ArticlesAre the clinical effects of homoeopathy placebo effects? A meta-analysis of placebo-controlled trials
Introduction
Bh1tween 30 and 70% of patients in developed countries use complementary, alternatior unconventional medicine,1, 2, 3 even though high-quality scientific research on these practices is lacking.4 Homoeopathy is one of the most widespread and controversial of these therapies. There are two main theoretical tenets: the principle of “similars” and the use of dilutions called “potencies”.5 The principle of similars states that patients with particular signs and symptoms can be cured if given a drug that produces the same signs and symptoms in a healthy individual. The second principle is that remedies retain biological activity if they are repeatedly diluted and agitated or shaken between each dilution. These dilutions are said to produce effects even when diluted beyond Avogadro's number in which no original molecules of the starting substance remain. How the solution “remembers” information from the original substance is speculative.6
Mh1ny scientists think that homoeopathy violates natural laws7 and thus effect must be a placebo effect.8, 9 But use of and belief in the effectiveness of homoeopathy is widespread and growing among physicians and the public,10, 11, 12, 13 and advocates claim that there are measurable and reproducible effects over placebo.14 A systematic review of 107 controlled clinical trials in homoeopathy by Kleijnen et al in 1991 showed a surprising number of positive results, even among those that received high quality-ratings for randomisation, blinding, sample size, and other methodological criteria.15 Vote counts of positive and negative trials, as used in that review, can be misleading without a quantitative summary of results. Since that study was published, at least 50 more controlled trials in homoeopathy have been reported.
Wh1 aimed to assess whether the effect seen with homoeopathic remedies is equivalto that seen with placebo. If the hypothesis that all clinical effects of homoeopathy are due to placebo is correct, it would mean that in all properly conducted placebo-controlled trials on homoeopathy, one placebo had been compared with another. The overall results of these trials, in any disease, should vary randomly around a zero difference between groups. This placebo hypothesis would be falsified if all properly conducted comparisons of homoeopathy and placebo showed a pooled effect significantly different from zero difference, or if there is independently replicated evidence for an effect over placebo in at least one consistently applied homoeopathic approach. Of course, evidence of an effect over placebo would be stronger if both approaches showed “positive” effects and “non-believers” were involved in the trials. We have tested both of the above strategies (overall comparison and reproducibility comparisons) with quantitative meta-analytic methods.
Section snippets
Literature search and data sources
All published reports of controlled clinical trials of homoeopathy were collected with use of multiple sources: (1) the review by Kleijnen et al,15 which used an extensive search strategy for MEDLINE and EMBASE up to 1990; (2) a MEDLINE search by an information specialist from 1966 to August, 1995, with the full-text terms homeop* and homoeop*, and the MeSH terms homoeopathy, homoeopathy, and alternative medicine, and screening of all citations found; (3) contacts with homoeopathic researchers,
Literature search and study selection
186 trials were identified (excluding drug “provings”). Of these, one was a time-series, placebo-controlled single-case experiment, 14 involved assessment of physiological measures on healthy volunteers, and 38 did not include a placebo group, leaving 133 placebo-controlled trials evaluating treatment or prevention. Of these, three trials were not randomised, nine were unclear about randomisation and double-blinding, and in two trials a single-blind design was used but a statement on treatment
Discussion
The results of our meta-analysis are not compatible with the hypothesis that the clinical effects of homoeopathy are completely due to placebo. But there is insufficient evidence from these studies that any single type of homoeopathic treatment is clearly effective in any one clinical condition. The evidence in our overall analysis would be more compelling if there were independently replicated, large-scale rigorous trials of defined homoeopathic approaches in at least a few specific disorders.
References (145)
- et al.
Prevalence and cost of alternative medicine in Australia
Lancet
(1996) Trials of homoeopathy
Lancet
(1993)- et al.
Belief in the efficacy of alternative medicine among general practitioners in the Netherlands
Soc Sci Med
(1990) - et al.
Is evidence for homoeopathy reproducible?
Lancet
(1994) - et al.
Assessing the quality of reports of randomized clinical trials: is blinding necessary?
Control Clin Trials
(1996) - et al.
