This book provides a practical, hands-on guide to designing group sequential clinical trials using the gsDesign web interface, built on the R package gsDesign. Group sequential designs allow pre-planned interim analyses during a clinical trial, offering the potential to stop early for efficacy or futility while maintaining control of the Type I error rate. The web interface enables users to generate flexible group sequential designs using only a web browser, with no need for specialized statistical software.
The book covers the full range of endpoint types supported by the interface, including time-to-event, binomial, and normal outcomes, with extensive worked examples throughout. Starting from the basics of converting a fixed design to a group sequential design, the presentation progresses through spending function families, information-based designs, and various testing configurations including two-sided, one-sided, and asymmetric designs with binding or non-binding futility bounds. Practical features such as saving and restoring designs, generating reproducible R Markdown reports, and updating bounds at the time of analysis are demonstrated in detail. An introduction to R coding with the gsDesign package extends the coverage to additional topics including conditional and predictive power, repeated confidence intervals, and sample size adaptation methods. Primarily written for biostatisticians and clinical trialists in academia and the pharmaceutical industry, the book keeps mathematical detail to a minimum, making group sequential design methods accessible to practitioners with a basic statistical background.
We publiceren alleen reviews die voldoen aan de voorwaarden voor reviews. Bekijk onze voorwaarden voor reviews.