Homeopathy is a controversial subject, but the positions of Markun et al.  are clearly determined. They claim that ‘placebo effects seem the most obvious explanation' for homoeopathic effects, because ‘explanation models for the effectiveness of homeopathy are not supported by natural sciences and the aggregated evidence from clinical trials is unconvincing' to them. The authors lament that ‘still, many physicians continue to prescribe homeopathic treatments' and view this as ethically problematic. Here, we argue that strong a priori standpoints may bear the risk of compromising the scientific discourse and may lead to imbalanced conclusions.
The survey by Markun et al.  among ambulatory care doctors in the canton of Zurich showed 2 remarkable findings: not only certified homeopathic doctors (only 2.4% of the participants) expected specific results from homeopathic remedies, but as many as 50.4% of all homeopathy prescribing physicians did so, and only a small number (21.4%) intended to achieve only placebo effects. This means that most of the homeopathy prescribers were expecting specific effects of homeopathic remedies. Second, 53% of all participants even endorsed the use of homeopathy in certain indications, particularly where the possibilities of conventional medicine were exhausted or did not exist.
Homeopathy, as many other medical therapies, cannot be explained in all aspects - however, ‘absence of evidence is not evidence of absence' . In conventional primary care, the majority of interventions, beyond doubt, has no proper evidence base. According to BMJ Clinical Evidence, a database collecting the best available evidence on common clinical interventions, there is evidence for beneficial effects for only 11% of the treatments, but for 50% of 3,000 treatments the effectiveness is unknown . We assume that most general practitioners (GPs) are aware of this gap, cope with these uncertainties, and try to find the optimal and least harmful treatment for their patients by taking into account clinical expertise, patients' expectations and preferences, and the best external evidence. Thus, it is not astonishing that ‘… three out of four prescribing physicians (who) use homeopathy as an opportunity to meet patients' expectations without exposing them to unnecessary side-effects …', and it is remarkable that 42.4% of non-prescribers report the need for further research.
In the questionnaires by Markun et al. , the participants were asked about their agreement with explanatory models regarding the effectiveness of homeopathy (question 10). The authors mixed up several of these, in our opinion inappropriate, ‘models': the law of similars relates to the prescribing of homeopathic remedies and is not an explanatory model. Water memory, quantum physics, chaos theory, etc. are explanatory models, but in large parts they are unproven hypotheses and not relevant for daily practice. From its very beginning, homeopathy has had an empirical base and has not been deduced from such theories as the authors assume. Also, items such as ‘satisfaction of patients' expectations' or ‘strengthening of the relationship between patient and physician' are not explanatory models but can be positive and intended side-effects of a homeopathic consultation .
The authors concluded that prescribing homeopathic remedies on a placebo basis is an ethical dilemma. However, the results from clinical studies do not support the conclusion that the effect of homeopathic remedies is only a placebo effect. Results from both basic research and clinical studies on different evidence levels including meta-analyses [5,6,7] are in favor of at least a possible specific homeopathic effect (survey in ). In particular, outcome and observational studies under real-practice conditions broadly show good or at least non-inferior results. On the contrary, we consider it an ethical problem if a safe and possibly helpful treatment is withheld from a patient because it is disliked or unknown by the treating physician. In their paper, the authors missed to discuss the impact on patients' safety of withholding therapies, which are requested by patients and might be helpful. Denying complementary medicine can provoke self-treatment or treatment by not adequately qualified therapists without GPs' and specialists' expertise, especially in patients with severe diseases.
Markun et al.  declared no conflict of interest regarding their study. However, 1 of the authors (T. Rosemann) has signed an appeal of ‘Sceptics Switzerland', an organization which campaigns against complementary medicine and especially homeopathy .
We agree with the authors that medical education should provide every physician with a sound basic knowledge of homeopathy. Since January 1, 2018, it is mandatory by federal law that basic knowledge of complementary medicine is conveyed to medical students . As usual in medical education, teaching should be provided by specialists with clinical and scientific expertise in the respective field. We also agree with the majority of the participants in that more research on homeopathy is needed. Looking back on many years of experience in homeopathic research and keeping in mind the controversial discussion about homeopathy, we would appreciate further research to be conducted in close cooperation of conventional and homeopathic physicians.
All authors have made substantial contributions to the conception and revision of this work. Particularly, K.v.A. was significantly involved in the research part, H.M. in the analysis of the study, and M.R. in all parts and steps.
The authors did not receive any funding for this paper, and there is no conflict of interest (apart from us being homeopathic doctors).