A Comparison of Data Quality Assessment Checks in Six Data Sharing Networks

Area of Science:

• Health Informatics
• Clinical Data Management
• Data Quality Assurance

Background:

• Clinical data sharing organizations utilize various data quality (DQ) assessment programs.
• Standardized terminology is crucial for comparing and harmonizing DQ efforts across these organizations.

Purpose of the Study:

• To compare rule-based data quality (DQ) assessment approaches across multiple national clinical data sharing organizations.
• To evaluate the utility of a harmonized DQ terminology for mapping and understanding DQ checks.

Main Methods:

• Six organizations provided documentation on their DQ checks.
• DQ checks were mapped to a harmonized DQ assessment (DQA) terminology through an iterative process.
• Conventions were developed to ensure consistent mapping, with consensus reached for difficult cases.

Main Results:

• A total of 11,026 DQ checks were provided, with 99.97% successfully mapped to DQA categories.
• The majority of DQ checks fell into Atemporal Plausibility (49.60%), Value Conformance (17.84%), and Atemporal Completeness (12.98%) categories.
• Significant differences in the distribution of DQ checks were observed among participating organizations.

Conclusions:

• A common DQA terminology achieved near-complete coverage of DQ programs, facilitating comparison.
• Observed variations in DQ check distributions suggest differences in organizational priorities and DQA maturity.
• Standardized terminology can guide DQA development and promote broader data quality initiatives across clinical networks.