MS liberation: in defence of Saskatchewan

You all know I’m sceptical of Paolo Zamboni’s vascular hypothesis about multiple sclerosis, so allow me to express support for Saskatchewan health minister Don McMorris, who is defending his province’s right to fund research into the idea. I didn’t like Saskatchewan’s politically-driven funding decision, but I don’t think there is any reason to believe that the funding cannot be useful even if the hypothesis is true. That seems to be what the MS Society is suggesting when it argues that Saskatchewan can’t possibly muster up a large enough sample to conduct a useful test of the hypothesis. From today’s Globe:

Yves Savoie, president of the MS Society of Canada, said a true clinical trial must be conducted at more than one institution and in more than one province. Because MS is so variable, “it will require well over 1,000 participants that will be recruited through a number of centres,” he said. “A single province or a single site would simply not be a way to get to the definitive answers that we all want.”

Logic and experience suggest an obvious rejoinder: a single trial, however large, won’t be the way to get an answer either. Pure statistical power is a good thing, but so is experimenter diversity. If there is any merit in the Zamboni hypothesis, the scientific community is likely to arrive at a consensus about it on the basis of many different kinds of tests, some of them modest in scale, most of them performed independently. There is danger in Savoie’s apparent insistence on a One True Trial and in his gratuitous, pre-emptive criticism of what Saskatchewan is doing: he is inadvertently encouraging the Zamboni believers’ cherished conviction that they are being ganged up on by a conspiratorial clique that desires a monopoly on truth.

And if the MS Society thinks a single, giant experimentum crucis is advisable, it would be fair to ask why it isn’t advocating one, instead of funding CCSVI research in dribs and drabs. The fact is that piecemeal accumulation is the usual means of accumulating scientific knowledge. There’s no unitary global Science Court where hypotheses can be hauled in for exoneration or hanging.

We don’t really need a trial with a sample size of thousands to confidently confirm or disconfirm the most basic claim of Dr. Zamboni: that an MS diagnosis can conceivably be verified, or falsified, from nothing more than medical images of a patient’s head and neck veins. As I’ve suggested before, we could arrive at a good initial answer to that question quickly and inexpensively. There is no sense in going ahead with inquiries into causality, or into the effectiveness of any particular therapeutic regime, until we have first established that yes, there is such a thing as “chronic cerebrospinal venous insufficiency”. Trap your unicorn first, then study its anatomy.