Most of those who work with Clinical Data Interchange Standards Consortium (CDISC) datasets have probably heard of the term “traceability.” But why is it important and how can it best be implemented?  

What is Traceability 

Traceability is the ability to “trace” or identify the downstream element(s) of a dataset record or analysis result. It allows one to go from the analysis results back to specific records in the Analysis Data Model (ADaM) dataset(s), back to the specific records in the Study Data Tabulation Model (SDTM) dataset(s), and back to the raw clinical dataset(s). It is essentially a clear audit trail for your analysis results. 

Importance of Traceability 

Why does this matter? One reason is entirely practical; clear traceability makes programming and review much easier. The other reason is regulatory authorities care. FDA emphasizes the importance of traceability in its Technical Conformance Guide, stating that “the regulatory review of a submission may be compromised” if the reviewer cannot follow the path from the original data collected to the analysis results.  

How to Succeed in Establishing Traceability 

Unfortunately, establishing traceability is not as simple as checking a box. FDA acknowledges that “establishing traceability is one of the most problematic issues associated with any data conversion.” A clear first step is to follow Clinical Data Acquisition Standards Harmonization (CDASH) standards for clinical data (which is harmonized with SDTM) and CDISC standards for your tabulation and analysis datasets (SDTM and ADaM). Running your SDTM and ADaM datasets through Pinnacle 21 (which provides compliance checks for CDISC output) is also an important step, but establishing traceability still requires additional thought and planning. 

The chart below provides a list of tips in terms of how to think about traceability at each step of the data development process. 

These tips are not exhaustive; as traceability is important to establish in other cases not mentioned here (e.g., an older “legacy” study with data that does not follow CDISC standards, datasets supporting integrated analyses, datasets supporting analyses using multiple imputation, etc.) Rho has extensive experience navigating traceability in these cases as well.  

Establishing traceability is not an insurmountable task. In the process, it is important to continually ask yourself if a reviewer can successfully “trace” back to the original data. Having this mindset will aid in creating a successful submission package.  

Need help establishing traceability? Contact us to speak with one of our Biometrics Regulatory Experts. 

References 

• FDA. Study Data Technical Conformance Guide. Version 5.9. October 2024. 
  • Link: https://www.fda.gov/media/153632/download 
•  CDISC. Analysis Data Model Implementation Guide. Version 1.3. 2021. 
•  PMDA, Technical Conformance Guide on Electronic Study Data Submissions (Provisional English Translation). 2024.  
  • Link: https://www.pmda.go.jp/files/000267935.pdf 
• Tinazzi, Angelo. How to be Traceable in ADaM. PHUSE EU Connect 2021. 
  • Link: https://www.lexjansen.com/phuse/2021/si/PAP_SI08.pdf 

Julie Gubitz, MS, MPH, JD, Senior Biostatistician, has six years of experience providing statistical support for Phase 1-3 clinical trials. Ms. Gubitz’s experience includes supporting NDA submissions, consulting on protocol development, writing detailed statistical analysis plans, preparing CDISC-compliant specifications for analysis (ADaM) databases, creating CDISC-compliant submissions packages, specifying and performing statistical analyses, preparing displays and reports, and providing safety evaluations for data monitoring committees. Ms. Gubitz has supported both federally funded projects and commercial projects with biotech/pharmaceutical companies. She has provided statistical support in multiple therapeutic areas, with a focus on pain and autoimmune disorders.