Statalist


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

RE: st: equivalence limit


From   "Joseph Coveney" <jcoveney@bigplanet.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: equivalence limit
Date   Mon, 17 Aug 2009 00:49:21 +0900

Ricardo Ovaldia wrote:

I am still trying to figure how to do an equivalence test in Stata.
All I have is a 2x2 table (Treatment x outcome). I can use -cs- or -rdci-
to get the CI, but how do I compute the equivalence limit assuming delta=25%,
for example?

Also, can -pkequiv- be trick into doing this analysis?

--------------------------------------------------------------------------------

If your delta is 25% for a risk difference (difference in proportions), then you
can reject the null hypothesis of nonequivalence in favor of the alternative
hypothesis of therapeutic equivalence if the confidence interval (CI) from -cs-
(Wald) or -rdci- (various other methods) lies between -0.25 and +0.25.  

For equivalence (as opposed to "noninferiority"), a 90% CI is conventionally
used.  This corresponds to a two, one-sided test (TOST) method with alpha =
0.05.

So, set -level- at 90% for -cs- or -rdci- (whichever command you choose), and if
the displayed CI is between the limits that you set for equivalence (e.g., -0.25
to +0.25, in your example), then you can declare bioequivalence.

Joseph Coveney



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



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