Population-based epidemiologic data on the incidence and progression of diabetic retinopathy are important in developing approaches to preventing diabetic retinopathy, in medical counselling and rehabilitative services. The objectives of the Wisconsin Epidemiological Study of Diabetic Retinopathy were as follows: (1) to describe the prevalence, incidence, and progression of diabetic retinopathy and its component lesions, and to determine the incidence of visual impairment in a large population-based cohort; and (2) to determine the relationships between incidence and progression of diabetic retinopathy and risk factors in this cohort. Grading of stereoscopic fundus photographs was performed at the University of Wisconsin Fundus Photograph Reading Centre for photographs from both the baseline examination in 1980-1982 and follow-up examination four years later. Insulin-taking persons diagnosed to have diabetes before 30 years of age had the highest prevalence (71%), four-year incidence (59%) and progression to proliferative diabetic retinopathy (11%) while older-onset persons diagnosed to have diabetes at or after 30 years of age and not using insulin had the lowest prevalence (39%), incidence (34%) and progression to proliferative diabetic retinopathy (3%). These incidence data indicate that a substantial proportion of older onset diabetic people, a group previously thought to be protected from retinopathy, developed it during the four-year interval and underscores the need for careful ophthalmologic examination.