Bayes Factor Binary

Single-arm binary trial monitoring with Bayes factors

This software implements the clinical trial monitoring method proposed by Valen Johnson and John Cook in "Bayesian Design of Single-Arm Phase II Clinical Trials with Continuous Monitoring," Clinical Trials 2009; 6(3):217-26. (preprint) It is the binary analog of the Bayes Factor TTE (time-to-event) software available here.

This method monitors a single-arm trial with binary outcomes with the option of stopping for inferiority or superiority. The method uses Bayes factors with non-local priors. The software supports continuous monitoring or monitoring in patient cohorts. It can produce a table of stopping boundaries as well as simulation results.

The software is easy to use because there are relatively few design parameters, and these parameters are intuitively meaningful. Most importantly, the decision whether to stop or continue a trial is based on the posterior probability of the null and alternative hypotheses. You could specify, for example, that a trial should stop for inferiority if there is a 90% probability in favor of the null hypothesis. This makes trials easier to design and makes the results easier to interpret.

BayesFactorBinary.exe is a command line program for Microsoft Windows 7. To install the software, unzip the file included here to an empty directory. The software requires the Microsoft .NET Framework version 4.0.

To use the software, open a command window and navigate to the directory containing the executable file. Running the program with no arguments will print instructions for using the software. For more information, see the BayesFactorBinary user's guide.

This software was developed by Xiudong (Jennifer) Lei, Ph.D.