R already provides the necessary low level building blocks to transform, manipulate and analyse clinical trial data. The Tidyverse makes this even easier. However, the clinical trial reporting process has evolved into a complex set of standards and reporting practices that need to be followed. The Pharmaverse collaboration aims to deliver a “curated and opinionated subset of open-source software packages and codebases, which would be based on the R language and those that we believe are the most suitable for the industry to adopt in order to deliver the clinical data pipeline”.
In this workshop you will be working with the Tidyverse and the first generation of Pharmaverse tools in order to ‘deliver’ typical clinical trial displays. Specifically, we will walk through an end-to-end focussing on a sample of demographic, safety and efficacy summaries. We will start at the planning stage, using Pharmaverse tools to create traditional tabular mock-ups before moving onto live data transformation by using the admiral package to transform SDTM data into ADaM. Next, we will summarise and transform our data into the new ‘Analysis Results Data’ format using a variety of Tidyverse and Pharmaverse tools. Finally, we will start to explore the possibilities of multifaceted reporting using the evolving tlang package and gt to create data ready for submissions, study reports, slide presentations and more.
If you are a clinical statistician, programmer or data scientist and competent in R and the Tidyverse then you will see how supporting tools in the ‘Pharmaverse’ can facilitate the data generation and transformation process. No prior knowledge of the Pharmaverse toolset is required. However, if you don’t work with clinical trial data and you’ve not programmed with the Tidyverse, this is probably not the workshop for you!
Andy Nicholls has been an R user and advocate for over 15 years. He is currently Head of Data Science within GSK’s Biostatistics function and is responsible for driving R adoption. Andy and his team have developed GSK’s world-wide R training programming and drove the creation of Biostatistics’ first dedicated analytics platform for R. The team now maintains several R-based tools and applications which were developed to facilitate the use of R for clinical reporting. Several of these tools have now been released into the public domain under the ‘Pharmaverse’ umbrella. Andy is the lead for the R Validation Hub, a collaboration to support the adoption of R within a biopharmaceutical regulation setting.
Christina Fillmore formally trained as a statistician, over her 5 years at GSK has transitioned into a data scientist. Her focus is on open source package development in a regulated environment. She also has a keen interesting in helping other develop R skills and become confident R programmers. She leads the development of metacore, metatools, and tlang as part of the Pharmaverse collaboration.
