Abstract Currently over 3 million Veterans have adopted the VA Patient Portal, My HealtheVet (MHV), however actual use of the portal varies considerably, and we have little data on how the portal features can best improve patient outcomes, or whether effective use varies by subgroups. We will work to define effective use, using clinical and administrative data, coding of secure messages, patient surveys, and in-depth interviews with extensive MHV experience. Lessons from this work will be used to refine an intervention to support effective use of MHV portal features for chronic disease management. We have chosen to anchor our research in a particular disease area: Type 2 diabetes, a complex disease affecting many Veterans associated with significant mental and physical comorbidities, complications and negative outcomes, and requiring high levels of self-management. Among a national sample of Veterans with uncontrolled diabetes, Aims 1 and 2 use a cohort-with-nested-case- control design to generate needed information to personalize interventions and to define effective use of the patient portal. Aim 3 is a small-scale experiment designed to test a MHV supported adoption intervention. Our Specific Aims are to: Aim 1. Generate new information for feature selection and personalizing interventions. In a national longitudinal cohort of Veterans with uncontrolled type 2 diabetes, we will test the association of patient portal tool use with improvements in physiologic measures of control (H1a), and the association will differ for patients with mental health conditions and those in rural areas (H1b). Aim 2. Generate details about how patients are effectively using the patient portal. We will collect data on patients with good glycemic (HbA1c) control (cases) and with poor control (controls). Using surveys and secure message coding, we will examine differences in communication with the clinical team (information sharing, information seeking, socio-emotional exchange), and in sharing information from their portals with family and/or non-VA providers. Aim 3. Conduct a small-scale experiment, using a randomized encouragement design, to test the feasibility of engaging Veterans in supported adoption intervention to promote effective MHV use, in a sample of 200 Veterans with diabetes (with or without comorbid mental health problems) For Aim 3, we will design a supported adoption intervention leveraging the features associated with the greatest evidence in Aims 1 and 2, seek feedback on the intervention from key stakeholders (e.g., primary care providers, staff, Veterans with diabetes), and then randomize patients at the Bedford/Boston VAMCs to be invited to participate in the intervention. Evaluation will focus on impact on increased use of MHV and feasibility of conducting a large scale implementation of the refined intervention in a future IIR. This research will develop and test an intervention to improve diabetes care, while strengthening the evide...