happyscientist wrote:
A single study, even if published in a peer reviewed journal, means nothing. It takes a lot of studies to be able to determine a trend. As a former professor of mine would say, once is a fluke, twice is a coincidence, and you can't begin to find a trend until the third consecutive time. I don't feel like finding the study in Lancet, but was it double blinded? Did they compare randomized performances of each individual with EPO vs placebo? N=48 is not enough for a clinical endpoint study.
I don't think you know 48 isn't enough until you try more. E.g. the people here claiming it's underpowered are really only speculating that it is. The other alternative is that there is really no effect.
What I don't understand is the protocol. T hey divided the 48 into two groups, gave half of them EPO for 8 weeks, then sent them up the mountain.
I don't get that. There's no baseline, as described. It seems the right thing to do would be to send all 48 up the mountain clean. Then inject them all, half with EPO, half with placebo. Then send them up a second time. And compare the differences in all the riders between the baseline run and the test run.
Otherwise you'd have no way of knowing if your division actually selected an underperforming group that was actually brought back to equal performance by EPO.