Research Methods Program

Department of Biostatistics

Vanderbilt University School of Medicine

Nashville TN USA

- General COVID-19 Therapeutics Trial Design
- Web conference 2020-05-08 recording | Talks
- Bayesian fully sequential design and analysis plan for the 7-level COVID ordinal outcome scale for rapid learning and decision making
- Simulating longitudinal ordinal clinical trials
- ðŸ†• Markov longitudinal ordinal model overview and discussion
- ðŸ†• Longitudinal ordinal models as a general framework for medical outcomes
- ðŸ†• Longitudinal ordinal analysis of ACTT-1 Remdesivir study and its analysis file creation and analysis by Max Rohde
- Longitudinal ordinal analysis of ORCHID and ðŸ†• Restricted mean days in state by Max Rohde
- Longitudinal ordinal analysis of VIOLET 2
- Markov modeling for longitudinal data with irregular time points
- Discussion about the statistical model for the outcomes and related Bayesian modeling discussion
- Controlling Î± vs.Â Probability of a Decision Error
- Continuous learning from data: No multiplicities from computing and using Bayesian posterior probabilities as often as desired
- Simulating frequentist and Bayesian operating characteristics for a 3-level ordinal longitudinal outcome, also demonstrating how to translate data from a published trial into a simulation model for a new trial. An example shows that the usual time to recovery analysis effectively assumes proportional odds when deaths occur.
- Examples of Markov partial proportional odds models and comparison with time to recovery analysis in two simulated datasets
- Template and examples of simulating Bayesian operating characteristics
- Power Simulation Templates for Ordinal Longitudinal Outcomes
- Comprehensive Markov model simulations (Bayesian and frequentist operating characteristics)
- 3-level ordinal response with model derived from published data including Bayesian sequential analysis and pre-simulation of treatment effect standard errors
- Bayesian sequential RCT with futility simulation
- Efficiency as a function of frequency of longitudinal assessments

- Reporting Bayesian results by David Rindskopf
- Using Bayesian methods to augment the interpretation of critical care trials
- Violation of proportional odds is not fatal
- If you like the Wilcoxon test you must like the proportional odds model
- Discussion Board, which includes new updates regarding
`Stan`

code for the Bayesian PO model - General rationale and methods for Bayesian clinical trial design and analysis and its overview
- Fundamental advantages of Bayes in drug development webinar
- R
`rms`

package Bayesian proportional odds model with random effects and implementing the partial proportional odds model to allow pre-specified departures from the proportional odds assumption (e.g.Â in a longitudinal ordinal outcome analysis the mix of outcomes may change over time but the treatment effect may be fairly constant) - Nathan Jamesâ€™
`R`

package for the Bayesian proportional odds model - Stan code from Ben Goodrich (Columbia University) for the PO model and Benâ€™s notes
- Inner workings of Stan ordinal models and Github
- Stan code repository from Ben Goodrich
- Discussion Board for COVID-19 research methods resources in general
- R code example of a frequentist mixed effects ordinal PO model vs.Â a Bayesian random effects ordinal model for longitudinal data
- Simulating ordinal outcome data by Keith Goldfeld
- Exploring the properties of a Bayesian model using high performance computing by Keith Goldfeld
- ðŸ†• References related to ordinal longitudinal and Markov models
- Papers about statistical modeling of longitudinal ordinal responses
- Biostatistics in Biomedical Research Tutorials - see sections below:

Topic |
Section |
---|---|

Ordinal outcomes in clinical trials | 3.6 and 5.12.5 |

Proportional odds model | 7.6 |

Power calculations tailored to the proportional odds model | 7.8.3 |

Bayesian approach for a single mean | 5.6.2 |

Bayesian logistic model | 6.10.3 |

Bayesian two-sample t-test | 5.9.3 and 5.10.5 |

Analysis plan for differential treatment effect (heteroeneity of treatment effect) and why subgroup analysis should be avoided | 13.6 |

Covariate adjustment in RCTs | Chapter 13 |

Overview of branches of statistics | 5.3 |

Problems with p-values | 5.4 |

- ðŸ†• Risk prediction models for discrete ordinal outcomes: Calibration and the impact of the proportional odds assumption by M Edlinger, M van Smeeden, H Alber, M Wanitschek, B Van Calster
- Information gain from using ordinal instead of binary outcomes
- Bayesian vs.Â frequentist statements about treatment efficacy
- Journey from frequentist to Bayesian statistics
- A litany of problems with p-values
- P-values and type I errors are not the probabilities we need
- Null hypothesis significance testing never worked

- ðŸ†• Paper exemplifying game playing in choice of RCT endpoint and how things go seriously wrong with time to symptom resolution as an endpoint (authors had to ignore emergency department visits)
- Detailed disease progression of 213 patients hospitalized with COVID-19 in the Czech Republic by ModrÃ¡k et al
- Analysis of the Czech data
- Clinical Trial Tracker from Gates Foundation and Cytel
- Bayesian adaptive trial
- The P value line dance: When does the music stop? by Marcus Bendtsen
- ðŸ†• Interim analysis for early stopping during the study by David Bock