Abstract
Abstract: :
Purpose: To determine the interobserver and intraobserver variability of the WARMGS in non-expert graduators. Methods: A prospective, observational and longitudinal study. A set of standard and example photographs from the WARMGS was used. Two ophthalmology residents were chosen as graduators. They were trained with 60 cases of Age-Related Maculopathy (ARM) from the set, standardizing methods and operational definitions. Later on they analyzed another 70 different cases. The revision and graduation of the lesions was carried out in an independent way. The answers of both graduators were analyzed in their agreement, exact agreement, agreement inside one step of difference in the category and Kappa. This same strategy was tested out one year later to compare the intraobserver variability. Results: The interobserver variability Kappa values were obtained from 0.589 to 0.9 with agreement forces of moderate to almost perfect. The exact agreements and inside one step in the category went from 72% to 100%. The intraobserver variability Kappa values for graduator 1 ranged from 0.62 to 1.0. For graduator 2 ranged from 0.81 a 1.0. The exact agreements and inside one step in the category for both graduators went from 75% to 100%. Conclusion: The obtained agreements are similar to those reported by WARMGS with expert graduators. The interobserver and intraobserver variability is low with this strategy of training.
Keywords: 308 age-related macular degeneration • 391 drusen • 354 clinical (human) or epidemiologic studies: prevalence/incidence