At least 16 million Americans have DM, mostly type 2, with the number expected to rise to 22 million by 2025. Both the American Academy of Ophthalmology (AAO) and American Diabetes Association (ADA) guidelines recommend examination upon diagnosis of type 2 diabetes mellitus (DM) and at least annually thereafter for detection and management of diabetic retinopathy. Yet, only about one quarter of those diagnosed with the disease actually receive the appropriate eye care. The facts that diabetic retinopathy (DR) is the most frequent cause of visual loss in Americans aged 20-74 and that disparities in morbidity and mortality due to diabetes are well documented in African-Americans and other minority populations, bear witness our inadequate detection method. To date, no population-based studies using digital cameras and a simpler (less than seven stereoscopic fields) acquisition protocol have been tested in the United States. The National Eye Institute's Visual Function Questionnaire (NEI-VFQ-25) is a relatively new screening tool, which assesses visual function and vision-specific quality of life. Its use in a high-risk minority population has yet to be studied. 1) Determine whether or not the proposed population-based screening program for detection of DR in high-risk patients with DM is more effective at meeting AAO and ADA screening guidelines than the standard of care (ophthalmology referral by PCP). This program will use digital fundus imaging (DFI) at the time of the primary care encounter, circumventing the need for ophthalmology referral for screening purposes. 2) To determine whether, in this population, glycosylated hemoglobin (HbA1c) is a predictor of the level of diabetic retinopathy based on DFI while controlling for concomitant eye disease, age, race and gender. 3) a) Correlate composite and domain scores of the National Eye Institute's visual function questionnaire-25 (NEI-VFQ-25) with visual acuity (VA) to determine whether an association exists between them in a high-risk, minority population, in which it has not yet been tested.b) To assess visual function and vision-specific quality of life in the DR screening cohort.