Research Database -
International Updates

Research Methodology

Issue 91

PARK and colleagues, Department of Complementary Medicine, School of Postgraduate Medicine and Health Sciences, University of Exeter, UK, J.B.Park@exeter.ac.uk, have reviewed (52 references) the nomenclature of Chinese medical formulae.
Background: Traditional Chinese Medicine (TCM) has been modified to some extent in other Asian countries such as Korea and Japan. Researchers in different countries seem to use different English names for the same TCM formula. Lack of knowledge of the Chinese characters is destined to increase this confusion.
Methods: All investigations of TCM published in the American Journal of Chinese Medicine were evaluated. PubMed was searched for TCM publications.
Results: 54 TCM formulae were identified in 45 reports published in AJCM. 32 were named in Chinese only (23 reports); 6 in Japanese (6 reports); 5 in Korean (5 reports). 10 formulae were named in Japanese with Chinese names in brackets (10 reports); and one in Chinese with the Japanese name in brackets.
Conclusions: By computerized literature search, different numbers of research papers were retrieved using keywords differentiated by language, location and number of hyphens. Such confusion may prevent progress in the evaluation of TCM. In order to increase the efficiency of studies on TCM, standardization of formulae is required.
Park J, Park HJ, Lee HJ, Ernst E. What’s in a name? A systematic review of the nomenclature of Chinese medical formulae. The American Journal of Chinese Medicine 30 (2-3): 419-427, 2002.

Issue 84

VICKERS, Integrative Medicine Service, Memorial Sloane-Kettering Cancer Center, New York, New York 10021, USA, vickersa@mskcc.org, shows how to reduce the number of patients needed for randomized trials.

Background: In a typical trial, patients are assessed for symptom severity at baseline, randomized to treatment or control and reassessed after a suitable follow-up period.

Methods: Medical statistics.

Results: It can be shown that when patients with chronic conditions are used, as is often the case with complementary therapies, the sample size can be reduced by two ways: repeat measurement and analysis of co-variance (ANCOVA). An example is given of a trial of acupuncture for back pain. Administering a pain questionnaire twice at baseline and four times at follow-up reduces the number of patients required by about 20%. A further 10% reduction is obtained by the use of ANCOVA.

Conclusions: It is desirable to reduce the number of patients required for a randomized controlled trial by a third; however ANCOVA is complex and expert statistical help is needed.

Vickers A. How to reduce the number of patients needed for randomized trials: a basic introduction. Complementary Therapies in Medicine 9 (4): 234-236, Dec 2001.

WHITE et al., Department of Complementary Medicine, School of Sport and Health Services, Exeter, UK, a.r.white@ex.ac.uk, make consensus recommendations for optimal treatment, controls and blinding in clinical trials of acupuncture.

Background: Acupuncture has been shown to be effective for the treatment of nausea and dental pain, and promising for migraine and osteoarthritis of the knee. For many other conditions the evidence is either insufficient or negative.

Methods: The authors discuss the reasons that misleading results may arise, such as inadequate treatment schedules and inappropriate control interventions.

Results: The authors make a number of recommendations which aim to improve the quality of sham-controlled acupuncture studies.

White AR, Filshie J, Cummings TM. Clinical trials of acupuncture: consensus recommendations for optimal treatment, sham controls and blinding. Complementary Therapies in Medicine 9(4): 237-245, Dec 2001.

Issue 83

MACPHERSON et al., Foundation for Traditional Chinese Medicine, York, UK, E: hugh@ftcm.org.uk, recommend standards for reporting interventions in controlled trials of acupuncture.

Background: Acupuncture treatment and control interventions in controlled trials of acupuncture are not always precisely reported.

Methods: An international group of experienced acupuncturists and researchers devised a set of recommendations for reporting acupuncture and control interventions. In a further consensus-building round, the editors of several journals helped redraft the recommendations.

Results: The recommendations are called STRICTA: STandards for Reporting Interventions in Controlled Trials of Acupuncture. They follow the Consolidated Standards for Reporting Trials (CONSORT) format. Participating journals are publishing the STRICTA recommendations and require prospective authors to adhere to them. Other journals are invited to adopt these recommendations.

Conclusions: The intended outcome is that interventions in controlled trials of acupuncture will be more adequately reported and lead to an improvement in critical appraisal, analysis and replication of results.

Background: Sham acupuncture treatments for use in controlled trials have to be evaluated for their suitability.

Methods: In a first experiment, acupuncture-naïve volunteers presenting with low back pain of at least 3 months’ standing received 6 insertions of real needles and 6 pokes with a toothpick in a guidetube in a two-period crossover design. In a second experiment, subjects were randomly assigned to receive either a complete treatment with acupuncture needles or a simulated treatment using a toothpick. Subjects’ perceptions about which implement was used were compared, while in the second experiment the patients’ perceptions of the treatment (acupuncturist’s warmth and caring, reasonableness of acupuncture as a treatment for low back pain) were compared.

