This article will examine the evidence for the use of homeopathy in the treatment of Rheumatoid Arthritis (RA). The successful treatment of RA using homeopathy has been recently highlighted by two published papers. Spence et al. (2005) undertook a six year university hospital outpatient observational study in which patients received homeopathic treatment for various ailments. Amongst the many health problems monitored, 10% of the participants constituted rheumatology department referrals. Outcome scores were assessed during the consultations. Of these, 70% of the RA patients benefited from adjunctive homeopathic treatment, with 30% recording that they felt ‘better’ and 19% recording that they felt ‘much better’ according to the outcome scale measurement. During a pilot study, using MYMOP 2 (Paterson, 1998) as a primary outcome measure, 6.9% of 482 patients surveyed, reported joint and mobility problems as the primary problem for which they had visited a homeopath (Relton et al., 2007). These two papers provide evidence that homeopaths have treated patients suffering from RA, and therefore it is a valid area in which to review the available literature.
RA is a chronic inflammatory condition which can be progressive and incapacitating. As the disorder is systemic by nature, it can affect extensive tissues (Gascoigne, 2001). Further health problems which may occur as a result of progressive RA are ankylosis and bone deformity (Weller, 2001). Onset of the condition usually occurs between the ages of 30 – 40, and it affects three times more women than men. Aetiology of the complaint is unknown, although an autoimmune process is recognised as being a significant factor. Occasionally a theory emerges that the onset of RA is caused by an infective factor, although this has not been corroborated (Gascoigne, 2001). Conventional treatment of RA concentrates on symptom suppression, consisting primarily of non-steroidal anti-inflammatory drugs (NSAIDs), disease modifying drugs (DMARDs) or slow-acting anti-rheumatic drugs (SAARDs) (National Rheumatoid Arthritis Society (NRAS), 2007). These therapies may produce side-effects such as indigestion, poor appetite, nausea, vomiting, blood disorders and kidney and liver damage (Smith, 2005).
The seriousness and the progressive nature of the disorder together with the lack of effective treatment and potential side-effects implies that complementary therapies may offer an alternative to controlling pain, the acute inflammatory response and the possible subsequent destruction of joints (Kenyon, 1994). Homeopathy provides treatment for patients without the side-effects associated with conventional drug therapy, therefore offering a safe, effective solution if the correct remedy is prescribed (Swayne, 1998). Classical homeopathy concentrates on prescribing for a patient’s individual symptoms. Analysis of the case concentrates on what is unique to the individual patient, and therefore a variety of remedies may be prescribed for a conventionally diagnosed condition (Roy, 1999). Having identified these aspects of ‘classical homeopathy’ it seems crucial to evaluate the potential of individualised homeopathic prescriptions for RA, and critically analyse the evidence already available.
This article will concentrate on three trials (Gibson, Gibson, MacNeill & Watson Buchanon, 1980; Andrade, Ferraz, Atra, Castro & Silva, 1991; Fisher & Scott, 2001) which have been conducted in order to assess the effectiveness of individualised homeopathy in the treatment of RA. All three trials are randomised controlled trials (RCTS) which are deemed the ‘gold standard’ for evaluating treatment efficacy in health research (Kane, 2004), but all three use individually prescribed homeopathic remedies, which resembles the true classical homeopathy scenario.
Critical Examination of selected studies on this topic
The three papers will be examined in date order starting with the Gibson et al. paper (1980).
