Data quality and completeness in a web stroke registry as the basis for data and process mining.

Area of Science:

• Health Informatics
• Clinical Data Management
• Stroke Neurology

Background:

• Electronic health records (EHRs) frequently contain missing values and errors, hindering their effective use.
• Maintaining high data quality in multicentric registries is crucial for reliable research and clinical practice.

Purpose of the Study:

• To re-engineer a web-based stroke registry to enhance data quality and user experience.
• To implement a knowledge-based system for consistent data interpretation and error prevention/detection.
• To improve stroke unit coordination and networking through ancillary tools.

Main Methods:

• Development of a knowledge-based data entry support system.
• Implementation of tools for monitoring patient enrollment, calculating care indicators, and analyzing guideline compliance.
• Integration of stroke unit profile data entry.
• Application of statistical methods, data mining, and process mining for analysis.

Main Results:

• Significant improvement in user experience during data entry.
• Substantially enhanced data quality, ensuring reliability for analyses.
• Better coordination and networking among stroke units.
• Development of indicators for assessing stroke care quality.

Conclusions:

• The re-engineering process effectively improves data quality and usability in electronic health records.
• The implemented system supports reliable data analysis, enhances clinical practice guideline adherence, and facilitates knowledge discovery in stroke care.