"To measure the p value, are we only comparing the events in each arm, or do the surviving patients in each arm attain p value "credit" for that arm? I was under the impression that when the 180th event is reached, any surviving patients will be counted for however many days post-randomization they are alive. Is this in fact the case, or do all surviving patients get thrown out of the p value equation when the 180th death occurs?"
Yes, we are only comparing the events in each arm and surviving patients do not get "credit" directly. However, they do count indirectly because the total number of patients enrolled in each arm determines the basic probability set which forms the basis of our calculations based on the null hypothesis.
For instance, if we have 501 patients in the trial and 334 of these are in the provenge arm while only 167 are in the Placebo arm, then the null hypothesis would assume that the probability of an event being due to a Provenge patient is 334/501 while the probability of an event being due to a Placebo patient is 167/501. These elemental probability assumptions show up indirectly in our calculation of the p value (for instance in a binomial formulation). Therefore, in an indirect sense, we are taking into consideration the total number of patients in each arm, and thus we are accounting for the surviving patients as well...
