Skip to main content Deutsch


Institute of Medical Statistics

survmixer: Design of clinical trials with survival endpoints based on binary response

by Marta Bofill Roig

Sample size and effect size calculations for survival endpoints based on mixture survival-by-response model. The methods implemented can be found in:
Bofill Roig, M., Shen, Y., & Gómez Melis, G. (2021). Design of phase III trials with long‐term survival outcomes based on short-term binary results. Statistics in Medicine 40(18), 4122–4135.

eselect: R package for adaptive trial designs with endpoint selection and sample size reassessment

by Marta Bofill Roig

Endpoint selection and sample size reassessment for multiple binary endpoints based on blinded and/or unblinded data. Trial design that allows an adaptive modification of the primary endpoint based on blinded information obtained at an interim analysis. The decision rule chooses the endpoint with the lower estimated required sample size. Additionally, the sample size is reassessed using the estimated event probabilities and correlation between endpoints. The implemented design is proposed in:
Bofill Roig, M., Gómez Melis, G., Posch, M. & Koenig, F. (2022) Adaptive clinical trial designs with blinded selection of binary composite endpoints and sample size reassessment. Biostatistics (in press).

NCC: Design and analysis of platform trials with non-concurrent controls

by Pavla Krotka, Marta Bofill Roig, Katharina Hees, Peter Jacko, Dominic Magirr

Design and analysis of flexible platform trials with non-concurrent controls. Functions for data generation, analysis, visualization and running simulation studies are provided. The implemented analysis methods are described in:
Bofill Roig M., Krotka P., Burman C.-F., Glimm E., Gold S.M., Hees K., Jacko P., Koenig F., Magirr D., Mesenbrink P., Viele K., Posch M. (2022) On Model-Based Time Trend Adjustments in Platform Trials with Non-Concurrent Controls. BMC Medical Research Methodology 22:228.
Saville B.R., Berry D.A., Berry N.S., Viele K., Berry S.M. (2022) The Bayesian Time Machine: Accounting for temporal drift in multi-arm platform trials. Clinical Trials 19(5),490–501.
Schmidli, H et al. (2014).

Omnibus Package

by Andreas Futschik, Thomas Taus, Sonja Zehetmayer

An omnibus test for the global null hypothesis: Global hypothesis tests are a useful tool in the context of, e.g., clinical trials, genetic studies or meta analyses, when researchers are not interested in testing individual hypotheses, but in testing whether none of the hypotheses is false. There are several possibilities how to test the global null hypothesis when the individual null hypotheses are independent. If it is assumed that many of the individual null hypotheses are false, combinations tests have been recommended to maximise power. If, however, it is assumed that only one or a few null hypotheses are false, global tests based on individual test statistics are more powerful (e.g., Bonferroni or Simes test). However, usually there is no a-priori knowledge on the number of false individual null hypotheses. We therefore propose an omnibus test based on the combination of p-values. We show that this test yields an impressive overall performance. The proposed method is implemented in the R-package omnibus.

subtee: Subgroup Treatment Effect Estimation in Clinical Trials

by Nicolas Ballarini

Naive and adjusted treatment effect estimation for subgroups. Model averaging (Bornkamp et al., 2016) and bagging (Rosenkranz, 2016) are proposed to address the problem of selection bias in treatment effect estimates for subgroups. The package can be used for all commonly encountered type of outcomes in clinical trials (continuous, binary, survival, count). Additional functions are provided to build the subgroup variables to be used and to plot the results using forest plots.

SubgrPlots: Graphical Displays for Subgroup Analysis in Clinical Trials

by Nicolas Ballarini

Provides functions for obtaining a variety of graphical displays that may be useful in the subgroup analysis setting. An example with a prostate cancer dataset is provided. The graphical techniques considered include level plots, mosaic plots, contour plots, bar charts, Venn diagrams, tree plots, forest plots, Galbraith plots, L’Abbé plots, the subpopulation treatment effect pattern plot, alluvial plots, circle plots and UpSet plots.


by Ristl Robin

Calculates exact hypothesis tests to compare a treatment and a reference group with respect to multiple binary endpoints. The tested null hypothesis is an identical multidimensional distribution of successes and failures in both groups. The alternative hypothesis is a larger success proportion in the treatment group in at least one endpoint. The tests are based on the multivariate permutation distribution of subjects between the two groups. For this permutation distribution, rejection regions are calculated that satisfy one of different possible optimization criteria. In particular, regions with maximal exhaustion of the nominal significance level, maximal power under a specified alternative or maximal number of elements can be found. Optimization is achieved by a branch-and-bound algorithm. By application of the closed testing principle, the global hypothesis tests are extended to multiple testing procedures.


by Ristl Robin

Provides global hypothesis tests, multiple testing procedures and simultaneous confidence intervals for multiple linear contrasts of regression coefficients in a single generalized estimating equation (GEE) model or across multiple GEE models. GEE models are fit by a modified version of the geeM package.


by Florian Frommlet

MOSGWA is a software tool for genome wide association studies (GWAS) using model selection based on modifications of information criteria suited for high dimensional settings. It can be used for binary traits using logistic regression (case control studies) and for quantitative traits using linear regression.


by Elias Meyer

Computes the most important properties of four ‘Bayesian’ early gating designs (two single arm and two randomized controlled designs), such as minimum required number of successes in the experimental group to make a GO decision, operating characteristics and average operating characteristics with respect to the sample size. These might aid in deciding what design to use for the early phase trial.