I've posted a link above, from the BMJ, who said that 81 trials yielded positive results. I don't have any trials, I just prescribe. Perhaps if someone would come up with a protocol that would be acceptable to homeopaths and conventional medicine before the test, it would be worth doing. All that happens otherwise is that people pick holes in the method rather than looking at the results.
Say, take 100 asthma patients, 50 for a control and 50 for treatment. How would the test go?
Firstly, all 100 would have to be interviewed by a homeopath. That would be a minimum of an hour each, possibly 2. Then the homeopaths specify which remedy for each patient. Fifty patients get a placebo, fifty get the remedies. Two weeks later, the procedure is repeated, except that the consultations would only be 15 - 20 minutes. Another remedy is prescribed, possibly the same one for the patient, possibly a different one, depending on the reaction the patient had to the first one. The problem here is that the homeopaths will know that those patients who had a reaction to their first remedy are not on placebos, so bias will creep in and critics will claim it is not a double blind test.
Or you can look at published data on the Spanish 'flu, for example, but when I posted that it was dissed because it was in a homeopathic journal. Saying that less than 1% of 27,000 cases died (compared to 30% mortality for conventional treatment). Well, you'd hardly expect the JAMA to publish such appalling statistics against themselves, would you?
Expand