Several discrepancies have been discovered in statistical analysis results between different programming languages, even in fully qualified statistical computing environments. Subtle differences exist between the fundamental approaches implemented by each language, yielding differences in results which are each correct in their own right. The fact that these differences exist causes unease on the behalf of sponsor companies when submitting to a regulatory agency, as it is uncertain if the agency will view these differences as problematic. Understanding the agency’s expectations will contribute significantly to enabling the broader adoption of multiple programming languages in the production of data submission packages for regulatory review
The Clinical Statistical Reporting in a Multilingual World project seeks to clearly define this problem and provide a framework for assessing the fundamental differences for a particular statistical analysis across languages. In this context, the risk of interpreting numerical differences in analysis results due solely to differences in programming language can be mitigated, instilling confidence in both the sponsor company and the agency during the review period. This will be accomplished by:
|Objectives and Deliverables||Timelines|
GitHub Repository documenting identified differences between statistical analysis implementations (based on R and SAS use cases as a starting point)
Expand repository to provide comparable syntax across languages (based on R and SAS use cases as a starting point)
Expand GitHub repository to incorporate Python and/or Julia
White Paper providing framework for addressing language discrepancies in statistical analysis implementations, including specific use cases as examples
CURRENT STATUS Q3/42021
Started draft writing of white paper with new sub-team.
Upcoming presentation at the EU Connect.
|Aiming Yang||Merck, Sharp & Dohme|
|Chung-kai Sun||Janssen Research & Development|
|Harshal Khanolkar||Novo Nordisk|
|Michael Kane||Yale University|
|Mia QI||Janssen Research & Development|
|Min-Hua Jen||Eli Lilly|