5 ms·
A family member of mine has tried to participate in a clinical trial. The trial also forces him to forgo any existing medication, basically making his illness i
by kesor 11y ago
A family member of mine has tried to participate in a clinical trial. The trial also forces him to forgo any existing medication, basically making his illness intolerable. There is no indication in the trial if he is being served with real medication or a placebo, but his life has been extremely affected just by participation. Naturally he quit the trial. How would this statistical analysis address these common behaviors where ill people would just not participate in trials when certain illnesses are involved?
- KingMob 11y agoThere's a whole statistical literature on survival bias. The basic equations all require "independent and identically distributed" data loss. As soon as that's not the case (e.g., more placebo patients drop out than treatment patients), you have to compensate for that. That being said, I suspect this analysis is operating at a higher level where survivor bias isn't as relevant, because it's looking across many studies at once.