Homeopathic Therapy in Rheumatoid Arthritis: Evaluation by Double-Blind Clinical Therapeutic Trial
Gibson, R.G., Gibson, S.L.M., MacNeill, A.D. & Watson Buchanan, W. (1980)
This trial was set at the Glasgow Homeopathic Hospital, Centre for Rheumatic Diseases. It was undertaken in response to the 1978 trial and subsequent paper entitled ‘Salicylates and Homeopathy in Rheumatoid Arthritis: Preliminary Observations’ (Gibson et al., 1978). Gibson et al. highlighted that the initial trial was methodologically flawed for the following reasons:-
In order to evaluate the importance of the methodological flaws, the 1980 trial was more rigidly controlled to limit potential variables which may have influenced the outcome. A randomised patient control group design was supervised by a third independent physician to determine reliability. Inclusion criteria are detailed in the table below:
|All patients satisfied diagnostic criteria of the American Rheumatism Association (ARA)|
|All patients were not using any more than one conventional anti-inflammatory drug|
|All patients were mobile|
|None had ‘advanced’ or ‘burnt out’ RA (except one patient who had gold treatment several years previously)|
|None had received corticosteroids, chrysotherapy, d-penicillamine or levamisole previously|
Patient progress was measured using the outcomes detailed below:
Gibson et al. (1980) recognised that homeopathic prescriptions are very individual to the patient. Homeopathic principles dictate that patient prescriptions should be established on a holistic individual basis. Although some symptoms are highly regarded in the analysis of the consultation, the homeopath prescribes on the whole symptom picture, not merely on certain characteristics (Wheeler, 1982). Although high internal validity is maintained by the division of patients into two groups, with ‘good’ prescribing symptoms (‘R’ group) or ‘poor’ prescribing symptoms (‘U’ group) (Appendix 1), ensuring precise distribution between placebo and active treatment groups, this evaluative practice is not reflective of homeopathic practice (Roberts, 1982).
As further variables were limited by the equal division between two physicians of patients using the same drug therapy and the maintenance of an unchanged dosage of anti-inflammatory agents, the quality of evidence produced is much improved from Gibson et al. (1978). However, this trial lasted only six months compared to the previous study of twelve months. The latter would be considered a more realistic timescale for the treatment of a chronic condition, potentially revealing statistically improved results in favour of homeopathy. In spite of this, and the small sample size (46), the enhanced response to homeopathy suggests that it may be a useful addition for the treatment of RA.
Ethics have been considered on two occasions. As some participants were prescribed placebo, conventional medication was not withdrawn. Therefore, the principle of non-maleficence has been observed. Those patients who had previously taken placebo were treated for a further three months with individualised homeopathy, observing the ethical principle of beneficence (Beauchamp and Childress, 2001).
Another factor considered important was the potential for improved response to homeopathic treatment due to the synergistic effects of the homeopathic consultation (Lewis, Jonas and Walach, 2002). It has been suggested that the whole experience of homeopathy including the homeopathic consultation, affects the response of the patient. The mere fact that a patient is at the centre of the homeopathic intervention may lead to improvements beyond the expectations of ingesting the remedy. However, this assumption was found to be inaccurate as the placebo groups for both physicians were not statistically improved.
A Randomised Controlled Trial to Evaluate the Effectiveness of Homeopathy in Rheumatoid Arthritis
Andrade, L.E.C., Ferraz, M.B., Atra, E., Castro, A. and Silva, M.S.M. (1991)
This trial was undertaken in 1991 at the Escola Paulista de Medicine in Sao Paulo. The authors claim that RCTs evaluating homeopathy are rare and often present conflicting results and believe that the application of appropriate scientific methodology will establish the RCT as a valuable tool for assessing the validity of homeopathy. However, analysis of the methodology of the six month, double blind trial demonstrated flaws and potential variables which may have seriously affected the outcome.
Patients were consecutively chosen from the rheumatic disease outpatient clinic where the trial occurred and met at least three specific inclusion criteria for classical RA:
|Ritchie’s Articular Index (7) > 20|
|Morning stiffness > 45 minutes|
|Tenderness of more than 6 joints|
|More than 3 swollen joints|
|Erythrocyte Sedimentation Rate (ESR) . 28mm/ 1st hour|
A number of medical tests for RA indicators were undertaken at baseline, three and six months, in order to establish whether the condition had worsened, stabilised or improved during the trial (Appendix 2). These tests concluded that there was not any statistical change from the baseline results in either group. Monthly follow ups continued and the remedy was maintained or changed according to patient response. Physical therapy was not undertaken during the trial. Outcomes were measured using the same criteria as the Gibson et al. (1980) trial.
Although remedy prescriptions were individualised according to patient requirements, the potencies used were limited to 5 – 30C. This is not reflective of homeopathic practice, as many factors are accounted for when choosing potency and dosage such as ‘energy’ of the patients, whether conventional medication is used regularly, and the length of duration of the chronic condition (Vithoulkas, 1980). The specific measurement of the homeopathic intervention is compromised by the fact that a range of potencies is available. It is unclear in the report whether patients received a constitutional and a therapeutic (local) remedy concurrently. The criteria for using this method are not detailed and its inclusion would have allowed for a better understanding of the philosophy behind it.
