Diagnosis of Eye Diseases in Children by Field Investigators Using External Photographs vs. Clinical Examination

Abstract

Purpose: : 1. To test the case detection rate by the Field Investigators (FI) when shown external photographs of eyes with and without ocular disorders. 2. To standardize the FI with the ophthalmologist and to analyze the interobserver agreement. 3. To compare the case detection rates with the above 2 techniques.

Methods: : Ten FI were chosen to screen 30,000 children for eye diseases in the Pavagada Pediatric Eye Disease Study in 2 taluks of Karnataka state in South India. The FI were trained in 3 phases: 1.Lectures and external photographs of children with ocular disorders were used to familiarize them with eye diseases. 2. Trained to detect eye diseases in children by the ophthalmologist at the rural hospital at Pavagada. 3. Trained by the pediatric ophthalmologist in a tertiary hospital. After this the FI's underwent a test wherein they were shown 30 slides of which 5 were normal eyes and 25 eyes with pathology and were asked to mark the probable abnormality.The FI were then standardized to the ophthalmologist in the Pavagada Rural Eye Hospital. The ophthalmologist examined 134 school children and the same were examined by 4 teams of FI. Results were analyzed using the kappa statistics for inter observer agreement.

Results: : In the test conducted by using photographs, 65.3% of the slides were answered correctly. True positive was 58.3%, 9% was false positive & false negative was 4%.In the standardization procedure, 82% were true positives, 6.7% were false positives and 11.1% were false negatives. The kappa statistics between the ophthalmologist and team 1, 2, 3 and 4 were 98.2, 98.5, 97.9 and 94.9 respectively.Using photographs has many disadvantages. The corneal reflex on the photo is mistaken for a white opacity on the cornea. Eyes not in primary position are mistaken for a squint. The grayish reflex of a lamellar cataract is missed.The advantages are that it is quick and easy to conduct.

Conclusions: : Though easy to conduct, using photographs of children with eye disorders to test FI is not an alternative to actual clinical testing since the case detection rate is less with this technique.

Keywords: clinical (human) or epidemiologic studies: biostatistics/epidemiology methodology • screening for ambylopia and strabismus • detection