Unambiguous data modeling to ensure higher accuracy term binding to clinical terminologies

Area of Science:

• Health Informatics
• Clinical Data Management
• Medical Error Reduction

Background:

• Standardized coded data recording is crucial for minimizing medical errors.
• Misinterpretation and misrepresentation of clinical data can lead to patient harm.
• Ambiguity in clinical data models compromises data quality and mapping accuracy.

Purpose of the Study:

• To address the need for unambiguous clinical data models.
• To improve the quality and accuracy of data mapping to terminology codes.
• To promote best practices in clinical data modeling for accurate patient data recording.

Main Methods:

• Selected an ambiguous clinical data model for revision.
• Remodeled the data model to clarify structure, intended use, and semantics.
• Evaluated the impact of the revised model on data mapping accuracy and clinical expert consensus.

Main Results:

• Data mapping accuracy improved from 64.7% to 80.55% after model revision.
• Clinical experts achieved 93.1% agreement on mappings with the revised model, compared to 48.57% previously.
• The revised model was found to be easier for clinical experts to use.

Conclusions:

• Unambiguous clinical data models are essential for accurate patient data recording and coding.
• Remodeling efforts can significantly enhance data quality and reduce diagnostic or treatment errors.
• Adopting good modeling practices supports clinicians in unambiguous and accurate patient data management.