Although the results for the active treatment group demonstrated that homeopathy may be useful in the treatment of RA, there was no statistical evidence that it was superior to placebo therapy. However, this result may be due to a number of methodological flaws within the trial. Again the study sample was small (44). Unlike the Gibson et al. (1980) trial, care was not taken in dividing the sample evenly, thus possibly skewing the results (Appendix 3). Furthermore, patients were authorised to continue with, or alter dosage of NSAIDs and steroidal drugs at the physician’s discretion, thereby not eliminating this variable. Although dose reduction was higher in the placebo group, patients within this group suffered more serious symptoms statistically than the active treatment group. Some patients received alternate homeopathic remedies on a daily basis, thereby not reflecting classical homeopathic philosophy. It is also unclear whether all patients received alternate remedies, or merely those whose symptoms were not precise enough to warrant a single remedy prescription.
A Randomised Controlled Trial of Homeopathy in Rheumatoid Arthritis
Fisher, P. & Scott, D.L. (2001)
Fisher and Scott undertook this trial to test the hypothesis that homeopathy is effective in reducing the symptoms of joint inflammation in RA using a six-month randomised, cross-over, double-blind, placebo-controlled study design. Ethical approval was granted by Saint Bartholomew’s and the Homerton Ethical Committee. All participants, recruited from a single rheumatology clinic, had classic symptoms of RA according to the ARA criteria (Appendix 4). Other trial inclusion and exclusion criteria are detailed below:
|the patient was seropositive for rheumatoid factor|
|the patient was receiving stable doses of single NSAIDs for more than three months or DMARDs with or without NSAIDs for more than six months|
|Patients with severe disability, had taken systemic steroids in the previous six months, or had withdrawn from DMARD therapy in the previous twelve months were excluded|
The main outcome measures for analysis were visual analogue scale pain scores, Ritchie articular index, duration of morning stiffness and ESR. Although the initial recruitment of participants was higher in number (112) compared to the previous trials analysed, there was a high attrition rate of 48.2%. It is unclear, however, whether intention to treat data was analysed. If it was not, although mean pain scores were significantly lower after three month’s placebo therapy, results could have been biased if the whole group was not analysed. Randomisation was stratified in order to ensure full representation in each group by participants in the trial (Kane, 2004). Remedy prescriptions were individualised and could be adapted according to disease progression, although potency used was low, either 6cH or 30cH. Although this design limits variables, it is once again unreflective of daily homeopathic practice.
A crossover trial design was utilised, and at the three month interval participants changed over to placebo or active therapy. This type of design is less common as there is a possibility of a residual effect of therapy which may continue during the next period of treatment, thereby potentially skewing results (Lewith et al., 2002). This may be true, as remedies prescribed may affect a person’s health levels for some time after the final dose (Gibson et al., 1980). Fisher and Scott (2001) debate the reasons that the placebo group showed significant improvement in pain scores. Although they offer the suggestion that this may be due to an aggravation in the initial stages of the treatment, which may occasionally occur, they do not seem to have considered the residual effect of active homeopathic treatment.
Finally Fisher and Scott conclude that a new investigational approach is required which will try to discover whether homeopaths can genuinely control patient symptoms rather than continuing to compare homeopathy to a placebo response.
The potential contribution other research designs could make to understanding the topic
The three trials which have been included within this article were all RCTs which had been adapted to account for the complexities of the homeopathic method. RCTs are deemed the ‘gold standard’ of evidence based medicine (EBM) in the current medical era (Bowling, 1997). However, the intricacies of homeopathy are not easily measured using this form of research, as RCTs are typically used to evaluate a single component of a treatment method. Lewith et al. (2002) believe that RCTs can render research a success when the most important outcome measure is agreed upon, the disease and its potential outcomes are easily defined and measured, and the intervention is simple, short and content based. Unfortunately the complex nature of the homeopathic intervention often confounds the RCT’s strict rules, leading to research results which often are conflicting and controversial (Bell and Koithan, 2006).
