Software for
Design and Analysis of Group Sequential Clinical Trials

(Last Update: Jan 1st, 2005. By Shuangge Ma)

Shuangge Ma, Assistant Professor
Department of Epidemiology and Public Health
Yale University
New Haven, CT 06520

Micheal R. Kosorok, Professor and Chair
Department of Biostatistics
University of North Carolina
Chapel Hill, NC 27599

Introduction:
In many phase III clinical trials, it is desirable to separately assess the treatment effects on more than one primary endpoints. C++ programs are provided for the computations of sample size, critical boundaries and test statistics for group sequential clinical trials with one or two primary endpoints. Our programs cover both the design and analysis phases. The computations are proper for any trial based on normally distributed test statistics, including those in which patients give a continuous response and in survival studies. All computations are based on the methodology proposed by Kosorok et al. For reference, see the documentation. This approach resolves the problems of how to stop early when the treatment effect is clear in all endpoints and how to control the many possible error rates for concluding wrong hypotheses. This page contains the documentations and the latest version of the software.

Special Notes from Shuangge:
The software (Windows version and Unix version) have been extensively tested with simulated data and real clinical data provided by the Statistical Data Analysis Center of Department of Biostatistics and Medical Informatics, University of Wisconsin. However, our empirical studies are by no means exhaustive. So if any abnormal numerical results are encountered during your study (using this software), consultations with us are suggested.
Please kindly inform us, if you see any error of the software. You are welcome to download the software and build improved version. Acknowledgement of us will be greatly appreciated. You are always welcome to collaborate with us to improve the software.

Documentation:

· Design and Analysis of Group Sequential Clinical Trials With Multiple Primary Endpoints. by M. R. Kosorok, Y. Shi, D. L. DeMets (2004). Biometrics, 60:134-145.

· Software for Design and Analysis of Group Sequential Clinical Trials With Multiple Primary Endpoints. by S. Ma, M. R. Kosorok, T. D. Cook (2003). Department of Biostatistics and Medical Informatics Technical Report 177. University of Wisconsin-Madison.

· Design and Analysis of Group Sequential Clinical Trials With Multiple Primary Endpoints. (ppt file) by S. Ma. Presented at ENAR 2004 (contributed session), Pittsburgh PA.

· Design and Analysis of Group Sequential Clinical Trials With Multiple Primary Endpoints. (ppt file) by S. Ma. Poster Presented at MBSW 2004, Muncie IN.

· Read Me.txt by S. Ma.

Source Code:

Unix Version (Now available)

· Download Design of Clinical Trial (Source Code and Makefile)

· Download Analysis of Clinical Trial (Source Code and Makefile)

Windows Version (Now available)

· Download Design of Clinical Trial (Source Code)

· Download Analysis of Clinical Trial (Source Code)
Download Sample Dataset:

Sample Files for Normally Distributed Data:

· Download Information File 1

· Download Information File 2

· Download Sample Data File
Sample Files for Time-to-event Type Data:

· Download Information File 1

· Download Information File 2

· Download Sample Data File


Acknowledgment:
Shuangge Ma would like to thank Ms. Mingyu Shi (Department of Mechanical Engineering, University of California-Berkeley) for consultations and discussions.


  For questions, comments or suggestions, email shuangge.ma@yale.edu