BookmarkSubscribeRSS Feed
rmacarthur
Pyrite | Level 9

Hi SAS Friends, 

 

Am looking for examples of how to use the PROC SEQDESIGN procedure when adding an interim FUTILITY analysis only to a study design.  All examples seen so far in the SAS PROC SEQDESIGN Manual and meeting papers include an interim analysis for efficacy with a futility analysis added.  However none with futility alone.

 

For example, the following code:

 
proc seqdesign pss stopprob errspend ;
onesidedtwostage: design nstages=2 info=cum(.6 1) alt=upper
alpha=0.025 beta=.20 
method(reject)=errfuncobf                                                            /* INTERIM EFFICACY BOUNDARY */
METHOD(ACCEPT)=ERRFUNCOBF STOP=BOTH ;              /* INTERIM FUTILITY  BOUNDARY */
samplesize model=twosamplefreq(nullprop=.5 prop=.70 ref=nullprop test=prop);
title1 "Stopping early for futility " ;
run;

 

The above code generates efficacy and futility boundaries,  however if the efficacy line is deactivated (as  /*method(reject)=errfuncobf */) then the code will not run and the following error is produced:

 

ERROR: Methods for all boundaries must be specified in the DESIGN statement if more than one
stage is specified.

 

Any suggestions how produce an interim futility analysis  only, using  the above code, or any SAS code, would be greatly appreciated.

 

Thank you !

 

  

1 REPLY 1
Rick_SAS
SAS Super FREQ

I do not use PROC SEQDESIGN myself, but I notice in the doc that the syntax supports

STOP=ACCEPT
STOP=REJECT

Is that useful?

SAS Innovate 2025: Call for Content

Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!

Submit your idea!

What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 1 reply
  • 2001 views
  • 0 likes
  • 2 in conversation