A Customizable Data Quality Tool for Global Observational Research Networks
View abstract on PubMed
Summary
This summary is machine-generated.This study generalized a data quality tool for multi-site clinical data analysis, making it easier for research consortia to use. The Harmonist Data Toolkit improves data consistency and reporting without requiring extensive technical expertise.
Area Of Science
- Clinical Data Management
- Health Informatics
- Observational Studies
Background
- Multi-site observational clinical data analysis requires robust data quality evaluation.
- Existing data quality tools often demand significant technical expertise, limiting their adoption by research consortia.
- Generalizing data quality solutions is crucial for improving the reliability of collaborative research.
Purpose Of The Study
- To generalize an established data quality checking and report generation tool for easier implementation across diverse research networks.
- To reduce the technical expertise needed at user sites for data quality assessment.
- To provide a scalable solution for data quality and reporting in multi-site research consortia.
Main Methods
- Leveraged REDCap (Research Electronic Data Capture) software to store data model details, variable expectations, and user group information.
- Developed an application using the REDCap API to retrieve stored details and assess dataset conformance, logical consistency, and completeness.
- Built the Harmonist Data Toolkit using freely available REDCap and R/Shiny platforms, with code accessible on GitHub.
Main Results
- The generalized Harmonist Data Toolkit successfully reduced the need for local programming expertise.
- All five collaborating consortia reported benefits in detecting inconsistencies and generating informative data reports and visualizations.
- The toolkit effectively assessed datasets for conformance to data models, logical consistency, and completeness.
Conclusions
- The Harmonist Data Toolkit provides a valuable, accessible solution for data quality and report generation in research consortia.
- The toolkit addresses the need for standardized data quality assessment in multi-site observational studies.
- Freely available platforms like REDCap and R/Shiny enable the development of powerful, generalizable research tools.