RCTs are designed to answer a simple question about the specificity of an intervention. Homeopathy, however, is not a simple intervention. The philosophy of homeopathy is founded upon the treatment of the whole person, rather than the specific disease (Roy, 1999). Therefore, during treatment a homeopath aims for improvement in many areas of a patient’s life, not just for the presenting complaint.
The high internal validity of RCTs which is essential in providing high quality trials often reduces the meaning of the research results in terms of ‘real world’ situations or external validity (Lewith et al., 2002). In two of the trials examined, although individual homeopathy was prescribed to patients, freedom to prescribe potencies of choice was curtailed, thus not reflecting the ‘real world’ situation of homeopathic prescribing where a practitioner bases the prescription on numerous factors, such as the length of duration of disease, the ‘energy’ of the patient, and whether allopathic medicine is currently used (Vithoulkas, 1980). Outcome measures in all three trials were narrow and specific, rather than patient-centred and multi-dimensional thus not reflecting the dynamic nature and ‘real world’ practice of homeopathy (Bell, 2003). The length of the trials was also not a true reflection of homeopathic practice. The treatment of chronic disease where patients are and have been taking conventional medicine for a number of years is often reflected in the time taken for homeopathy to have significant effect on the presenting complaint although non-specific mental and emotional improvements may occur previous to the improvement of physical problems (Grossinger, 1993). Homeopathic practitioners observe changes in the positive dimensions of a patient’s health, not merely improvement in the disease (Bell, 2003).
One reasonable suggestion which may provide important qualitative evidence for the use of homeopathy in RA is the large scale outcome study (Lewith et al., 2002). For conditions such as RA, where the natural course of the disease is known, and the prognosis can be estimated, documentation of the effects of homeopathic treatment for a longer time would provide important information about the usefulness of homeopathy in such a chronic condition. Lewith et al. (2002) suggest that in conditions such as RA, where there may often be a history of ineffective treatment, both doctors and patients are often desperate for interventions whether specific or non-specific, which may improve patient’s quality of life.
Bell (2003) suggests that the homeopathic profession currently has a paucity of systematic qualitative and observational data especially in the treatment of chronic conditions, and that large scale observational studies of such conditions could direct the future design of homeopathically orientated RCTs. She proposes a three stage research programme for homeopathy. Each stage would move the research process forwards culminating in the development of homeopathic research standards which could be implemented in large scale, multi-site controlled clinical trials. This specific longitudinal research design would be especially useful for chronic conditions providing a focus for the translation of homeopathic remedy effects into measurable changes in the holistic dynamics of a patient.
|Stage of Study||Objectives|
|First Wave of studies||Qualitative, observational|
|Second Wave of Studies||Feasibility, pilot controlled studies to establish effect size of specific study designs, accommodating specific homeopathic issues|
|Third Wave of Studies||Development of homeopathic research standards and implementation in large scale, multi-site controlled clinical trials|
Mixed methods research may provide a suitable solution for research into homeopathy and RA. Hailed as a third paradigm in research, this method entails the integration of quantitative and qualitative data collection and analysis in a single study (Cresswell et al., 2004). The provision of quantitative data analysing physical changes using measurements such as grip strength, and qualitative data which would concentrate on subjective data such as improvements in general well-being, would afford a more holistic research agenda for RA (Bell, 2003). Acupuncture research, already published using mixed methods research, has demonstrated the phenomenon of ’emergence’ where patients describe multi-dimensional changes in their health, occasionally omitting to mention the state of their initial complaint (Paterson, 2006). This type of research may allow researchers to develop techniques which may precisely represent ‘real world’ homeopathy, ensuring external validity but also honouring internal validity by employing more precise and less subjective quantitative data collection techniques.
Another development in homeopathic research which Bell and Kothian (2006) suggest is the utilisation of homeopathic theories such as Hering’s Law of Cure, which is considered during homeopathic consultations to ascertain whether healing has occurred. As Hering’s Law of Cure follows a traceable distinct pattern in patients, it could be applied in a research situation to assess whether healing in certain chronic diseases follows the same sequence. Thus the utilisation of a specific homeopathic theory could provide the basis for a theory-driven outcome for controlled research. This type of research would identify patterns of healing in a whole system, rather than reducing the analysis of a specific point in time to a specific outcome. The outcomes would be probabilistic not certain, as whole systems react differently to interventions and by their very nature are non-reductionist.
