The accuracy of the Edinburgh Red Eye Diagnostic Algorithm

Abstract

Purpose To assess the diagnostic accuracy of the Edinburgh Red Eye Algorithm. Methods This was a prospective study. A questionnaire was designed and made available to clinicians referring patients to the acute ophthalmology service within Edinburgh. The questionnaire involved them using the algorithm to reach a diagnosis in patients presenting with red eye(s). Patients were then referred to the emergency eye clinic and the questionnaire faxed to the clinic or sent with the patients. Patients were then examined by an experienced ophthalmologist (not blinded) to reach a 'gold standard' diagnosis. The concordance between the 'algorithm assisted' diagnosis and the 'gold standard' was then assessed. Results All patients presenting with red eye(s) were eligible for inclusion. Forty-one questionnaires were completed, two were excluded. The algorithm assisted diagnosis was correct 72% (28/39) of the time. It correctly diagnosed: acute angle closure glaucoma in 100% of cases (4/4); iritis in 82% (9/11); stromal keratitis in 63% (5/8); epithelial keratitis in 70% (7/10); and infective conjunctivitis in 50% (3/6). Discussion The diagnostic accuracy of The Edinburgh Red Eye Diagnostic Algorithm is 72, rising to 76% when only the most serious red eye(s) causes are included. The diagnostic accuracy of nonophthalmologists when assessing patients presenting with red eye(s) is greater when the algorithm is used. We hope that the use of this algorithm will prevent delayed presentations of certain serious eye conditions and reduce the morbidity from delayed treatment.

Peek Community Eye Health - mHealth system to increase access and efficiency of eye health services in Trans Nzoia County, Kenya: study protocol for a cluster randomised controlled trial

“…Effective task shifting with clear referral criteria and management plans has been successfully delivered through algorithms such as the Integrated Management of Childhood Illness (IMCI) at primary level [29,30]. In eye care, decision trees and algorithms have been developed, mostly outside Africa, and focused on identifying the diagnosis and treatment at a secondary level [31][32][33]. The World Health Organization (WHO) recently developed similar algorithms and training manual for use at the PHC facilities in Africa [34].…”

Comment on ‘The accuracy of the Edinburgh Red Eye Diagnostic Algorithm’

“…We congratulate the authors on the creation and evaluation of their Red Eye Algorithm, for use in primary care settings. 1 This algorithm suggests that bilateral red eyes, with signs and symptoms of infective conjunctivitis, should be given 2 weeks of chloramphenicol drops before review by their GP. However, we would suggest that, because of its prevalence and wider health implications, primary care practitioners should be encouraged to consider the possibility of chlamydial conjunctivitis and initiate investigations when appropriate.…”

The accuracy of the Edinburgh diplopia diagnostic algorithm

“…This is the last in a series of three studies assessing the accuracy of diagnostic algorithms for each of the three most commonly encountered ophthalmic presentations: diplopia, red eye(s) 1 and visual loss. 2 Algorithms have been used in medicine for over 30 years.…”