Purpose
Discus is a freely accessible online software package that allows observers to assess their skills at interpreting non-stereoscopic optic disc photographs for signs of glaucomatous damage. The aim of this study was to validate Discus software and to establish reference data for performance and repeatability with Discus software in participants of a four-year Ophthalmic residency training program in Saudi Arabia.
Methods
Fifty four residents participated in this study. All residents tested themselves three times over five days. The software display non-stereoscopic optic disc images, 80 images without evidence of repeatable visual field (VF) loss (VF negatives) and 20 with repeatable VF loss (VF positives), for up to 60 seconds. The software displays 2 VF positive and 24 VF negative images were repeated to asses consistency. Each observer will rate optic disc image on a 5-point scale (definitely healthy, probably healthy, not sure, probably damaged, definitely damaged). At the end of each test, online feedback was given to each participant.
Results
The mean AUROC was 0.68, 0.68, 0.69 and 0.71 for first, second, third and fourth years residents respectively (p=0.04, Kruskal-Wallis test). The overall performance of residents was 0.69 (95% CI, 0.65, 0.74) compared to a panel of experts which was 0.79 (95% CI, 0.58, 0.96). Fourth years residents had the highest correlation with experts(r=0.64,p=0.02). There was no difference in the performance on the three repeated tests (P=0.6) with high degree of reliability (ICC=0.76). But there was more variability among junior residents which did not influence the mean performance (relationship between mean performance and mean difference in performance was -0.056, p=0.64).
Conclusions
Senior residents had better performance and less variability on repeated tests compared to junior residents . Discus allows efficient and precise assessment of resident’s performance at assessing optic disc photographs in patients with glaucoma.