Progression models for repeated measures (PMRMs) are continuous-time nonlinear mixed-effects models for longitudinal data in progressive diseases. Unlike mixed models for repeated measures (MMRMs), which estimate treatment effects as linear combinations of additive effects on the outcome scale, PMRMs characterize treatment effects in terms of the underlying disease trajectory. This framing yields clinically interpretable quantities such as average time saved or percent reduction in decline due to treatment. These time-based measures are becoming increasingly popular in clinical studies of neurodegenerative diseases (https://www.nature.com/articles/d41586-024-00756-8).
{pmrm} is an open-source R package for fitting frequentist PMRMs. Under the hood, {pmrm} leverages {RTMB}, a high-performance user-friendly R framework for fitting complex statistical models, achieving orders-of-magnitude speedups over equivalent implementations with nlme::gnls(). {pmrm} provides first-class functionality for simulation, post-processing, visualization, marginal mean estimation, and S3 methods for standard generics, making PMRMs accessible to clinical statisticians and analysts. Like the {mmrm} package, {pmrm} is a member of the OpenPharma GitHub organization.