Content area
Full Text
Eye (2004) 18, 38
& 2004 Nature Publishing Group All rights reserved 0950-222X/04 $25.00www.nature.com/eyeKL Tu1, P Palimar1, S Sen1, P Mathew1 andA Khaleeli2Comparison of
optometry vs digital
photography
screening for
diabetic retinopathy
in a single districtCLINICALSTUDYAbstractPurpose To compare (a) the clinicaleffectiveness and (b) cost effectiveness of thetwo models in screening for diabeticretinopathy.Methods (a) Retrospective analysis of referraldiagnoses of each screening model in theirfirst respective years of operation and an auditof screen positive patients and a sample ofscreen negatives referred to the hospital eyeservice from both screening programmes. (b)Cost effectiveness study.Participants (1) A total of 1643 patientsscreened in the community and in digitalphotography clinics; (2) 109 consecutivepatients referred to the Diabetic Eye Clinicthrough the two existing models of diabeticretinopathy screening; (3) 55 screen negativepatients from the optometry model; (4) 68screen negative patients audited from thedigital photography model.Results The compliance rate was 45% foroptometry (O) vs 50% for the digital imagingsystem (I). Background retinopathy wasrecorded at screening in 22% (O) vs 17% (I)(P 0.03) and maculopathy in 3.8% (O) vs 1.7%(I) (P 0.02). Hospital referral rates were 3.8%(O) vs 4.2% (I) Sensitivity (75% for optometry,80% for digital photography) and specificity(98% for optometry and digital photography)were similar in both models. The cost ofscreening each patient was d23.99 (O) vs d29.29(I). The cost effectiveness was d832 (O)vs d853(I) in the first year.Conclusion The imaging system was notalways able to detect early retinopathy andmaculopathy; it was equally specific inidentifying sight-threatening disease. Costeffectiveness was poor in both models, in theirfirst operational year largely as a result of poorcompliance rates in the newly introduced
screening programme. Cost effectiveness ofthe imaging model should further improvewith falling costs of imaging systems. Untilthen, it is essential to continue any existingwell-coordinated optometry model.Eye (2004) 18, 38. doi:10.1038/sj.eye.6700497Keywords: diabetes; retinopathy; screening;optometry; digital photographyIntroductionDiabetic retinopathy is the leading cause ofblindness of working age patients in the UK.1Laser photocoagulation is effective ifretinopathy is detected before irreversiblechanges take place.2 Improving metaboliccontrol in patients with mild retinopathy slowsthe progression of retinopathy,3 indicating acase for the detection of early backgroundretinopathy. In a survey of ophthalmologists inEngland and Wales carried out in 1999, therewas no programme of diabetic retinopathyscreening in 23% and in only 64% were theresults routinely sent to the generalpractitioner.4 In 1999, the UK NationalScreening Committee asked the British DiabeticAssociation (now...