Using forest plots to interpret covariate effects in pharmacometric models

Pharmacometrics

Methodology

CPT Pharmacometrics Syst Pharmacol. 2024 May;13(5):743-758.

Jonsson EN, Nyberg J

The tutorial addresses the important and complicated challenge of how to communicate the impact of covariates.

In contrast to what may be expected, forest plots rely very much on transparency in the communication of the assumptions leading up to the plot - from study design and conduct, via model development choices and the subjective choices made when creating the plot.

However, it is not sufficient for the creator of the forest plot to be transparent, it is also important that the receiver of the information is willing to incorporate the impact of these assumptions in the interpretation of the plot. The tutorial aims to facilitate the collaboration by discussing the various aspects to keep in mind and by supplying checklists for both the creator and the receiver of the forest plots, as well as suggesting a structured way of formulating informative figure captions. This tutorial aims to support a rational use of forest plots to support a clear communication of covariate effects in pharmacometric models.