X hits on this document

554 views

0 shares

0 downloads

0 comments

152 / 170

12

So, I think it was good that they did some analyses that try to take into

13

account the fact that you had no events, tried to use that information in some way and

2

commenting on earlier, I think in a case like this with not that many events, it’s very

3

important to look at a sensitivity analysis and to do it multiple ways. So I saw this as a

4

definite plus. Like I said, the only thing I might do in addition would be this Barnard’s

5

Test, which can sometimes do - of course that’s a test, it’s not really for confidence

6

intervals. I think the issues around like what analysis to do are kind of tricky, because if

7

you condition on the total number of events, like I was saying, in some sense it was very

8

natural, then you can’t include trials with no events, because you just won’t get any

9

information out of those. in a way you see a trial with no events and you say, well, that’s

10

evidence that there may not be a problem, I mean or it’s evidence we had such a low risk

11

population that, you know, of course you didn’t have a chance to see a problem.

14

then some other analyses where you throw out the trial with no events. So, I thought it

15

was very good seeing the analyses done many different ways and seeing similar results,

16

what you don’t see similar or not too similar results on are the custom MACE versus the

17

more general MACE, I mean because the custom MACE has those - seems to show

18

benefit where the more general MACE seems to show basically no effect. I don’t exactly

19

know what to make of that, I don’t know how seriously to take those CPK elevations. In

20

general, I thought it was very good to do the analyses many different ways and mostly

21

getting similar results.

23

DR. TEERLINK: So, I’m not a statistician. So I think when I try to hear

24

about sensitivity analyses, I just have the clinical trial experience where we have a whole

1

DR. PROSCHAN: Right, yeah, no, I mean this is what I was

151

22

DR. BURMAN: Thank you. John.

Scribes, LLC Toll Free 1-800-675-8846 www.scribesllc.com

Document info
Document views554
Page views554
Page last viewedSun Dec 11 04:44:34 UTC 2016
Pages170
Paragraphs7975
Words54271

Comments