Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: re: non-inferiority trials


From   Philip Jones <pjones.statalist@gmail.com>
To   Statalist <statalist@hsphsun2.harvard.edu>
Subject   st: re: non-inferiority trials
Date   Fri, 5 Oct 2012 12:37:34 -0400

Andrzej asked about non-inferiority trials:

On Thursday, October 4, 2012, Niemierko, Andrzej wrote:

> Are there any user-written Stata programs to calculate:
> 1. the confidence interval for the difference of median survival times, and
> 2. p values in non-inferiority trials?
>
> Thanks,
> Andrzej


Regarding #2, how can one calculate a P value for a non-inferiority trial?
I believe there is a logical error when discussing this, as explained below.

Null hypothesis in a non-inferiority trial: there is a difference between
groups.

Alternate hypothesis: there is no difference between groups, up to a
certain acceptable difference (the "delta"). Note this is a one-sided
check, as opposed to an equivalence trial.

Therefore, the P value will be dependent on your delta. If you choose a
ridiculously high, implausible delta, you may have a significant P value
but clear evidence of a difference between groups. Better to use a
one-sided confidence interval approach as recommended in the CONSORT
guidelines at
http://www.consort-statement.org/extensions/designs/non-inferiority-and-equivalence-trials/

Phil
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index