More than 65% of patients with blast-related and non-blast-related traumatic brain injuries report vision problems. Traumatic brain injury (TBI) related visual impairment can damage not only the eye, but the visual (sensory and ocular-motor) pathways and/or cortical processing areas. Thus, the traditional objectives and treatment plans for eye-related blindness or low-vision may not be appropriate for combat and TBI related vision injuries. In 2008, the Veteran Health Administration (VHA) published Directive 2008-065 titled “Performance of Traumatic Brain Injury Specific Ocular Health and Visual Functioning Examinations for Polytrauma Rehabilitation Center Patients” specifying that all patients with a diagnosis of TBI who are admitted to VHA Polytrauma Rehabilitation Centers (PRCs) have a TBI-specific ocular health and visual functioning examination performed by an optometrist or ophthalmologist. Several studies have identified the prevalence and symptomology of this cohort; however, the service delivery data generated by this mandated exam have not been analyzed. The purposed study will evaluate how a selected attempt to organize the structure of care (Directive 2008-065) affects access to and utilization of vision health care services. The Specific Aims are to (1) Determine the pattern of access to and utilization of services among Veterans with TBI with visual and ocular deficits; (2) Describe rehabilitation service delivery; and (3) Explore clinical recovery as measured by visual and ocular outcomes. This population-based study will use a retrospective cohort design with a prospective survey component to identify access barriers to follow up appointments. The population is defined as all Veterans and active duty service persons with TBI admitted to one of the five VHA PRCs (approximate n=2,500). Patient identifiers (both Veteran and active duty) are provided by the Physical Medicine and Rehabilitation National Program Office. Using the VA Informatics and Computing Infrastructure workspace, both structured and unstructured data will be used to create an analytic data set. Structured data will be extracted from relevant data tables from the Corporate Data Warehouse. Unstructured data will be extracted from text documents stored in the electronic health record, also housed in the Corporate Data Warehouse, using an Information Extraction system. Extraction methods include manual chart review by experts, natural language processing, named entity recognition, machine learning and ad-hoc methods like handcrafted rules or regular expressions. Extracted data will be validated by an Expert Panel composed of oculomotor and vision experts from each of the five PRCs, the Polytrauma System of Care, and pre-911 vision and blind services clinics. Geographic Information System tools will be used to determine geographic access (both drive distance and drive time) to health services. The final dataset will consist of diverse variables including patient characte...