ABSTRACT This proposal responds to the National Eye Institute’s (NOT-EY-22-004) call for a better understanding of the connection between social determinants of health (SDOH) and eye health. SDOH refers to “conditions in the environments where people are born, live, learn, work, play, worship, and age that affect a wide range of health, functioning, and quality-of-life outcomes and risks”. Through a systematic review, we identified research gaps regarding SDOH and their relationship with dry eye. Dry eye, a multifactorial disease of the tear film and ocular surface, is a global health problem with considerable impact on quality of life. Dry eye is one of the most common reasons for patients to seek eye care and the societal cost of managing dry eye is substantial. Leveraging multi-center electronic health records (EHRs), we propose to identify indicators of SDOH that are associated with disparities in the diagnosis, management, and prognosis of dry eye. In Aim 1, utilizing a single-center EHR dataset from the Sue Anschutz-Rogers Eye Center, we will explore factors contributing to differential access to care and prognosis among dry eye patients. These potential indicators (or their upstream factors) for SDOH include race and ethnicity, preferred language spoken, and other individual characteristics that are routinely documented in the EHR. We aim to build multivariable prognostic models for generating personalized risk assessment on dry eye progression over a 2-year period, considering important individual and structural barriers to accessing care. In Aim 2, leveraging the multi-center EHR datasets curated by the Sight Outcomes Research Collaborative (SOURCE) Consortium, we will quantify geographical variations in the impact of SDOH on dry eye prognosis. We will validate our initial findings and the prediction models developed in Aim 1 while accounting for additional center-specific factors such as density of eye care professionals. With data from more than 4 million unique patients, the multi-center SOURCE datasets offer substantial sample sizes suitable for external validation. We anticipate providing up-to-date evidence on the association between SDOH and the diagnosis, management, and prognosis of dry eye. By incorporating these routinely collected information as predictors in the rigorously constructed and externally validated prognostic models, we aim to provide physicians with updated evidence for disease management and enhance long-term outcomes and quality of life for dry eye patients. Our findings will inform the development of intervention programs addressing specific factors contributing to disparities in access to dry eye care.