Diabetic retinopathy (DR) is a leading cause of blindness in working-aged persons in the United States. Black, Latinx, Indigenous and persons living in low resource communities are less likely to be screened present at later stages in the disease and are less likely to follow-up. There is a gap in our understanding of how to optimize DR screening in persons historically under-screened. Digital health applications have the potential to increase awareness and increase healthy behaviors. We leverage our DR screening determinants model and risk calculator for the development of sight threatening DR to address health disparities. The long-term goal of our research is to contribute to a decrease in the incidence of preventable blindness from DR. The overall objective of this proposal is to develop a novel community-responsive DR digital health tool. Our central hypothesis is that we can design a tool optimized for real-word usability to facilitate DR screening in persons with diabetes. Our hypothesis is built on DR risk stratification models, known health disparities in DR, and our determinants framework for DR screening in New Haven.