Results: Toothpick insertions were perceived as slightly more real than real acupuncture needling (p = 0.08). Comparable proportions of respondents in the second experiment believed they were receiving the real treatment. Those receiving real acupuncture were more likely to report immediate pain relief.

Conclusions: The placebo treatment described here represents a reasonable control treatment for acupuncture-naïve patients in randomized controlled trials of acupuncture.

Comment: In recent years, there has been considerable publication activity regarding the lack of suitable research protocols to investigate acupuncture, particularly the lack of a suitable placebo or sham acupuncture technique. The above published studies demonstrate that progress is being made in addressing these issues regarding shortcomings in methodology.

Issue 82

REILLY, Glasgow Homoeopathic Hospital, Glasgow, Scotland, davidreilly@compuserve.com, reviewed (8 references) recent studies showing evidence for effectiveness of homeopathic medicines.

Background: Conceptual and scientific clashes make it difficult for orthodox Western medicine to accept homeopathy. However in recent years there has been a renaissance of interest in the approach. Approximately 20% of Scottish GPs have completed basic training. Homeopathy uses microdoses of potential toxins to provoke self-regulatory responses. This hints at its clinical scope: it can help, at times resolve, conditions that are intrinsically reversible rather than mechanical problems, deficiencies, or irreversible breakdowns in the body’s functions.

Discussion: Trials and meta-analyses of controlled trials are pointing towards real effects of homeopathy, mechanism of action unknown. Clinical outcome studies suggest useful clinical impact and excellent safety. There seems to be a potential to enhance patient care by integrating orthodox and homeopathic medicine.

Reilly D. The puzzle of homeopathy. The Journal of Alternative and Complementary Medicine 7 (Suppl 1): S103-9. 2001.

PLAUGHER and colleagues, Director of Research, Life Chiropractic College West, 25001 Industrial Boulevard, Hayward, California 94545, USA, examined the feasibility of carrying out a randomized controlled clinical trial of short- and long-term effects of chiropractic treatments given in the private practice setting to patients with high blood pressure compared with brief soft tissue massage or non-treatment.

Methods: This feasibility study was conducted in a private practice out-patient chiropractic clinic. It was a randomized, controlled, comparative trial involving three parallel groups of subjects. Subjects were 23 patients aged 24-50 years with systolic or diastolic essential hypertension. The chiropractic group received 2 months of full-spine chiropractic care (i.e. Gonstead) involving mainly specific contact, short-lever-arm adjustments to motion segments that showed signs of subluxation. The massage group received brief effleurage at localized areas of the spine believed to be showing signs of subluxation. The control group rested alone for about 5 minutes in an adjustment room during study treatment sessions. The investigators recorded the cost per enrolled subject and measured systolic and diastolic blood pressure (BP) using a random-0 sphygmomanometer and patient-reported health status (using SF-36). They also assessed subjects’ cooperation with the randomization procedures and measured drop-out rates, recruitment effectiveness, temporal stability of BPs at the start of the study and the influences of the inclusion and exclusion criteria on the subject pool.

Results: 30 subjects enrolled. One was later deemed to be ineligible and six subjects dropped out during the study. Cost per subject enrolled was US$161. All subjects in the chiropractic and massage therapy groups were classified as either overweight or obese, whereas only two were classified thus in the control group. Patient-reported health status (SF-36 profile) was similar to that of a normal population for all three experimental groups. Mean changes in diastolic BP were as follows: chiropractic group, -4; brief massage group, 0.5; control group, -4.9. At the end of the study, mean changes in diastolic BP were as follows: chiropractic group, -6.3; brief massage group, -1.0; control group, -7.2 Mean improvements in the chiropractic and control groups remained consistent over the follow-up period.

Conclusion: This pilot/feasibility study uncovered a number of issues needing to be addressed before proceeding to a full-scale clinical trial. Recruitment may require a multidisciplinary approach to identify sufficient numbers of suitable patients with hypertension who are not taking medication to control their condition. Methods to ensure comparable treatment groups with regard to prognostic variables (e.g. weight) need to be implemented. The findings of this study, however, indicate that it is feasible to conduct a full-scale, three-group, randomized clinical trial in the private practice setting to investigate the effects of chiropractic adjustments.

Plaugher G et al. Practice-based randomized controlled-comparison clinical trial of chiropractic adjustments and brief massage treatment at sites of subluxation in subjects with essential hypertension: pilot study. Journal of Manipulative and Physiological Therapeutics 25 (4): 221-39. May 2002.

Comments: The above research study (which was actually a feasibility study prior to conducting a fully-fledged clinical trial) demonstrates the myriad of details which have to be considered and implemented in designing a truly randomized clinical trial in which all groups of patients are as equal as possible, so that one or more treatment groups don’t turn out to be obese or overweight and the control group of normal weight.

