[R] group sequential and adaptive designs

Frank E Harrell Jr fharrell at virginia.edu
Thu Jul 10 15:43:30 CEST 2003


On Thu, 10 Jul 2003 14:13:20 +0200 (MEST)
Marc Vandemeulebroecke <vandemem at gmx.de> wrote:

> Hello R users,
> 
> I am looking for R (or S) code related to group sequential or adaptive
> designs for clinical trials. (The most prominent examples are the designs of
> Pocock or O'Brien/Fleming, the alpha-spending function approach, or Fisher's
> combination test and the inverse normal method.) I am particularly interested in
> the calculation of the critical boundaries, the handling of spending functions
> and, most of all, power calculations. All I could find were announcements of
> courses or postings related to the S-Plus SeqTrial module. No code. Can
> anyone give a hint?
> 
> Many thanks in advance,
> Marc
> 

If you get the U. of Wisconsin Biostatistics program ld98 executable, which implements several spending functions in the Lan DeMets paradigm, you can use the S function ldBands and its plot and summary methods, which call ld98.  ldBands is in the Hmisc package at http://hesweb1.med.virginia.edu/biostat/s/Hmisc.html

ldBands works under Linux/Unix but has not been tested under Windows yet.
Get ld98 and its documentation from http://www.medsch.wisc.edu/landemets

ldBands makes ld98 easier to use.  Examples show how to do power calculations.
---
Frank E Harrell Jr              Prof. of Biostatistics & Statistics
Div. of Biostatistics & Epidem. Dept. of Health Evaluation Sciences
U. Virginia School of Medicine  http://hesweb1.med.virginia.edu/biostat




More information about the R-help mailing list