Project Summary Caring for the aging population at the end of life presents both a major challenge and opportunity to improve patient-centered care. Aging and dying often necessitate difficult decisions about the receipt of medical care, particularly among older adults with Alzheimer’s disease or Alzheimer’s related dementias Further, these decisions are made without an understanding of the trade-offs between disability and mortality that commonly underlie them. Optimizing advance care planning (ACP) by including such patient-centered data has the potential to enhance shared decision making and avoid undesired outcomes. ACP identifies and clarifies patients’ values and expectations to formulate end-of-life treatment preferences which later inform treatment decisions at the end of life. Too often, though, ACP occurs in a data-vacuum. Providers struggle to accurately prognosticate and patients, particularly those with Alzheimer’s disease or Alzheimer’s related dementias, may misestimate how their preferences will influence outcomes. This project seeks to improve ACP by accomplishing three key goals. First, to determine the influence of end-of-life care preferences on older adult and caregiver outcomes. Second, to explore factors associated with changing end of life preferences with a particular emphasis on Alzheimer’s disease or Alzheimer’s related dementias. Finally, to determine the patient (e.g. Alzheimer’s disease or Alzheimer’s related dementias), provider, and regional predictors of ACP visits and implications on patient-centered outcomes. Only recently has the data needed to inform these critical questions been collected. The National Health and Aging Trends Study (NHATS), a longitudinal survey of Medicare beneficiaries, obtained the first nationally representative longitudinal data on end-of-life care preferences. Additionally, data linkages with NHATS will inform caregiver and end-of-life experiences amongst NHATS respondents. This proposal will be particularly important for older adults with Alzheimer’s disease or Alzheimer’s related dementias because it will inform the optimal timing of ACP in patients with cognitive decline and better inform proxy decision makers. This proposal will serve as the basis for future patient-centered, data-driven ACP with results that can be immediately incorporated by clinicians into existing ACP and will inform researchers and policy-makers on how to optimize ACP.