A Pragmatic Method to Integrate Data from Pre-existing Cohort Studies using the Clinical Data Interchange Standards...

Area of Science:

• Data management
• Clinical informatics
• Database integration

Background:

• Legacy data utilization is increasing, but methods for integrating pre-existing databases for different purposes are not established.
• Previous work developed a tool for generating Study Data Tabulation Model (SDTM) data from hypothetical clinical trial data.
• The Clinical Data Interchange Standards Consortium (CDISC) SDTM is a key standard in clinical data management.

Purpose of the Study:

• To design and demonstrate a practical model for integrating multiple pre-existing databases.
• To leverage the CDISC SDTM for harmonizing data from disparate sources.
• To establish a methodology for legacy data utilization in pooled analyses.

Main Methods:

• Data integration involved three phases: variable confirmation, SDTM mapping, and SDTM data generation.
• REDCap field annotations were used to embed metadata (domain name, variable name, test code) based on CDISC SDTM.
• The REDCap2SDTM v2 tool was utilized to generate mapped SDTM data in ODM format.

Main Results:

• SDTM data were successfully generated as CSV files for seven metadata-defined domains.
• Twenty-two common variables were mapped across three independently existing databases.
• The successful integration of three pre-existing databases into a single CDISC SDTM-formatted database was achieved.

Conclusions:

• The CDISC SDTM is a valuable tool for the integration of multiple pre-existing databases.
• The proposed model offers a practical approach to legacy data harmonization.
• This methodology facilitates pooled analyses and enhances the utility of existing research data.