TechGuru, did you see the slide today?!?!?!? FORTY
Post# of 148154
45 total deaths drives the p-value towards zero if the deaths are equal in placebo and LL arms.
I am using the WolframAlpha Two Binomial Distribution test linked below.
i don't know what I am doing on this WolframAlpha Two Binomial Distribution Test.
I don't know whether its one or two tailed.
link[url=https://www.wolframalpha.com/input/?i=two+binomial+distribution+test&assumption=%7B%22F%22%2C+%22ProportionDifferenceTest%22%2C+%22phat1%22%7D+-%3E%2223%2F65%22&assumption=%7B%22F%22%2C+%22ProportionDifferenceTest%22%2C+%22p0%22%7D+-%3E%220%22&assumption=%7B%22F%22%2C+%22ProportionDifferenceTest%22%2C+%22n1%22%7D+-%3E%2265%22&assumption=%7B%22F%22%2C+%22ProportionDifferenceTest%22%2C+%22n2%22%7D+-%3E%22130%22&assumption=%7B%22F%22%2C+%22ProportionDifferenceTest%22%2C+%22phat2%22%7D+-%3E%2222%2F130%22[/url]
When I run the above calculator with 23 placebo deaths and 22 LL deaths, 45 total, p=,003.
What "p-value" do you get for 45 deaths total if the deaths are equal between the Ll and placebo arms?
How do they not stop for efficacy if deaths are equal in the two arms???
That's 50% mortality reduction.
That's 30,000 lives saved over the next 60 days in the US alone if the trial doesnt run to its end.
Only reason to let the trial run at 45 deaths total, equally distributed between betweenb LL and placebo, is if manufacturing and/or distribution need 60 days to prepare for delivery?
i have no clue how it is possible to not stop for efficacy if mortality reduction is 50% and the deaths are 45 total. I wonder what possible logic there could be for continuing the trial with those interim results.