Related Experiment Videos
Representation of ophthalmology concepts by electronic systems: intercoder agreement among physicians using
John C Hwang1, Alexander C Yu, Daniel S Casper
1Department of Ophthalmology, Columbia University College of Physicians and Surgeons, New York, New York 10032, USA.
Ophthalmology
|February 21, 2006
Summary
Intercoder agreement for ophthalmology concepts is imperfect across five controlled medical terminologies. Systematized Nomenclature of Medicine, Clinical Terms (SNOMED-CT) showed higher agreement, highlighting the need for improved reproducibility in electronic health records.
Area of Science:
- Medical Informatics
- Health Information Management
- Ophthalmology
Background:
- Accurate coding of medical concepts is crucial for data analysis and electronic health records.
- Evaluating the consistency of different controlled terminologies in representing ophthalmology concepts is essential.
Purpose of the Study:
- To assess intercoder agreement among physician coders for ophthalmology concepts using five distinct controlled terminologies.
- To compare the performance of International Classification of Diseases 9, Clinical Modification (ICD9CM), Current Procedural Terminology, fourth edition, Logical Observation Identifiers, Names, and Codes (LOINC), Systematized Nomenclature of Medicine, Clinical Terms (SNOMED-CT), and Medical Entities Dictionary.
Main Methods:
- A noncomparative case series involving five ophthalmology case presentations.
- Three physician coders independently assigned codes to discrete concepts within each of the five terminologies.
- Intercoder agreement was assessed through exact code matching and manual review for semantic equivalence.
Main Results:
- Analysis of 242 unique concepts revealed complete intercoder agreement ranging from 12% (LOINC) to 44% (SNOMED-CT).
- When considering only concepts with adequate coverage, complete agreement ranged from 33% (LOINC) to 64% (ICD9CM), with no significant differences between terminologies.
- LOINC showed a statistically significant lower intercoder agreement compared to other terminologies when all concepts were analyzed.
Conclusions:
- Existing controlled medical terminologies demonstrate imperfect intercoder agreement for ophthalmology concepts.
- Enhancing intercoder reproducibility is vital for the accurate and consistent electronic representation of medical data in ophthalmology.
- The findings underscore the need for standardization and potential improvements in medical terminologies to ensure data integrity.