The R Consortium’s R Tables for Regulatory Submissions (RTRS) Working Group has made considerable progress in identifying and working through the issues involved with developing a modern R based framework for creating tables. The goal is to be able to make it easy for statistical programmers working in pharmaceutical companies to find the right R resources for creating any type of table that may be required to support a submission to the FDA or any other regulatory agency.
Currently there are at least six R packages that have the functionality to support some portion of the panoply of tables that might comprise an essential part of a statistical report. These packages include
- flextable
- gt
- huxtable
- mmtable2
- rtables
- Tplyr
The trick is to establish a framework that makes it easy to use the combined features of relevant R packages to produce any table that is likely to show up in a production environment.
RTRS wants examples of tables, those that are already part of your standard statistical submissions, and those that you would use if you could make them.
A: Drug X B: Placebo C: Combination
(N=134) (N=134) (N=132)
---------------------------------------------------------
AGE
n 134 134 132
Mean (sd) 33.77 (6.55) 35.43 (7.9) 35.43 (7.72)
IQR 11 10 10
min - max 21 - 50 21 - 62 20 - 69
BMRKR2
LOW 50 45 40
MEDIUM 37 56 42
HIGH 47 33 50
We want to make sure that we understand and explore the entire, conceivable space of production tables and want to see to which extent we are already able to build them in R. So please send us your tables.
We will take them any way that is easy for you to provide. Open up an issue on our GitHub repository (https://github.com/RConsortium/rtrs-wg/issues) and drop them in. Something computable would be best, but we will take text or even screenshots. If there is some part of your example table that is particularly vexing to create, please point that out. Be careful not to include any proprietary information.
If we can already build your difficult table, we will show you how to do it in R. This is meant to be a benchmarking process to showcase abilities of the packages so far and to reveal current limitations. If it turns out that we cannot build your table with R’s current table making capabilities, then there is a good chance that we will add it to our “To Do” list.
If you think that you would like to become involved with our work, please let us know that, too. Include your email address and we will invite you to the next meeting.