Aims/hypothesis Individuals with type 2 diabetes mellitus may experience an asymptomatic period of hyperglycaemia, and complications may already be present at the time of diagnosis. We aimed to determine the prevalence of diabetic retinopathy in patients with newly diagnosed (screeningdetected) type 2 diabetes. Methods The Gutenberg Health Study is a population-based study with 15,010 participants aged between 35 and 74 years. We determined the weighted prevalence of diabetic retinopathy by assessing fundus photographs. Screeningdetected type 2 diabetes was defined as an HbA 1c concentration of 6.5% (47.5 mmol/mol) or more, no medical diagnosis of diabetes and no intake of insulin or oral glucoselowering agents. Results Of 14,948 participants, 1377 (9.2%) had diabetes mellitus. Of these, 347 (25.2%) had newly diagnosed type 2 diabetes detected by the screening. Overall, the weighted prevalence of screening-detected type 2 diabetes was 2.1%. Fundus photos were evaluable for 285 (82.1%) participants with newly diagnosed diabetes. The weighted prevalence of diabetic retinopathy in screening-detected type 2 diabetes was 13.0%; 12% of participants had a mild non-proliferative diabetic retinopathy and 0.6% had a moderate nonproliferative diabetic retinopathy. Diabetic retinopathy was proliferative in 0.3%. No cases of severe non-proliferative diabetic retinopathy or diabetic maculopathy were found. Thirty (14.9%) of 202 and six (7.2%) of 83 individuals with and without concomitant arterial hypertension, respectively, had diabetic retinopathy (OR 2.54, 95% CI 1.06, 7.14). Visual acuity did not differ between individuals with and without diabetic retinopathy . Conclusions/interpretation In this large European study, the prevalence of diabetic retinopathy in screening-detected type 2 diabetes was 13%. Only a very small proportion of participants with detected diabetic retinopathy needed treatment.