Abstract
Abstract: :
Purpose: The purpose of this work is to describe differences in examination of the optic disk by experts and beginners and to use these differences in the generation of an algorithm for basic competency in the assessment of the optic disk. Methods: A Topcon 6NWS fundus camera and Haag–Streit Eyecap Imaging System captured high quality images of the optic disk. One–hundred fundus images of consecutive patients were used in the analysis. Four experienced clinicians served as experts and thirty–nine second–year optometry students served as beginners who reviewed the images without time limit. Each clinician recorded the horizontal and vertical cup–to–disk (CD) ratios, an assessment of the color, integrity of the neuro–retinal rim and an overall assessment of the integrity of the optic disk. Except for the CD ratio, assessments used a scale of 1, unremarkable; 2, questionable; and, 3, problem. The experts were judged by the median of the other experts and the beginners were judged by the median of all the experts. Results: Experts and beginners differed in their judgments of horizontal and vertical CD ratios (Kruskal–Wallis non–parametric analysis of variance {KW}, N=4149, 4148, p=0.0001, 0.0001). Experts and beginners also differed in their assessment of the color of the optic disk (KW, N=4149, p=0.0001) and the integrity of the neuro–retinal rim (KW, N=4149, p=0.0001). Beginners differed from the standard mean absolute color and neuro–retinal rim integrity by 1.4 and 2.1 times those of the experts, respectively. The two classes differed in their overall assessment of the optic disk (KW, N=4149, p=0.0001). A basic competency algorithm using absolute differences from the standard and approximately double the values of the experts as criteria indicates that all of the experts and eleven of the thirty–seven beginners demonstrate basic competency in the assessment of the optic disk. Conclusions: The basic competency algorithm used mean absolute differences of 0.10, 0.10, 0.35, 0.60 and 0.80 for horizontal and vertical CD ratios, assessments of color, integrity of the neuro–retinal rim and overall integrity of the optic disk as single pass–fail criteria for these components. Since substantial variability occurs in clinical judgments, the algorithm allows one of the five single differences to fall below the criterion and still pass the overall task. This work suggests that such an algorithm can be developed and may be useful in the certification of individual clinicians examining the optic disk.
Keywords: clinical research methodology • optic disc