental design studies, expanded into large-scale RCTs if they merit allocation of substantial resources, may be a valuable tool for clinical investigation as well as a source of credible evidence supporting the practice of conscientious acupuncture practitioners.

REFERENCES
1. Lewith GT, Machin D. On the evaluation of the clinical effects of acupuncture. Pain. 1983;16:111-127.
2. Levin JS, Glass TA, Kushi LH, Schuck JR, Steele L, Jonas WB. Quantitative methods in research on complementary and alternative medicine: a methodological manifesto. Med Care. 1997;35:1079-1094.
3. Margolin A, Avants SK, Kleber HD. Investigating alternative medicine therapies in randomized controlled trials. JAMA. 1998;280:1626-1628.
4. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS. Evidence based medicine: what it is and what it isn't. BMJ. 1996;312:71-72.
5. Kaptchuk TJ. Web That Has No Weaver: Understanding Chinese Medicine. New York, NY: Congdon & Weed; 1983.
6. Huang-fu M, Shou-Zhong Y, Chace C, trans. The Systematic Classic of Acupuncture and Moxibustion. Boulder, Colo: Blue Poppy Press; 1995.
7. Davidoff F, Case K, Fried FW. Evidence-based medicine: why all the fuss? Ann Intern Med. 1995;122:727.
8. Evidence-Based Medicine Working Group. Evidence-based medicine: a new approach to teaching the practice of medicine. JAMA. 1992;268:2420-2425.
9. Medical Research Council Investigation. Streptomycin treatment of tuberculosis. BMJ. 1948;2:769-782.
10. L'Abbe KA, Detsky AS, O'Rourke K. Meta-analysis in clinical research. Ann Intern Med. 1987;107:224-233.
11. Thacker SB. Meta-analysis: a quantitative approach to research integration. JAMA. 1988;259:1685-1689.
12. Sharp SJ, Thompson SG, Altman DG. The relation between treatment benefit and underlying risk in meta-analysis. BMJ. 1996;313:735-738.
13. Chalmers TC, Levin H, Sacks HS, Reitman D, Berrier J, Nagalingam R. Meta-analysis of clinical trials as a scientific discipline, I: control of bias and comparison with large co-operative trials. Stat Med. 1987; 6:315-328.
14. LeLorier J, Gregoire G, Benhaddad A, Lapierre J, Derderian F. Discrepancies between meta-analyses and subsequent large randomized, controlled trials. N Engl J Med. 1997;337:536-542.
15. Ioannidis JP, Cappelleri JC, Lau J. Issues in comparisons between meta-analyses and large trials. JAMA. 1998;279:1089-1093.
16. Patel M, Gutzwiller F, Paccaud F, Marazzi A. A meta-analysis of acupuncture for chronic pain. Int J Epidemiol. 1989;18:900-906.
17. Ter Reit G, Kleijnen J, Knipschild P. Acupuncture and chronic pain: a criteria-based meta-analysis. J Clin Epidemiol. 1990;43:1191-1199.
18. Balas EA. From appropriate care to evidence-based medicine. Pediatr Ann. 1998;27:581-584.
19. NIH Consensus Development Panel on Acupuncture. Acupuncture. JAMA. 1998;280:1518-1524.
20. Hammerschlag R. Methodological and ethical issues in clinical trials of acupuncture. J Altern Complement Med. 1998;4:159-171.
21. Eskinazi D. Methodologic considerations for research in traditional (alternative) medicine. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 1998;86:678-681.
22. Kawakita K. Oral presentation at: International Meeting of the World Federations of Acupuncture and Moxibustion Societies; Kyots; 1993.
23. Kazdin AE. Single Case Research Designs: Methods for Clinical and Applied Settings. New York, NY: Oxford University Press; 1982:368.
24. Barlow DH, Hersen MH. Single Case Experimental Designs: Strategies for Studying Behavior Change. New York, NY: Pergamon Press; 1984:419.
25. Guyatt G, Sackett D, Taylor DW, Chong J, Roberts R, Pugsley S. Determining optimal therapy: randomized trials in individual patients. N Engl J Med. 1986;314:889-892.
26. Portney LG, Watkins MP. Single-case experimental design. In: Foundations of Clinical Research - Applications to Practice. Norwalk, Conn: Appleton & Lange; 1993:191-232.
27. Rochon J. A statistical model for the "N-of-1" study. J Clin Epidemiol. 1990;43:499-508.
28. Balon J, Aker PD, Crowther ER, et al. A comparison of active and simulated chiropractic manipulation as adjunctive treatment for childhood asthma. N Engl J Med. 1998;339:1013-1020.

AUTHOR INFORMATION
Dr Lynnae Schwartz is a Pediatric Anesthesiologist and a Pediatric Research Fellow through the Department of Anesthesiology, National Institutes of Health. She was formerly a staff physician at Children's National Medical Center in Washington, DC. Dr Schwartz received her formal training in medical acupuncture at the New England School of Acupuncture.

Lynnae Schwartz, MD, MAc, FAAP*
3218 Brooklawn Ct
Chevy Chase, MD 20815
Phone: 301-718-1757
*Reprint requests to above address.

Back to Top | Table of Contents | On-line Journal Index