How Much Anecdotal Evidence?

Say I give a homeopathic remedy to a patient who complains of, among other things, rheumatoid arthritis. Her rheumatoid symptoms improve over six weeks. Can I really say that the remedy caused this improvement?

What if she’s had rheumatoid arthritis for many years and had tried conventional treatments and some other alternative healing modalities without much effect, but felt better after taking a homeopathic remedy?

What if, in addition to the rheumatoid symptoms, after taking the remedy she also had improvement with some other problematic but non-pathological things, such as that she was less bothered by something that used to cause her stress, and that a disturbing recurring dream changed?

What if she took the remedy and felt better for a time, then started to worsen again and took the remedy again and improved a second time?

What if after those six weeks her rheumatoid arthritis symptoms were gone altogether, requiring no more remedies or conventional medications? (Not the kind of result you get from a placebo effect, although I’m open to the possibility.)

I’m not intending to put this out there as evidence for homeopathy. I understand that this is hypothetical and anecdotal. (And for the record, I’m not saying all my patients have this kind of result either.) I am saying, what if you experienced this? Would that not convince you personally that homeopathy “works?” What level of personal evidence do I need to have before it’s not just my “belief?”

Now what if I observe similar kinds of things on a regular basis?

And then what if other homeopaths tell me they have similar experiences?

The next question is how much of this “anecdotal evidence” do I need to accumulate before it raises to the level of scientific evidence? I could try to create a randomized double-blind controlled trial to “prove” this. But every patient might get a different remedy. Some might not need the repetition, and some might need more. The levels of improvement would vary. And the other “symptoms” of each patient that improved would be different in each patient. It would not be a successful “Remedy X is effective in reducing rheumatoid arthritis symptoms” kind of study. It would have to be: “homeopathic treatment in general produces a general improvement in health that causes a significant number of patients with rheumatoid arthritis to experience improvement.”

Apart from the epistemic questions, there is also a practical question here for homeopaths: how many “clues” do we need before we feel positive that the remedy has worked for a patient. Sometimes it’s overwhelmingly obvious. But when I only have the first two “what if”s I listed above, I remain skeptical. Many of my patients are doing other things for their health as well as homeopathy. For me, it’s mainly the “pattern shift” and “improvement on repetition” clues that make me feel sure. Only truly holistic influences cause the pattern shift, in my experience. And the repetitive effect is just rational—more evidence increases certainty.

However, I should point out, with a patient my goal is not to prove myself right or accumulate data. The goal is the health of the patient. If the patient is better and I'm not trying to decide whether to change to a different remedy or not, the certainty factor is just a note in the case. Just a reminder that the questions I'm asking in this blog are secondary…