Homeopathy does not profess to act with local specificity, but rather on the whole system. Therefore, research, such as RCTs, which measure specific effects, are not truly reflective of the homeopathic system, and therefore often produce contentious results. Although RCTs have historically been viewed as the ‘gold standard’ of research into conventional medical interventions, it is time that homeopathic research truly reflected the uniqueness of its delivery, allowing homeopathy to be more accurately examined, thus enabling its true value to be revealed.
Whole systems research seems to provide a distinct way forward in homeopathic research since it would provide a more holistic form of research. According to homeopathic understanding of disease and health, many patients express illness not only in their physical body, but also mentally and emotionally. Therefore when assessing the action of homeopathic treatment, not only should changes in physical symptoms be evaluated, but also changes in the mental and emotional state (Vithoulkas, 1980).
Andrade, L.E.C., Ferraz, M.B., Atra, E., Castro, A. & Silva, M.S.M. (1991) A Randomized Controlled Trial to Evaluate the Effectiveness of Homeopathy in Rheumatoid Arthritis Scand J Rheumatology 20 pp.204-208
Apte, S.S. (1989) Clinical Research in Homeopathy, An Experience with Rheumatoid Arthritis Indian Journal of Homeopathic Medicine pp.27- 30
Beauchamp, T.L. & Childress, J.F. (2001) Principles of Biomedical Ethics 5th ed. New York: Oxford University Press
Bell, I.R. (2003) Evidence-Based Homeopathy AJHM 96 (1) pp.17-31
Belladonna I.R. & Koithan, M. (2006) Models for the Study of Whole Systems Integrative Cancer Therapies 5(4) pp.293-307
Bowling, A. (1997) Research Methods in Health Buckingham: Open University Press
Brien, S., Lachance, L. & Lewith, G.T. (2004) Are the Therapeutic Effects of Homeopathy Attributed to the Consultation, the Homeopathic Remedy, or Both? A Protocol for a Future Exploratory Feasibility Trial in Patients with Rheumatoid Arthritis The Journal of Alternative and Complementary Medicine 10(3) pp. 499-502
Cresswell, J.W., Fetters, M.D. & Ivankova, N.V. (2004) Designing A Mixed Methods Study in Primary Care Annals of Family Medicine 2 (1) pp.7-12
Ernst, E. (2003) Complementary Medicine Current Opinion in Rheumatology 15(2) pp.151-155
Fisher, P. (1990) Research into Homeopathic Treatment of Rheumatological Disease: Why and How? Complementary Medical Research 4(3) pp.34-40
Fisher, P. & Scott, D.L. (2001) A Randomized Controlled Trial of Homeopathy in Rheumatoid Arthritis Rheumatology 40 pp.1052-1055
Gascoigne, S. (2001) The Clinical Medicine Guide Cork: Jigme Press
Gibson, R.G., Gibson, S.L.M., MacNeill, A.D., Gray, G.H., Carson Dick, W. & Watson Buchanan, W. (1978) Salicylates and Homeopathy in Rheumatoid Arthritis: Preliminary Observations British Journal of Pharmacology 6(5) pp.391-395
Gibson, R.G., Gibson, S.L.M., MacNeill, A.D. & Watson Buchanan, W. (1980) The Place for Non-pharmaceutical Therapy in Chronic Rheumatoid Arthritis: A Critical Study of Homeopathy The British Homeopathic Journal 69 (3) pp.121-133
Grossinger, R. (1993) Homeopathy: An Introduction for Skeptics and Beginners California: North Atlantic Books
Haila, S., Koskinen, A. & Tenovuo, J. (2005) Effects of Homeopathic Treatment on Salivary Flow Rate and Subjective Symptoms in Patients with Oral Dryness: A Randomized Trial Homeopathy 94 pp.175-181
Hornung, J. (1990) Misconception or Guidepost? The 1978 Study by Gibson et al. on Rheumatoid Arthritis – reviewed The Berlin Journal on Research in Homeopathy 1(1) pp.77-84
Kane, M. (2004) Research Made Easy in Complementary and Alternative Medicine London: Churchill Livingstone
Kenyon, J.N. (1994) Rheumatoid Arthritis: The Alternatives Considered Complementary Therapies in Medicine 3 pp.75-78
