Herbal remedies and standards of proof

When I was a boy scout, one of our favourite campfire skits involved a Chinese medicine man and a patient. The patient has an ailment for which the medicine man has a cure, which is a bundle of leaves and twigs. "Boil this in a big pot of water for hours until one cup of liquid remains; drink it and your troubles will be over". This was traditional medicine in our grandparents' time and we thought it was funny.

Last week, I met my old scout troop leader and, at one point, our conversation turned on cholesterol. He had been taking pills for a long time to keep his high cholesterol level down. Then a friend recommended a herbal cure: a bundle of leaves and twigs, to be boiled in a big pot of water for 2 hours, together with dried figs (optional) for flavour. He drank this for two weeks and his cholesterol came down to 4 and has remained so for months without any further treatment. So utterly impressed is he that he has been talking to doctors to conduct medical research on this plant, and cannot understand their total lack of interest.

I explained to my friend that scientific research is very complicated. One is supposed to assemble a large sample of people, divide them into two evenly matched groups, give the treatment to one group, and a bogus treatment to the other. The people in the experiment are not to know which group they have been placed into. The people administering the test should not know either. The volunteers get treatment A or B and only the designer of the experiment know what is in A and B. Then there are tough rules governing the use of humans as test subjects. If you use animals, the rules are almost as tough.

If these conditions had prevailed in the past, Louis Pasteur would not have been able to develop a cure for rabies. Having successfully developed vaccines against various fatal diseases of animals, Pasteur developed what he believed would work against rabies. No one had yet been known to survive rabies. But there was no way Pasteur could have assembled a test group of infected people, some of whom would be given the treatment and some not. So he waited until one victim was brought to him. He administered treatment. The victim survived and Pasteur became a hero. The next victim also recovered. And the next. After that, Pasteur's treatment became the standard treatment for rabies, to the regret of statisticians who feel this is a rotten example of experimental procedure.

I asked my friend to show me the plant. It turned out to be a dark-coloured form of Alternanthera sessilis, a weed of ditches and wet places in many parts of the world. The amount of stems and leaves he gave me was enough to fill a pot 25 cm diameter, to a depth of 7 cm. I added water and boiled it for two hours, topping up with water as necessary. After one hour, I added 200 gm of dried figs (labelled 'dried dates' on the packet, sold in Chinese herbal shops). I ended up with 3 cups of dark-coloured liquid. I drank one cup last night and put the rest in the fridge. I found the drink pleasant but too sweet. 100 gm of dried figs would have been enough.

I have no cholesterol problem so this test was merely to satisfy myself that the concoction is pleasant to drink and harmless (I think I still make sense as I type this). So now I am ready to test this on volunteers with high cholesterol.