Analysis of cohort stepped wedge cluster-randomized trials with non-ignorable dropout via joint modeling

This repository contains additional material for the manuscript titled "Analysis of cohort stepped wedge cluster-randomized trials with non-ignorable dropout via joint modeling", currently published in Statistics in Medicine (DOI: 10.1002/sim.10347).

A preprint is available on arXiv, and the latest released version of this repository is archived on Zenodo as well.

Table of Content

The content of this repository is organised as follows:

1. The 01-example-swjm folder contains simulated data and code to analyse a stepped wedge trial with the joint modelling approach that we propose in the manuscript;

2. The 02-example-swjm-individual folder contains simulated data and code to analyse an individually-randomised stepped wedge trial (e.g., without any cluster effect);

3. The 03-simulation-code folder contains code that could be used to reproduce the Monte Carlo simulation studies reported in the paper;

4. Finally, the 04-convergence-issues folder contains code and steps that could be used to improve convergence of the joint model.

{simswjm} Package

Note that the simulation code requires the {simswjm} R package, which can be installed from GitHub with the following code:

# install.packages("remotes")
remotes::install_github("RedDoorAnalytics/simswjm")

License

All code included in this repository is released under the MIT License. If you find this useful, please cite the manuscript and/or this repository in your work.