Developing a controlled vocabulary for use in a hospital information system

Area of Science:

• Health Informatics
• Medical Terminology
• Data Management

Background:

• Controlled vocabularies are essential for health information systems.
• User-generated terms are crucial for accurate data input.
• Existing systems may lack comprehensive, user-friendly terminology.

Purpose of the Study:

• To develop a methodology for incorporating user-inputted terms into a controlled vocabulary.
• To enhance the usability and accuracy of health information systems.
• To bridge the gap between user language and standardized medical codes.

Main Methods:

• Collecting "free text" diagnosis descriptions from health information system users.
• Sorting and identifying appropriate "atomic" terms from user input.
• Submitting terms to domain experts for validation of appropriateness and fidelity.
• Coding identified concepts into SNOMED International for integration into controlled pick lists.

Main Results:

• A systematic methodology was established to extract and standardize user-entered diagnostic terms.
• User-generated terms were successfully mapped to atomic concepts.
• Expert review confirmed the appropriateness and fidelity of the selected terms.
• The process facilitates the creation of a comprehensive, user-centric controlled vocabulary.

Conclusions:

• The developed methodology effectively transforms user-entered free text into standardized medical concepts.
• This approach enhances health information system data quality and user experience.
• Integration with SNOMED International ensures international coding compatibility and future system enhancement.