PROJECT SUMMARY Patients with Alzheimer’s disease & related dementia (ADRD) experience 400,000 avoidable hospitalizations annually, amounting to $5.4 billion in preventable healthcare costs. In addition to financial costs, avoidable hospitalizations increase risks for adverse outcomes, such as secondary infections, deliriums, or acute distress. To prevent hospitalizations, clinicians rely on caregivers of patients with ADRD to report any physical, behavioral, and emotional changes that caregivers observe. If communicated in a timely manner, many such changes can be addressed without hospitalization, but through pharmacological interventions, home visits, or clinic visits. Currently, communication between clinicians and family caregivers depends on the caregiver knowing when to call, and clinicians sorting through messages of various importance to find and address worrisome changes associated with an impending hospitalization. This strategy is insufficient, as it delays communication of potentially significant changes in patients with ADRD to the clinician. It also relies on caregivers’ ability to distinguish innocuous changes from those that can lead to hospitalizations. To minimize occurrence of avoidable hospitalizations in patients with ADRD, we will develop and test the first remote patient-monitoring platform, Digital Outpost, based on caregiver-reported information. The platform will contain two parts: the native Caregiver App and web-based Clinician Action Dashboard. Guided by user-centered design principles and rapid qualitative research methods, we will query 12 caregivers of patients with ADRD regarding content, look, feel, and experience of the Caregiver App. Using input of 20 clinicians through Delphi and discrete choice methodology, we will define the daily clinical surveys through which caregivers will report physical, behavioral, and emotional changes in patients with ADRD. We will also define the algorithms for analyzing and displaying caregiver-reported data in the Clinician Action Dashboard. Then, we will develop the supporting platform prototype using Agile/Kanban-based Software Development methodology and conduct a 14-day pilot study of Digital Outpost with ten primary caregivers of patients with moderate ADRD. Primary outcome of the pilot will be usability, measured by the System Usability Scale, with a follow-up assessment through rapid qualitative methods. Secondary outcome will be feasibility, measured through platform usage statistics. Upon completion, we will be poised to update the platform to include integration of other smartphone data (e.g. walking steps), use a machine-learning based predictive algorithm for hospitalization, and test efficacy of Digital Outpost in reducing all-cause hospitalization through multi-site trial in a Phase II STTR application.