Use of Diagnostic Codes for Primary Open-Angle Glaucoma Polygenic Risk Score Construction in Electronic Health...

Area of Science:

• Ophthalmology
• Genetics
• Bioinformatics

Background:

• Polygenic risk scores (PRS) are increasingly utilized to predict disease risk and prognosis.
• Primary open-angle glaucoma (POAG) is a leading cause of irreversible blindness worldwide.
• Accurate case ascertainment is crucial for developing and validating PRS.

Purpose of the Study:

• To compare the performance of a POAG PRS when cases are defined by International Classification of Diseases (ICD) codes versus manual medical record review.
• To evaluate the utility of ICD codes for POAG case identification in large biobank cohorts.

Main Methods:

• A retrospective cohort study was conducted using data from the Mount Sinai BioMe and Mass General Brigham (MGB) biobanks.
• POAG cases were initially identified using ICD codes and subsequently confirmed through optical coherence tomograms and visual fields.
• A genome-wide POAG PRS was computed using the lassosum method and genotype data from the UK Biobank.

Main Results:

• The positive predictive value of ICD codes for POAG ranged from 53% to 55%, with a negative predictive value of 96% to 97%.
• Sensitivity was high (97%) for both methods, but specificity was lower for ICD codes (44%-53%) compared to manual review.
• Area Under the Curve (AUC) values for POAG prediction by PRS showed no statistically significant differences between ICD code and manual review definitions across various ancestries.

Conclusions:

• A POAG PRS demonstrates comparable predictive performance regardless of whether International Classification of Diseases (ICD) codes or manual medical record review is used for case definition in electronic health record-linked biobanks.
• Manual assessment of glaucoma status may not be essential for all PRS studies, potentially streamlining research processes.
• Caution is advised when relying solely on ICD codes for glaucoma diagnosis due to their limited specificity in accurately identifying confirmed glaucoma cases.