Recording of Alcohol Use Disorder in Electronic Health Records: Developing a Recommended Codelist for Research

Area of Science:

• Medical Informatics
• Public Health
• Clinical Informatics

Background:

• Electronic health records (EHR) are crucial for health research but present data utilization challenges.
• A validated codelist for alcohol use disorder (AUD) in UK primary care is essential for robust population-based research.

Purpose of the Study:

• To develop a comprehensive AUD codelist for the Clinical Practice Research Datalink (CPRD) Aurum database.
• To enhance the accuracy and consistency of identifying AUD in UK primary care EHR data.

Main Methods:

• Searched the CPRD code browser using alcohol-related keywords and a predefined strategy.
• Categorized codes as AUD (diagnosis, withdrawal, chronic harm) or related alcohol use (status, acute harm, screening).
• Independent review by multiple clinicians (GPs) with consensus reached on disagreements; assessed reliability using kappa statistics.

Main Results:

• Identified 556 potential alcohol-related codes; moderate inter-coder reliability (Kappa AUD: 0.72, all categories: 0.62).
• Developed a final codelist of 138 codes indicating AUD.
• Included 38 codes for AUD diagnosis, 14 for withdrawal, and 85 for chronic alcohol-related harm (physical and mental health).

Conclusions:

• Significant variability exists in how primary care clinicians code alcohol use and related harms.
• The developed AUD codelist provides a valuable reference for future UK primary care research.
• Further validation through clinical reviews and qualitative studies is recommended to refine understanding of coding practices.