Thanks Mojo for your reply,
The question about how to deal with patients who step out of Sunrise still remains unclear. I called them “censored”, but that is not exactly the right word. I know that there are patients who have voluntarily stepped out of Sunrise for whichever reason (not necessarily because Bavi is no good).
Let’s suppose that 582, as you said, is the magic number. And let’s also suppose that PPHM knows that a number “D” of patients have dropped out Sunrise.
a.)Will all known dropped outs be overwritten as enrolled and be replaced by new enrolled patients, so “582” remains regardless of the dropped outs?… Or
b.)The number of Sunrise participants will consistently decrease as dropped outs increase and therefore it would always be (582 – D) participants. In that case it wouldn’t make sense to apply the 33% (for the 1st look-in) to “582” but to (582 – D). i.e. (suppose) there is a 50% dropped outs it would take ages to arrive to 582 x 0.33
If (b) is the case then the model I made is not taking into account the number of dropped outs and therefore giving a greater number of eventings than in reality. Then dropped outs should be taken into account in the equation which would mean that all along the enrolled patients should be reduced by a percentage of dropped outs (15%?)
It would be interesting to know if it is a) or b). I wish it is a).
I would appreciate if you or some other expert on the board has an answer.
