Hanson at Overcoming Bias posts the graph below from the British Medical Journal. It says they looked at 2500 treatments, so I suspect this isn't just medicines. I'd like to see the list to determine whether it only included standard treatments like antibiotics for pneumonia or stents for coronary artery blockages, or whether it includes things like homeopathy and massage therapy.
I don't know what to make of the Terra Incognita of the 46% of "unknown effectiveness." I want to see the details.