Repository files navigation

Themixed model for repeated measures(MMRM) isa popular model for longitudinal clinical trial data with continuousendpoints, andbrms is powerful andversatile package for fitting Bayesian regression models. Thebrms.mmrm R package leveragesbrms to runMMRMs,and it supports a simplified interface to reduce difficulty and alignwith best practices for the life sciences.

The documentation website athttps://openpharma.github.io/brms.mmrm/has a complete function reference and tutorial vignettes.

To ensure the correctness of the model and its implementation, thispackage has been validated using simulation-based calibration andcomparisons against the frequentistmmrm package on twoexample datasets. The analyses and results are described in the packagevignettes linked below:

• Simulation-basedcalibration
• FEV1 data comparison between Bayesian and frequentistMMRMs.
• BCVA data comparison between Bayesian and frequentistMMRMs.

Notably,FEV1andBCVAare the same datasets thatmmrm uses to compareitself against SAS inthisvignette.For additional validation in your functional area or domain ofexpertise, you may choose to run similar analyses on your own datasetsto comparebrms.mmrm againstmmrm and/or SAS.

Please report questions and problems asGitHubdiscussions andGitHubissues, respectively.

Thanks to theopenstatsware andRConsortium for providing professionalnetworks to recruit skilled statisticians and developers.

Please note that the brms.mmrm project is released with aContributorCode ofConduct.By contributing to this project, you agree to abide by its terms.

To cite package 'brms.mmrm' in publications use:  Landau WM, Kunzmann K, Sidi Y, Stock C (????). _brms.mmrm: Bayesian  MMRMs using 'brms'_. R package version 1.1.0.9002,  <https://github.com/openpharma/brms.mmrm>.A BibTeX entry for LaTeX users is  @Manual{,    title = {brms.mmrm: Bayesian MMRMs using 'brms'},    author = {William Michael Landau and Kevin Kunzmann and Yoni Sidi and Christian Stock},    note = {R package version 1.1.0.9002},    url = {https://github.com/openpharma/brms.mmrm},  }

• Paul-Christian Bürkner (2017). brms: An R Package for BayesianMultilevel Models Using Stan. Journal of Statistical Software, 80(1),1-28.
• Mallinckrodt, C.H., Lane, P.W., Schnell, D. et al. Recommendations forthe Primary Analysis of Continuous Endpoints in Longitudinal ClinicalTrials. Ther Innov Regul Sci 42, 303–319 (2008).
• Holzhauer, B., and Weber, S. (2024), “Bayesian mixed effects model forrepeated measures,” in Applied Modeling in Drug Development, NovartisAG.https://opensource.nibr.com/bamdd/src/02h_mmrm.html.