Consolidating CCDs from multiple data sources: a modular approach.

Area of Science:

• Health Informatics
• Clinical Data Management
• Information Systems Engineering

Background:

• Healthcare providers often encounter multiple, fragmented Continuity of Care Documents (CCDs) for a single patient.
• Navigating disparate CCDs is cumbersome, leading to data redundancy and potential conflicts.
• Existing systems lack efficient methods for integrating and de-duplicating patient data from multiple sources.

Purpose of the Study:

• To develop a modular, scalable, and open-source system for consolidating and de-duplicating multiple CCDs.
• To create a unified patient-level data view for improved clinical decision-making.
• To address the challenge of data duplication inherent in health information exchange.

Main Methods:

• A prototype system was engineered for automated CCD consolidation and de-duplication.
• A corpus of 150 de-identified CCDs for 50 unique patients was used for testing.
• System performance was evaluated based on document throughput, file size, and data volume reduction.

Main Results:

• Successful consolidation of all input CCDs with no data loss.
• Significant reduction in data entries: 49% in Problems, 60.6% in Medications, 79% in Allergies.
• Overall document size reduced by 57.5% and line count by 58%.

Conclusions:

• Automated consolidation and de-duplication of clinical documents show significant promise for managing data in health information exchange (HIE).
• The developed system effectively reduces data redundancy and improves data manageability.
• Further research is needed to test the system with heterogeneous vocabularies and across diverse HIE scenarios.