I was recently made aware of the release of the mmrm package in R. It has been developed by a group of programmers and statisticians at a number of pharmaceutical companies, led by Daniel Sabanes Bove at Roche, as part of the ASA Biopharmaceutical Section Software Engineering Working Group. I’ve written previously about fitting mixed models for repeated measures (MMRM) using R, Stata and SAS. In R, this can be done using the gls function in the nlme package, but there are a number of limitations with this approach. For example, it is difficult (or impossible) to fit models where you allow the covariance parameters to be distinct between treatment groups. In this post, I’ll take a very quick look at the new mmrm package in R.
MMRM vs LME model
Following my recent post on fitting an MMRM in SAS, R, and Stata, someone recently asked me about when it is preferable to use a Mixed Model Repeated Measures (MMRM) analysis as opposed to a a linear mixed effects model (LME) which includes subject level random effects (e.g. intercepts).
Mixed model repeated measures (MMRM) in Stata, SAS and R
Linear mixed models are a popular modelling approach for longitudinal or repeated measures data. They extend standard linear regression models through the introduction of random effects and/or correlated residual errors. In the context of randomised trials which repeatedly measure patients over time, linear mixed models are a popular approach of analysis, not least because they handle missing data in the outcome ‘automatically’, under the missing at random assumption. Because of this a mixed model analysis has in many cases become the default method of analysis in clinical trials with a repeatedly measured outcome.