Accuracy of vision technicians in screening ocular pathology at rural vision centres of southern India

• Published in: Clinical and experimental optometry 2016-03, Vol.99 (2), p.183-187
• Main Authors: Suram, Vasantha, Addepalli, Uday Kumar, Krishnaiah, Sannapaneni, Kovai, Vilas, Khanna, Rohit C
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Child; Clinical Competence - standards; Eye Diseases - diagnosis; Female; Humans; India; Male; Middle Aged; Observer Variation; Ophthalmic Assistants - standards; ophthalmologist; Ophthalmology - standards; referral; Referral and Consultation; Reproducibility of Results; Rural Health Services - statistics & numerical data; Rural Population; screening; Sensitivity and Specificity; vision centre; Vision Screening - standards; vision technician; Young Adult
• ISSN: 0816-4622; 1444-0938
• DOI: 10.1111/cxo.12345

There is paucity of data on the inter-observer agreement and diagnostic accuracy between a primary-care technician (vision technician) and an ophthalmologist. Hence, the current study was conducted to assess the accuracy of vision technicians, to screen potentially sight-threatening ocular conditions at rural vision centres of southern India and their agreement with an ophthalmologist. In July to August 2010, patients presenting to seven vision centres in Adilabad district (Andhra Pradesh) were selected and screened in a masked manner by seven vision technicians followed by an examination by a consultant ophthalmologist. Agreement was assessed between vision technicians and the ophthalmologist for screening of potential sight-threatening ocular conditions and decisions for referral. The ophthalmologist's findings were considered as the reference standard. Two hundred and seventy-nine patients were enrolled at seven vision centres with a mean age of 32.9 ± 21.8-years. Agreement for screening of ocular pathology was 0.82 (95 per cent CI, 0.8-0.83). There was excellent agreement for cataract (0.97; 95 per cent CI, 0.93-1), refractive error (0.98; 95 per cent CI, 0.96-1), corneal pathology (1.0; 95 per cent CI, 1.0-1.0) and other anterior segment pathology (0.95; 95 per cent CI, 0.9-1); the agreement was moderate to fair for detection of glaucoma suspects (0.43; 95 per cent CI, 0.28-0.60) and retinal pathology (0.39; 95 per cent CI, 0.14-0.63). Sensitivity for screening of anterior segment pathology was 94.6-100 per cent. There was a fair to moderate sensitivity for glaucoma suspect; 35.6 per cent (95 per cent CI, 21.9-51.2) and retinal pathology 26.3 per cent (95 per cent CI, 9.2-51.2). Specificity for screening of ocular pathology was 98.2 to 100 per cent. The kappa (κ) agreement for referral for any pathology was 0.82 (0.8-0.83) As there is good agreement between the vision technicians and the ophthalmologist for screening and referral of anterior segment pathology but moderate to fair for glaucoma suspects and retinal pathology, vision technicians would be a good resource at the primary level to screen for anterior segment pathology; however, they may need further training to detect posterior segment pathology.