1 Background: Every day, older Veterans receiving VA-financed home care experience unacceptably high rates 2 of health-related adverse events (e.g., infections and injuries), many of which, if their early warning signs are 3 caught, can be prevented. Patients with both dementia and other chronic conditions are at even higher risk. 4 Home care prevents costly nursing home placement and maintains patients in their environment of choice. Yet 5 family caregivers and home care agency workers, who provide home care do not receive training to detect 6 early warning signs, because available training does not target these providers. And VA patient safety 7 initiatives using high-reliability organization (HRO) principles targets only VA-owned health care facilities. This 8 CDA-2, informed by HRO principles, will fill this gap and addresses early warning signs of adverse events in 9 home care. The CDA-2 will estimate prevalence rates of potentially avoidable hospitalization in home care; 10 assess the roles of family caregivers, home care agency workers, and VA clinical teams, in the recognition of 11 early warning signs; and adapt a solution to detect early warning signs. My background in gerontology and 12 preliminary work funded by my VISN 1 CDA partially prepare me for this work. But I need additional training in 13 non-VA data sources and advanced statistical methods for causal inference, mixed methods, care 14 coordination, and cutting-edge user-centered design to advance my career and have the CDA-2 succeed. 15 Specific Aims: My proposal fills training gaps and provides VA with rigorous, actionable research. With 16 operational partner Office of Geriatrics and Extended Care, I will achieve 3 aims with my mentors and 17 statistical experts: (1) estimate rates of potentially avoidable hospitalizations in home care and assess how 18 their prevention is related to home care and VA clinical team characteristics; (2) produce in-depth knowledge of 19 home care processes through which family caregivers, home care agency workers/supervisors, and VA 20 clinicians prevent early warning signs from escalating; and (3) adapt and pilot an existing tool to bolster the 21 detection of early warning signs in the home. 22 Methods: Aim 1: Use of Medicare (non-VA) and VA databases will enable me to estimate potentially avoidable 23 hospitalizations in my study population and assess whether their prevention is associated with types of home 24 care and VA clinical teams. Aim 2: Using mixed-methods (in-depth interviews and surveys) at 3 VAMCs 25 selected for rates estimated from Aim 1 data, I will examine multiple processes, including care coordination, 26 through which family caregivers, home care agency workers, and VA clinical team members recognize and 27 manage early warning signs. Aim 3: I will use evidence from the literature and Aim 2 findings to implement a 28 user-centered design approach to adapt and pilot an existing tool to enhance family caregivers and...