Comprehensive ophthalmologists underestimated the likelihood that optic discs had signs of glaucoma almost one-fourth of the time compared with judgments by glaucoma specialists, according to the results of a new study.

Ophthalmology trainees and comprehensive ophthalmologists were twice as likely to underestimate glaucoma likelihood as they were to overestimate it, according to the results of a study published in the May 2014 issue of JAMA Ophthalmology.
Diagnosing glaucoma is challenging, and clinicians often disagree as to whether an optic disc appears to have signs of glaucoma or not. The Glaucomatous Optic Neuropathy Evaluation (GONE) Project aimed to help eliminate diagnostic barriers by gathering information about how clinicians determine whether patients have the condition.
The analyzed results from the GONE Project to determine which signs and symptoms most often lead clinicians to make a correct or incorrect glaucoma diagnosis. Ophthalmology trainees, ophthalmologists, and glaucoma specialists registered for the GONE Project from December 2008 through June 2010 and assessed 42 monoscopic optic disc photographs for physiological and glaucoma-related characteristics through an Internet-based system. Clinicians were then asked to assess glaucoma likelihood for each photograph as unlikely, possible, probable, or certain.
The results indicated that both ophthalmology trainees and comprehensive ophthalmologists often underestimated the likelihood of glaucoma. For optic discs that glaucoma specialists judged as having signs of probable or certain glaucoma, ophthalmology trainees underestimated glaucoma likelihood for 22.1% of discs and ophthalmologists underestimated likelihood for 23.8% of discs.  Trainees overestimated glaucoma likelihood for 13.0% of the discs, while ophthalmologists overestimated for just 8.9%.
The analysis also found that underestimation of vertical cup-disc ratio and failure to identify retinal nerve fiber layer loss, disc hemorrhage, or rim loss were the most influential factors that led to underestimation of glaucoma. When all of these occurred, the average rate of underestimation of glaucoma likelihood rose to 43.0% among trainees and 42.9% among ophthalmologists. When trainees and ophthalmologists accurately assessed all 4 features, underestimation of glaucoma likelihood significantly decreased, to 10.9% for trainees and 3.1% for ophthalmologists, and overestimation decreased to 2.9% for trainees and 1.6% for ophthalmologists. Overestimation of retinal nerve fiber layer loss, rim loss, vertical cup-disc ratio, cup depth, and disc hemorrhage, and incorrect assessment of disc tilt were all associated with overestimation of glaucoma likelihood. In addition, clinicians were more likely to overestimate glaucoma likelihood in large discs.
“The results of this study will facilitate creation of targeted teaching tools for ophthalmology trainees and continual medical education modules for comprehensive ophthalmologists, focusing on specific aspects of disc examination,” the authors of the study conclude. “We recommend that the assessment parameters be highlighted in optic nerve head examination teaching and continuing medical education modules.”