Using SNOMED-CT to Help the Transition from Microbiological Data to ICD-10 Sepsis Codes

Area of Science:

• Medical Informatics
• Clinical Microbiology
• Health Information Management

Background:

• Accurate coding of sepsis is crucial for patient care and research.
• Assigning International Classification of Diseases, Tenth Revision (ICD-10) codes for sepsis requires detailed microbiological knowledge due to varying granularity.
• Current manual coding processes can be time-consuming and prone to errors.

Purpose of the Study:

• To develop and evaluate an automated system for assigning ICD-10 codes for sepsis.
• To leverage the Systematized Nomenclature of Medicine - Clinical Terms (SNOMED-CT) for enhanced coding accuracy.
• To create a dichotomous classification of sepsis-causing bacteria aligned with ICD-10 requirements.

Main Methods:

• A dichotomous classification of bacteria causing sepsis was generated based on ICD-10 granularity.
• An algorithm was developed to explore SNOMED-CT and assign appropriate ICD-10 sepsis codes.
• The system was tested on a dataset of 164 bacterial pathogens.

Main Results:

• The automated system demonstrated a low error rate of 1.22% in assigning ICD-10 sepsis codes.
• The developed bacterial classification effectively mapped microbiological data to coding requirements.
• SNOMED-CT proved a valuable resource for automating this complex coding task.

Conclusions:

• Automated coding using SNOMED-CT can significantly improve the accuracy and efficiency of assigning ICD-10 sepsis codes.
• The proposed algorithm and bacterial classification provide a robust framework for clinical informatics applications.
• This approach has the potential to streamline healthcare data management and improve sepsis surveillance.