Lewith, G., Jonas, W.B. & Walach, H. (2002) eds. Clinical Research in Complementary Therapies London: Harcourt Publishers Ltd
MacNeill, A.D. (1977) Rheumatoid Arthritis: Report on an On-going Trial British Homeopathic Journal 71(1) pp.13-19
National Rheumatoid Arthritis Society (2007) Treatment for Rheumatoid Arthritis [online] Available at: <URL: > [Accessed: 14th May 2007]
Paterson, C. (1998) Measuring Outcomes in Primary Care: a Patient Generated Measure MYMOP2 Compared with the SF36 Health Survey British Medical Journal pp. 312-316
Paterson, C. (2006) Measuring changes in self-concept: a qualitative evaluation of outcome questionnaires in people having acupuncture for their chronic health problems BMC Complementary and Alternative Medicine 6(1) pp.7
Relton, C., Chatfield, K., Partington, H. & Foulkes, L. (2007) Patients Treated by Homeopaths Registered with the Society of Homeopaths: a pilot study Homeopathy 96 pp.87-89
Roberts, H.A. (1982) The Principles and Art of Cure by Homeopathy reprint ed. Saffron Walden: Health Science Press
Roy, M. (1999) The Principles of Homeopathic Philosophy London: Winter Press
Smith, S. (2005) Medical Homeopathy Brighton: The South Downs Publishing Company
Spence, D.S., Thompson, E.A., Barron, S.J. (2005) Homeopathic Treatment for Chronic Disease: A 6-Year University-Hospital Outpatient Observational Study The Journal of Alternative and Complementary Medicine 11(5) pp.793-798
Swayne, J. (1998) Homeopathic Method New York: Churchill Livingstone
Van Haselen, R.A., Graves, N., Dahiha, S. (1999) The Costs of Treating Rheumatoid Arthritis Patients with Complementary Medicine: Exploring the Issue Complementary Therapies in Medicine 7 pp.217-221
Vithoulkas, G. (1980) The Science of Homeopathy New York: Grove Press
Weller, B.F. ed. (2001) Balliere’s Nurses’ Dictionary 23rd ed. Edinburgh: Harcourt Publishers Limited
Wheeler, C. (1982) An Introduction to the Principles and Practice of Homeopathy 4th ed. Saffron Walden: The C.W. Daniel Co. Ltd
Good Prescribing Symptoms (‘R’ patients)
Onset of symptoms following-
- a sudden fright
- physical injury
- profound emotional or physical trauma
Complaint affected by-
- climatic conditions e.g. damp or dry weather, heat, frost or wind
- other factors such as movement, rest or time of day
- marked craving or aversion to certain foods
For female patients –
- physical, emotional, mental changes during or after the menstrual period
Other anomalies –
- Marked mental or emotional peculiarities e.g. extreme tidiness, fear of heights or unusual reactions to sympathy
(Gibson et al., 1980)
Medical Tests conducted during the trial
|Test||Baseline||3 Months||6 Months|
|Full Blood Count||X||X||X|
|Serum Protein Electrophoresis||X||X|
|Glutamic – oxalacetic transferase||X||X|
(Andrade et al., 1991)
Weaknesses in Baseline Demographics
|Previous Use of 2nd line drugs||9||7|
|Morning Stiffness (mins)||107||92|
|15 Metre Walking Time (secs)||21.9||15.9|
|Ritchie Articular Index||23.7||18.7|
|Grip Strength (mm Hg)||68.0||79.5|
|Functional Class 1||2||3|
|Functional Class 2||12||14|
|Functional Class 3||9||4|
(Andrade et al.,1991)
American College of Rheumatology
Classification of Global Functional Status in Rheumatoid Arthritis
Class I Completely able to perform usual activities of daily living (self-care, vocational, and avocational
Class II Able to perform usual self-care and vocational activities, but limited in avocational activities
Class III Able to perform usual self-care activities, but limited in vocational and avocational activities
Class IV Limited in ability to perform usual self-care, vocational and avocational activities
Usual self-care activities include dressing, feeding, bathing, grooming and going to the toilet. Avocational (recreational and/or leisure) and vocational (work, school, homework) activities are patient-desired and age- and sex-specific.