OBJECTIVE To investigate the association of intraocular pressure (IOP), ocular perfusion pressure, and myopia with the incidence and progression of diabetic retinopathy. METHODS The design is a cohort study of a population-based sample (n = 1210) of persons with younger-onset diabetes (diagnosis was made before 30 years of age, and subjects were taking insulin) and a random sample (n = 1780) of persons with older-onset diabetes (diagnosis made after 30 years of age). Baseline and 4-year follow-up examinations were completed by 891 younger-onset and 987 older-onset persons. Retinopathy was graded from stereoscopic fundus photographs. Endpoints were incidence of retinopathy, progression of retinopathy, and progression to proliferative diabetic retinopathy (PDR). Ocular perfusion pressure was calculated from IOP and blood pressure. Myopia was a refractive error of -2 diopters or less. RESULTS In univariate analyses, ocular perfusion pressure was associated with incidence of retinopathy (P < 0.005), progression of retinopathy (P = 0.07), and progression to PDR (P < 0.001) in the younger-onset group but not in older-onset subjects taking or not taking insulin. Intraocular pressure and myopia were not associated with any endpoint in any group. Using logistic regression to control for covariates, ocular perfusion pressure was significantly associated only with incidence of retinopathy in younger-onset persons. The odds ratio for a 10-mmHg increase in ocular perfusion pressure was 2.13 (95% confidence interval, 1.30-3.50). Also, myopia was protective for progression to PDR in younger-onset persons with an odds ratio of 0.40 (95% confidence interval, 0.18-0.86). CONCLUSIONS These results suggest that pressure phenomena may be related to the development of retinopathy in younger-onset persons. This would have implications for treatments affecting both IOP and blood pressure.