Complexity of biomedical data models in cardiology: the Intranet-based AF registry

Area of Science:

• Cardiology and Medical Informatics
• Development of clinical data management systems

Background:

• Atrial fibrillation (AF) is a prevalent cardiac arrhythmia with significant complications.
• Current research explores novel pacing devices for AF treatment.
• Effective data management is crucial for analyzing clinical information, quality of life, and device parameters in AF patients.

Purpose of the Study:

• To develop and implement a high-granular, XML-based documentation scheme for a prospective atrial fibrillation registry.
• To establish a platform for long-term follow-up and detailed statistical analysis of AF patient data.
• To integrate a complex research database into a busy clinical workflow.

Main Methods:

• An XML-based documentation scheme with 619 items across eight tables was developed.
• State-of-the-art intranet technology was used for implementation.
• Data from pacing devices (400-500 parameters per visit) were transferred via a specific interface into the database.
• Iterative software engineering and plausibility checks were employed for data quality assurance.

Main Results:

• Detailed information on 88 patients is currently recorded.
• A dedicated visualization tool was developed to manage the complex dataset.
• Success factors included interfaces for non-redundant data entry and clinical user benefits like patient summaries.

Conclusions:

• The implemented system facilitates the integration of complex research data into routine clinical workflows.
• Interinstitutional cooperation is essential for establishing common documentation standards to enable data pooling for robust statistical evaluation.
• Further recruitment of a large patient collective is necessary for comprehensive statistical analysis.