Prevalence of diabetic retinopathy at first presentation to the retinal screening service in the greater Wellington region of New Zealand 2006-2015, and implications for models of retinal screening.

AIM: To describe the prevalence of diabetic retinopathy (DR) in patients at first presentation for diabetic retinal screening in the greater Wellington region with the intent of service evaluation.

METHODS: This is a retrospective study using data collected from patients newly referred for diabetic retinal screening between 2006-2015 (prevalence analysis, n=12667). The prevalence of DR was calculated by gender, ethnicity, age, type of diabetes and glycaemic control (HbA1c). Chi-square test and multiple logistic regression was used for data analysis.

RESULTS: The prevalence of any DR was 22.5% (n=2852) (non-sight-threatening (NST-DR) n=2562, 20.2%, sight-threatening (ST-DR) n=290, 2.3%). Type 1 diabetes and poor HbA1c control were strongly associated with any degree of DR. Old-age (>65 years), and Asian and Pacific Island (PI) ethnicity had moderately greater odds compared with European. Male gender had marginally increased odds for any DR.

CONCLUSION: This study identified a large proportion (97.7%) of patients (no DR n=9815, 77.5%, NST-DR n=2562, 20.2%) who can be managed in the community by appropriately supported primary care providers, and do not require referral to secondary care ophthalmology. In addition to early detection of ST-DR (2.3%), retinal screening is an early opportunity for education of patients with no DR or NST-DR.