11

guidance although I know there has been some statistical discussions about this, is that if

12

you just focus on the upper bound of the 95% confidence interval and you don’t take into

13

account the rest of the confidence interval and where the point estimate falls, which is our

14

best guess of what the treatment is effected based on that sample data that you may be

15

missing something and so, for example the guidance says an example that if a point

16

estimate was 1.5, even if your upper bound was 1.8 and your lower bound was above 1,

17

that wouldn’t be as reassuring as if your point estimate fell around unity with the upper

18

bound of 1.8.

4

events occurred in the clinical programs and those were the analyses that FDA requested

5

of the company. For this reassuring point estimate to prevent myself from getting into

6

trouble with the statisticians, I'll ask them perhaps to - if they have a comment on that.

23

DR. JOFFE: Close to one.

24

DR. FLEGAL: Close to one, but no specific definition?

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

1

thought would most likely represent an acute important cardiovascular event as

2

reported by the investigator.

9

DR. FLEGAL: Yes, the guidance uses the term reassuring point estimate.

10

DR. JOFFE: Maybe I will start speaking about that. The point with the

20

have upper bound of 1.3 with a reassuring point estimate, then a post marketing trial

21

would generally not be necessary. So in that context, the reassuring point estimate would

22

be something below one I suppose would be reassuring, about...

3

87

Now that was done just looking at preferred terms without knowing what

19

DR. FLEGAL: Can I follow up on that because the slide says that if you

7

8

DR. MELE: I’m not quite sure at what point in time we said reassuring

the point estimate. Was it in Hilton’s presentation?