Diagnostic Decision-Making Variability in Glaucoma: Comparative Analysis of Novice and Expert Optometrists to Inform AI System Design (Preprint)

BACKGROUND This study examines the diagnostic decision-making differences between novice and expert optometrists in glaucoma diagnosis. OBJECTIVE The objective of this study is to identify and analyze the variations in diagnostic decision-making approaches between novice and expert optometrists, with the aim of informing the development of AI systems that can enhance diagnostic accuracy and consistency across different levels of expertise. METHODS We conducted in-depth interviews with 14 optometrists including both novices and experts, focusing on their approaches to glaucoma diagnosis. Responses were coded and analyzed qualitatively and quantitatively, and themes were extracted to understand their decision-making patterns and find out variations in their decision-making approaches RESULTS Experts showed higher concordance rates with clinical decisions with limited data, highlighting the impact of experience and data availability on clinical judgment. The accuracy gap narrowed with patient data access to complete historical data. Approaches to the exams assessment and decision differed significantly: experts emphasized comprehensive risk assessments and progression analysis, demonstrating cognitive efficiency and intuitive decision-making, while novices relied more on structured, analytical methods and external references. CONCLUSIONS Understanding these differences can inform the future design of AI systems to enhance diagnostic accuracy and consistency across different expertise levels, improving patient outcomes in optometric practice.