Please use this identifier to cite or link to this item:
Title: Strength of agreement among examiners assessing optic disk stereophotos [Concordancia inter e intra-observador en la evaluación clínica de los discos ópticos mediante estereofotografías]
Author: Paczka Z, J.A.
Gonzalez, J.
Gonzalez Flores, I.
Ochoa-Tabares, J.C.
Ruiz-Orozco, H.
De Los Rios-Obregon, D.
Issue Date: 2002
Abstract: Purpose: To determine the strength of agreement among examiners assessing optic disk stereo-photos, for establishing c/d ratio and presence of glaucoma. Methods: Eighty optic disk photos from 80 patients were assessed in two different sessions by three expert graders (E1,E2 y E3) and three inexperienced ones (NE1, NE2 y NE3); each examiner determined vertical c/d ratio, as well as the presence of glaucoma. Student's t test, ANOVA, bi-variate simple regressions, kappa statistics and coefficient of variation were used. Sensitivity and specificity of assessing stereo-photos were estimated for each examiner. Results: Mean c/d ratio values ranged from 0.48 to 0.62 (first session), and from 0.55 to 0.70 (second session). The weakest agreement between examiners for establishing the presence of glaucoma occurred between NE3 y E1 in the first session (k = 0.31); in contrast, the strongest agreement was observed between NE2 and NE3 during the second session (k = 0.77). Correlation c/d ratio values (r) ranged from 0.6 to 0.89. The sensitivity values for different examiners were reported between 79% and 87%. Specificity values were smaller than the sensitivity values for each examiner (range, 40-80%). A c/d ratio of 0.71 presented optimal sensitivity/specificity values (80% / 79%). Conclusions: Subjective assessment of optic disk stereo-photos has a wide inter- and intra-observer variability that seems to be influenced by examiner level of experience.
Appears in Collections:Producción científica UdeG

Files in This Item:
There are no files associated with this item.

Items in RIUdeG are protected by copyright, with all rights reserved, unless otherwise indicated.