Evaluation of screening schemes for eye disease in a primary care setting

Abstract

Background and purpose: Screening in a primary care setting could be an effective method for detection of eye disease. This study was designed to evaluate a questionnaire and a battery of tests for their performance in eye disease screening at a primary care clinic. Methods: 405 patients aged 40 years or older were interviewed and received a comprehensive eye examination including visual acuity and visual field testing, tonometry, slit-lamp examination, dilated fundus examination and photography. Sensitivity and specificity for the identification of eye disease were calculated for each test and various combinations of tests. Results: A questionnaire-based algorithm for detection of overall eye disease was sufficiently sensitive (90%) but less specific (44%) than an ideal screening test. Distance visual acuity with presenting correction of =20/40 had a sensitivity of 61% and specificity of 72%. A dilated fundus examination had a sensitivity of 79% and specificity of 82%. In screening for glaucoma, tonometry was ineffective (sensitivity=27% and specificity=96%), while visual field testing by suprathreshold screening had a sensitivity of 70% and specificity of 67%. Among a variety of combinations, a two-stage strategy with the questionnaire as a first-stage and visual acuity and ophthalmoscopy as second-stage tests provided the best balance of sensitivity (83%) and specificity (76%). Conclusion: Among currently available ophthalmic tests, an eye examination including a thorough fundus examination is critical in detection of eye disease. A five-item questionnaire may be useful to identify patients at high risk in primary care practice. More effective tests are needed to improve performance of eye disease screening.