Adaptive Randomization

This software simulates randomized trials in which the randomization probabilities adapt in response to the outcome data. More patients are treated with the better treatment while retaining the benefits of randomization. The software supports both binary and time-to-event outcomes. Numerous design options are supported.

The software supports up to 10 arms. Stopping rules may be based on a maximum patient accrual or maximum trial length. Optionally one may specify a minimum number of patients in the trial. Also, one may specify the number of patients to randomize fairly before adaptive randomization begins. Stopping rules are based on posterior probabilities.

The trial design contains a tuning parameter to control the degree to which the randomization probabilities respond to data. If this parameter is set to zero, the "adaptive" randomization is actually equal randomization. The larger the value of this parameter, the more the randomization favors what appears to be the better treatment. See the report [1] examining the role of this parameter on operating characteristics and giving guidance for selecting its value. See also [2] for a comparison of this tuning parameter to other methods of controlling the randomization probabilities.

The Adaptive Randomization user's guide is available here and also included as part of the download. See [3] for a study of the operating characteristics of adaptively randomized clinical trials.

See also our Predictive Probability software interim analysis of randomized trials.

See [5] for a description of the algorithm used for calculating time-to-event randomization probabilities. Algorithms for calculating the other randomization probabilities are described in [6].

For question concerning this software please contact J. Kyle Wathen.

Odis Wooten and J. Kyle Wathen developed the user interface using C# and the Microsoft .NET framework version 2.0.

J. Kyle Wathen implemented the simulation kernel using Visual C++.

Thomas Liu and James Martin assisted in writing unit tests for the project.

References

[1] John D. Cook, Understanding the Exponential Tuning Parameter in Adaptively Randomized Trials (2006). UT MD Anderson Cancer Center Department of Biostatistics Working Paper Series. Working Paper 27.

[2] John D. Cook, Comparing Methods of Tuning Adaptively Randomized Trials (2007). UT MD Anderson Cancer Center Department of Biostatistics Working Paper Series. Working Paper 32.

[3] J. Kyle Wathen and John D. Cook. Power and bias in adaptively randomized clinical trials (2006). Technical Report UTMDABTR-002-06.

[4] Peter F. Thall and Jay K. Wathen, Practical Bayesian Adaptive Randomization in Clinical Trials (December 2006). UT MD Anderson Cancer Center Department of Biostatistics Working Paper Series. Working Paper 31.

[5] John D. Cook. Numerical evaluation of gamma inequalities (2006). Technical Report UTMDABTR-001-06.

[6] John D. Cook Numerical Computation of Stochastic Inequality Probabilities (2003) Technical report UTMDABTR-008-03