Comparison of Diabetic Retinopathy Detection by Clinical Examinations and Photograph Gradings

Objective: To compare the use of clinical examination and fundus photograph gradings in detecting diabetic retinopathy in a population that includes persons with and without diabetes. Design: Population-based epidemiologic study. Setting: Sir Winston Scott Polyclinic, Bridgetown, Barbados, West Indies. Participants: Subset of a random sample of the country's population aged 40 to 86 years. Results: Among 1168 black persons with fundus photograph evaluations, ophthalmologic examinations, diabetes history, and glycated hemoglobin data, 21% reported a history of diabetes; 9.5% had definite diabetes (glycated hemoglobin >11.5%); and 13.3% had a diabetes history and glycated hemoglobin value less than or equal to 11.5%. The frequency of diabetic retinopathy in this group was 7.7% (90/1168) by clinical examination, 8.7% (102/1168) by photograph gradings, and 6.7% (78/1168) by both methods. Conclusions: These results suggest that in certain populations that include diabetics and nondiabetics, a clinical examination by an ophthalmologist will detect most cases of diabetic retinopathy identified by disc and macula photographs read by skilled graders. However, it will lead to an underestimate of prevalence. Staff availability and cost, issues not examined in this study, should determine which approach is selected.