The central question: “how likely is this range of results if the point estimate of efficacy is the true underlying efficacy?". Point estimates of efficacy are obtained from the entire cohort and from individual patient age groups. The question here is whether a simulation model with the reported overall efficacy is able to reproduce the ranges of reported per patient-group efficacies. If it is difficult to produce ranges as precise as reported, the interpretation is that it is difficult to achieve that level of consistency in the ideal case, and unlikely to be generated from experimental data without other supporting evidence.
epitoolstidyversepatchworkfurrrhere
-
All code is in the
Rdirectory -
Study data is in the
datadirectory -
Results are in the
resultsdirectory -
Figure 1 and Figure 2 can be reproduced by running
R/head-to-head.R -
Individual trial simulations can be examined and modified by looking for the appropriately named file in the
Rdirectory.