What would you need in your study to make it… for how long would you need to do it for to make it valid?
It depends very much what you’re looking for. So if you’re looking at disease modification you generally need 300 patients.
Over what time period?
As long as you want, and it’s a big problem with MS trials. So MS normally takes decades to do its worst, whereas your longest study is three years. Conversely, if you do a study that goes on for 20 years, you generally find… the longest blinded study in neurology is a five-year study, and that’s for Parkinson’s disease, and that’s just not long enough. However, in other areas we know that if you ask a question over, say, ten years, by the time you’ve answered the question you realise you hadn’t asked the right question in the first place. So it’s rather as if you read a review of cars and so you discover that people who bought a Ford Focus in 2007, it lasted longer than a VW Astra, for example. But you can’t buy that car now, so it’s meaningless.
But you’ve got to start somewhere.
I mean even if you think it’s going to be meaningless, you can’t discount something unless you study it, right?
Exactly. I mean, for example, the first time I learnt this was a study of blood pressure in older people, we know it’s a good thing to treat blood pressure, we believe it is. They had a ten-year study. Now, as it happens, lowering the blood pressure did not save life. Huh. Why? Because the dose of the drug they were using was too high, it caused side effects. Now, we later realised that a dose a quarter of what they used is much safer. So this is a ten-year study, and I hate to think what that cost, became meaningless. Now I think what we should be doing is everybody with MS is monitored and measured and we build up a picture of what happens in MS.
Isn’t there an MS Register?
Yeah, but not everyone’s on it, and you can’t force people to be on a register. And we all feel a bit sensitive about being on databases, I don’t want them to know about me.