Completeness of reporting of trials published in languages other than English: implications for conduct and reporting of systematic reviews
Lancet
(1996) The “file drawer problem” and tolerance for null results
Psychol Bull
(1979)Two pilot controlled trials of Arnica montana
Br Hom J
(1976)Clinical investigations into the actions of potencies
Br Horn J
(1971)- et al.
Homoeopathic treatment of acute childhood diarrhoea: a randomized clinical trial in Nicaragua
Br Hom J
(1993)
Metronidazole (Flagyl) and Arnica montana in the prevention of post-surgical complications, a comparative placebo controlled clinical trial
Br J Oral Maxillofacial Surg
A controlled trial
Br Hom J
An experimental double-blind study to evaluate the use of Euphrasia in preventing conjunctivis
Br Horn J
Potent placebo or potency?
Br Horn J
A double blind trial to assess the benefit of Arnica montana in acute stroke illness
Br Hom J
Unconventional medicine in the United States—prevalence, costs, and patterns of use
N Engl J Med
Complementary medicine in Europe
BM J
Organon of medicine
Ultra high dilution: physiology and physics
Homoeopathy does not work
Alt Ther
Is homoeopathy a placebo response?
Lancet
Beurteilung und Verbreitung komplementärmedizinscher Verfahren - eine Befragung von 793 Ärzten in Praxis und Klinik
Gesundh-Wes
Die Verbreitung der Homöopathie unter Ärzten in Deutschland
Forsch KomplementÄrmed
Complementary medicine—what physicians think of it: a meta-analysis
Arch Intern Med
Clinical trials of homoeopathy
BMJ
A coefficient of agreement for nominal scales
Educ Psychol Meas
The importance of quality of primary studies in producing unbiased systematic reviews
Arch Intern Med
Randomized clinical trials of acupuncture for asthma—a systematic review
Forsch Komplementämed
St John's wort for depression–an overview and meta-analysis of randomised clinical trials
BMJ
Intraclass correlations: uses in assessing rater reliability
Psychol Bull
Meta-analysis of diagnostic and screening tests
Psychol Bull
Fixed effects models
Measures of effect size for categorical data
Statistical methods for rates and proportions
Meta-analysis in clinical trials
Control Clin Trail
The existence of publication bias and risk factors for its occurrence
JAMA
Publication bias: a problem in interpreting medical data
J Roy Stat Soc
Summing up: The science of reviewing research
Modeling publication selection effects in random effects models in meta-analysis
Stat Sci
A general linear model for estimating effect size in the presence of publication bias
Psychometrika
Estimating effects size under publication bias: small sample properties and robustness of a random effects selection
model J Educ Behav Stat
A fail-safe-N for effect size in meta-analysis
J Educ Stat
Bilan de 60 observations randomisees—Arnica contre placebo dans les névralgies dentaires
Homéopathie
Aconit en dilution homéopathique et agitation post-operatoire de l'enfant
Pédiatrie
A randomised controlled trial to evaluate the effectiveness of homoeopathy in rheumatoid arthritis
Scand J Rheumatol
Action d'un traitement homeopathique sur la reprise du transit post-opératoire
Homéopathie
Bedriegt schone schijn? Een onderzoek om de gerapporteerde werking van een homeopathisch middel te objectiveren
Huisarts Wetenschap
Homoeopathic treatment of anal fissures using Nitricum acidum
Berl J Res Horn
Evaluation de l'activité antitussive du sirop Drosetux: étude en double aveugle versus placebo
Cahiers d'O R L
Cited by (808)
Systematic reviews and meta-analyses in Homeopathy: Recommendations for summarising evidence from homeopathic intervention studies (Sum-HomIS recommendations)
2023, Complementary Therapies in MedicinePrevalence of Chronic Pain Among People with Dementia: A Nationwide Study Using French Administrative Data
2023, American Journal of Geriatric PsychiatryYes to pluralistic health system, but no to homeopathy
2023, The Lancet Regional Health - Southeast AsiaHomeopathy can offer empirical insights on treatment effects in a null field
2023, Journal of Clinical Epidemiology