||PURPOSE: The aim of the study was to describe the prevalence of visual impairment and retinopathy and to investigate risk factors for retinopathy in persons with diabetes, screen-detected diabetes, impaired glucose tolerance and normal glucose tolerance in a subpopulation of the HUNT study. METHODS: We used a sample (n = 163) from a population-based screening survey of hyperglycaemia, undertaken in 2004-2005 in Verdal, Norway. Baseline information was accessible through the second Nord-Trondelag Health Study (HUNT2), 1995-97. Data collection was made in 2005 and included patient history, refraction, visual acuity, cataract assessment and single-field, nonmydriatic retinal photography. Retinal photographs were graded independently by two graders blinded to patient information. Data were analysed with standard statistical methods, and p < 0.05 was considered significant. RESULTS: In all, 126 (77%) persons participated, 55% were women. The mean (SD) age was 59 (+/- 14) years. Four (3%) had correctable visual impairment, and none were visually impaired. Retinal photographs were gradable for both eyes in 109 (87%) participants. The prevalence of retinopathy was 11% in persons with known diabetes, 4% in persons with screen-detected diabetes, 3% in persons with impaired glucose tolerance and 10% in persons with normal glucose tolerance. Retinopathy was not associated with known history of diabetes or current glycaemic status. Nonfasting plasma glucose (in 1995-97) was an independent risk factor for retinopathy (in 2005), OR (95% CI) 1.5 (1.01, 2.13), p = 0.046. CONCLUSION: The prevalence of diabetic retinopathy in persons with diabetes in this study was low. Appropriate optical correction and regular eye examination can prevent unnecessary visual impairment in both persons with and without diabetes.