BERMAN, Complementary Medicine Program, University of Maryland at Kernan Hospital, Baltimore 21207-6697, USA, Bberman@compmed.ummc.ab.umd.edu, reviewed (24 references) studies that are of sound methodological quality and contribute to an understanding of acupuncture.

Background: It is difficult to design an appropriate control group for acupuncture studies, because acupuncture is an invasive, physical modality. No single control group can answer all research questions.

Discussion: Many types of control groups, each with advantages and disadvantages, have been used, depending on the specific questions asked. Examples are taken from the different categories of control groups, and a model for future acupuncture research which is at the same time comprehensive and cost-effective, is described.

Berman BM. Clinical applications of acupuncture: an overview of the evidence. The Journal of Alternative and Complementary Medicine 7 (Suppl 1): S111-8. 2001.

SHERMAN and colleagues, Northwest Institute of Acupuncture and Oriental Medicine, Seattle, Washington, USA, sherman.k@ghc.org, investigated the diagnoses and treatments provided by acupuncturists practising Traditional Chinese Medicine (TCM) for patients suffering from chronic low-back pain.

Background: In order to select treatments that are valid for testing in clinical trials, it is necessary to understand how acupuncturists diagnose and treat specific health problems.

Methods: The investigators examined two separate sets of treatment records that provided information on the diagnosis and treatment of patients with chronic low-back pain by TCM methods and acupuncture. The records covered more than 150 initial patient visits.

Results: 85% of patients received a diagnosis of Qi and Blood Stagnation or Qi Stagnation. 33-51% of patients had a diagnosis of Kidney Deficiency or one of its three subtypes. Less than 20% of patients received other specific diagnoses. On average, 12-13 needles were used per treatment. More than 85 different acupuncture points were used in each of the two data sets. However, only five to six acupuncture points were used in more than 20 treatments in each data set, and only two of these points (UB23 and UB40) were the same in both data sets. More than 50% of patients received other treatments in addition to acupuncture needling, such as heat (36-67%) and cupping (16-21%).

Conclusion: There was a great deal of variability in the treatments provided to different patients for the condition of chronic low-back pain. This presents a substantial challenge for selecting a single treatment for investigation in a clinical trial. The authors recommend that researchers attempt to develop a treatment programme that is considered valid by expert acupuncturists and that has ‘broad features characteristic of patterns of common clinical practice’.

Sherman KJ et al. The diagnosis and treatment of patients with chronic low-back pain by traditional Chinese medical acupuncturists. The Journal of Alternative and Complementary Medicine 7 (6): 641-50. Dec 2001.

Comments: Again, the above study demonstrates how difficult it can be to come up with a treatment programme, this time with acupuncture, which would be recognized as valid by acupuncturists and at the same time deals with the type of treatments usually delivered in clinical practice. This study showed that there was very little similarity between treatments carried out in the 150 patient visits. Of course, it may be that there is no such thing as a characteristic acupuncture treatment protocol, and that there has to be great variation in the way treatments are delivered. If this were to be the case, it would argue against using a single treatment regimen to validate acupuncture treatment for low back pain, but would support using clinical case records.

Issue 78

Issue 75

LINDE and colleagues, Institute for Social Medicine Epidemiology, Charite Hospital, Humboldt-Universitaet, 10098, Berlin, Germany, Klaus.Linde@lrz.tu-muenchen.de, investigated differences between randomized and non- randomized clinical trials in terms of patient characteristics, intervention characteristics, quality aspects, response rates and the provision of additional useful information (e.g. long-term effects, side effects, prognostic indicators), in order to determine whether non-randomized studies should be included in systematic reviews. They used the case of acupuncture for the treatment of chronic headache.
Methods: The researchers performed meta-analysis of clinical trials of acupuncture for chronic headache. Studies had to include at least 5 patients and report clinical outcome data. Studies were identified through searches of Medline, Embase, the Cochrane Controlled Trials Register, other databases and bibliographies.
Results: They identified 24 randomized trials and 35 non-randomized studies (5 non-randomized, controlled cohort studies; 10 prospective, uncontrolled studies; 10 case series; and 10 cross-sectional surveys). Randomized trials had smaller sample sizes, met more quality criteria and produced lower response rates (0.59) versus 0.78. Studies that met more quality criteria produced lower response rates. There was no difference in follow-up periods between randomized and non-randomized studies. 3 non- randomized studies included an analysis of prognostic variables. Only 1 non-randomized study reported on side effects. Non-randomized studies confirmed the finding of a systematic review of randomized trials that acupuncture is likely to be effective in the treatment of chronic headache. However, non-randomized studies provided little relevant additional information on long-term effects, prognostic factors or side effects.
Linde K et al. Should systematic reviews include non-randomized and uncontrolled studies? The case of acupuncture for chronic headache. Journal of Clinical Epidemiology 55 (1): 77-85. Jan 2002.

Issue 65