11

make any good statistical point, but we don’t want to keep setting the bar higher and

12

higher, we started out with 1.8 as encompassing the range and then we say, well, we want

13

to have below 1.8 in all the subgroups or now we want to say, well, 1.8 if we had had

14

more events, but I think we need to be careful about what we are - our goal is here is

15

again not to show that this drug has a cardiovascular benefit, but to try to rule out major

16

harm. So I think we need to be cautious about trying to impose too many conditions and

17

really actually ending up making it more difficult than the guidance suggests to meet this

18

criteria.

23

issue on question number four, and in my summary to see if this is accurate. Multiple

24

statistical methods were used, please discuss the adequacy of these methods for

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

1

sensitivity analyses per se, to say oh does it matter on, is the result sensitive to the

2

different type of statistical test you implement, but rather how sensitive is it to possible

3

random changes in patients from one group to the other, because it’s so small, the

4

number.

9

DR. ALXEXANDER: Thank you.

10

DR. FLEGAL: I’m not a statistician either, but I may not be going to

21

(No response.)

22

All right. Then, let me, since we have time, let me open to just review the

155

8

groups.

5

DR. BURMAN: John, excuse me. What was the trial you just quoted, I

19

20

# DR. BURMAN: Thank you. Other comments on this statistical question?

# Question No. 4?

6

7

didn’t understand you?

DR. TEERLINK: The Vesnarinone trials is one of them in terms of the