Reference Based Multiple Imputation

Implements standard and reference based multiple imputation methods for continuous longitudinal endpoints (Gower-Page et al. (2022) ). In particular, this package supports deterministic conditional mean imputation and jackknifing as described in Wolbers et al. (2022) , Bayesian multiple imputation as described in Carpenter et al. (2013) , and bootstrapped maximum likelihood imputation as described in von Hippel and Bartlett (2021) .

Reference Based Multiple Imputation (rbmi)

Overview

The rbmi package is used for the imputation of missing data in clinical trials with continuous multivariate normal longitudinal outcomes. It supports imputation under a missing at random (MAR) assumption, reference-based imputation methods, and delta adjustments (as required for sensitivity analysis such as tipping point analyses). The package implements both Bayesian and approximate Bayesian multiple imputation combined with Rubin's rules for inference, and frequentist conditional mean imputation combined with (jackknife or bootstrap) resampling.

Installation

The package can be installed directly from CRAN via:

install.packages("rbmi")

Note that the usage of Bayesian multiple imputation requires the installation of the suggested package rstan.

install.packages("rstan")

Usage

The package is designed around its 4 core functions:

• draws() - Fits multiple imputation models
• impute() - Imputes multiple datasets
• analyse() - Analyses multiple datasets
• pool() - Pools multiple results into a single statistic

The basic usage of these core functions is described in the quickstart vignette:

vignette(topic = "quickstart", package = "rbmi")

Validation

For clarification on the current validation status of rbmi please see the FAQ vignette.

Support

For any help with regards to using the package or if you find a bug please create a GitHub issue