Epidemiological study of diabetic retinopathy in a primary care setting in Hong Kong.

ABSTRACT

OBJECTIVES: To estimate the prevalence and risk factors of diabetic retinopathy in type 2 diabetic patients, and to investigate the difference in retinopathy progression in patients with normal fundi or established retinopathy at baseline and the risk factors implicated in the progression. DESIGN: Retrospective community-based study. SETTING: Ten primary care clinics in Hong Kong. PATIENTS Type 2 diabetic patients; subsidiary analysis included subjects with more than one screening event. MAIN OUTCOME MEASURES: Patient demographics, baseline prevalence, and risk factors of diabetic retinopathy; progression of retinopathy in patients with normal fundi and established retinopathy at baseline, and the associated risk factors. RESULTS: A total of 6165 patients were recruited from January 1998 to May 2004. Primary analysis included 4423 patients with good-quality retinal photographs. The mean age of the patients was 60.36 years (standard deviation, 10.80 years; range, 28-94 years), the mean duration of diabetes was 4.71 years (standard deviation, 4.67 years; range, 0.1-40.6 years), and the mean level of glycated haemoglobin was 7.47% (standard deviation, 1.44%). The prevalence of retinopathy at baseline was 28.4%. Subsidiary analysis showed progression to sight-threatening retinopathy was more common in the group with baseline retinopathy than that without (7.9% vs 0.7%), and occurred at a faster rate (mean, 1.5 [range, 0.5-3.0] vs 2.0 [1.0-4.2] years). Logistic regression revealed that the level of glycated haemoglobin was positively associated with both the onset (P<0.001) and progression of retinopathy (P=0.03). CONCLUSION: Optimal glycaemic control is important for reducing sight-threatening retinopathy. Close observation is required for patients with established retinopathy as progression occurs more rapidly.