Recap: Exploring Clinical Submissions With Admiral: An R-Based ADaM Solution With Ben Straub

Have you been looking for a more efficient way to create ADaM (Analysis Data Model) datasets for your clinical trial submissions? Look no further than Admiral, an open-source R package.

In our latest Shiny Gathering, hosted in collaboration with Pharmaverse, featured Ben Straub, Principal Programmer at GSK and one of the maintainers of the package.

Collaborative projects led by pharmaceutical companies are propelling advancements in research. Explore R Packages for Clinical Trial Data in our blog post.

Ben shared his insights and expertise on the admiral R package, highlighting its utility for creating ADaM datasets in clinical submissions. Here’s a quick overview of the session for those who missed it.

The Admiral Package: An Overview

Admiral is an R package designed to facilitate the creation of ADaM datasets, a critical component in clinical drug submissions. Ben walked through the package’s history, its alignment with CDISC standards, and its collaborative development involving multiple pharmaceutical companies.

Watch the Video

Key Features of Admiral

1. Modular Functions: Admiral offers modular functions for adding variables and records to datasets. This modularity simplifies the process of constructing ADaM datasets, making it easier to follow and less prone to errors.
2. Documentation and Examples: The package includes extensive documentation and examples, aiding users in understanding and utilizing its functions effectively.
3. Unit Testing: Admiral incorporates unit testing to ensure code reliability, crucial in the highly regulated pharmaceutical industry.
4. GitHub Integration: The package leverages GitHub for collaborative development, with features like issue tracking, pull requests, and continuous integration ensuring high-quality code and documentation.

Practical Demonstration

Ben provided a practical demonstration of using the admiral package, showcasing how to derive variables and merge datasets. He highlighted the simplicity and efficiency of admiral functions compared to base R and tidyverse alternatives, particularly in reducing repetitive coding tasks.

1. Deriving Variables: Using the derive_vars functions, users can easily add new variables to datasets.
2. Adding Records: The derive_param_computed functions allow for the addition of new records based on existing dataset information.

Future Goals for Admiral

Ben outlined some of the future goals for the admiral package, including:

1. Reducing Complexity: Simplifying the datetime functions to enhance maintainability and reduce code complexity.
2. Improving Error Messaging: Enhancing user messaging to provide clearer guidance when errors occur, making it easier for users to troubleshoot issues.

Contributing to Admiral

He encouraged the community to contribute to the admiral package, highlighting resources like the dummy issue for onboarding new contributors. He emphasized the collaborative nature of admiral’s development and the valuable learning opportunities it offers.

Conclusion

Admiral represents a significant step forward in using R for clinical submissions. By providing a standardized, well-documented, and collaboratively developed solution, it’s helping to pave the way for more R-based submissions in the pharmaceutical industry alongside other open-source packages in the Pharmaverse. 

Here are some helpful links:

• Join the Pharmaverse Slack channel
• Subscribe to the Pharmaverse newsletter
• Blog post about the latest admiral 1.1 update

Our next Shiny Gathering promises to share more insights for the pharma industry. Sign up for our newsletter to find out the details. 

The post appeared first on appsilon.com/blog/.