Dear Piersente,
I agree with you in that it seems to me that the key point is 'what should we believe the study showed'. This comes down to the analyses and ultimately the data. So what I would like to know is does changing the endpoint have an effect on the interpretation. In the NEJM study I gave as an example some posts back changing the analysis did change the interpretation. The FDA advisory panel unanimously rejected the analysis that was eventually published in the NEJM. In the very same issue the FDA published a letter saying (effectively) do not believe the results of this study. Whether the NEJM is right or the FDA is right (and I side with the latter) this is truly extraordinary.
Regards
Stephen
Stephen Senn
Professor of Statistics
School of Mathematics and Statistics
Direct line: +44 (0)141 330 5141
Fax: +44 (0)141 330 4814
Private Webpage: http://www.senns.demon.co.uk/home.html
University of Glasgow
15 University Gardens
Glasgow G12 8QW
The University of Glasgow, charity number SC004401
________________________________________
From: Piersante Sestini [[log in to unmask]]
Sent: 22 August 2011 02:40
To: Stephen Senn
Cc: [log in to unmask]
Subject: Re: Ezetimibe/Simvastatin
On 21/08/2011 20.39, Stephen Senn wrote:
> Dear Ash,
> Agreed that there is a lot to read. Has anybody done an analysis of the other endpoint (the one that was initially primary) to see if the message is different. My opinion is that if the two differ but medical opinion is divided as to which is important then it is more or less irrelevant as to which was pre-specified. There is a problem with the trial.
>
As I wrote in a previous message (I am unsure whether it went through,
since I had no confirmation from the listserv), my current understanding
is that the change in "primary outcome" was proposed, but eventually
dropped.
Indeed the article on the NEJM reports the results of the "primary
outcome" as the mean of several points as originally planned, and not as
a single point as for the proposed change.
Furthermore, the data are also presented separately for each of the
points, and in none there is a noticeable effect, making it unlikely
that the change had been proposed based on the knowledge of a better
result.
As I explained in the message, I don't think that this makes the sponsor
freed from all the charges, though
regards,
Piersante Sestini
|