Antimicrobial resistance (AMR) is already a big problem in medicine, causing more than 700,000 deaths per year worldwide. It is expected to become the main cause of death within 30 years with 10 million deaths each year. About 75% of all antibiotics is used in respiratory tract infections (RTI), but we need to treat more than 4,000 patients with antibiotics to prevent one serious complication. Severe adverse effects of antibiotics outweigh this benefit and AMR comes on top of this trade-off of benefit and harm. Despite the clear harm of antibiotics in RTI, authoritative bodies in medicine claim that antibiotics should be preferred over homeopathy. They state that there is “insufficient reliable proof” for homeopathy in RTI. Homeopathy could even be harmful because it would “cause delay of effective conventional treatment”. A closer look at the existing evidence shows the absurdity of such claims. There is no proof of direct or indirect harm of homeopathy. Shang/Egger's meta-analysis of homeopathy in eight RTI trials shows “substantial beneficial effect” (odds ratio (OR) 0.36, 95% CI 0.26-0.50) without proof of quality bias. Other meta-analysis show positive - but not statistically significant - effect after post-hoc selection of subsets of one or two trials. Trying to confirm the placebo-hypothesis for homeopathy by creating small subsets with statistically non-significant positive results is unscientific. Like the philosopher Popper stated: “No number of sightings of white swans can prove the theory that all swans are white [homeopathy is placebo]. The sighting of just one black one may disprove it”. Shang/Egger's finding is obviously a black swan. This finding is corroborated by meta-analysis of all trials for all indications except the indication ‘muscle soreness after marathon’; even if we confine ourselves to high quality trials with greater than median sample size. The proof for homeopathy can therefore be compared with the best proof in conventional medicine. Many individual patients experience no relief from medicines despite RCT proof. The selection of homeopathic medicines for individual patients is a reproducible solution for such patients–using prognostic factors (symptoms)–like in the diagnostic process. Medical science should further develop in that direction. Prognostic factor research can be applied to develop treatment algorithms for improved and more accessible application of homeopathy; it can also be used for better selection of patients that need antibiotics.Le texte complet de cet article est disponible en PDF.
Keywords : Homeopathy, Antimicrobial resistance, Randomised Controlled Trials, Plausibility, Statistical significance