Project Abstract All persons with Alzheimer’s disease and related dementias (ADRD) will experience behavioral symptoms of dementia (BSD). Management of BSD is one of the most frequent daily tasks caregivers perform, which includes deciding when to administer an “as needed” medication to address BSD. Non-pharmacologic strategies are the recommended first-line approach to manage BSD, though “as needed” medications can be warranted when non- pharmacologic strategies fail and the BSD are either causing distress to the care recipient or jeopardizing the physical safety of the care recipient or caregiver. Conversely, chemical restraint means a drug is being administered to make care more convenient for the caregiver, not because it is medically indicated. Our team has reported that 22% of ADRD family caregivers inappropriately administer medications as chemical restraints. Importantly, there are no evidence-based indications for pharmacotherapeutic management of a patient symptom for the benefit of a caregiver, aside from safety. Thus, this high prevalence is concerning, and represents an emerging public health crisis and major gap in knowledge. Given the novelty of this newly described phenomenon, the purpose of this study is to describe patterns of medication administration, chemical restraint, and associated mortality outcomes in ADRD family caregiving. To address this, we will conduct a 31- day micro-longitudinal study in which family caregivers (N=300) complete daily diaries describing BSD management including use of non-pharmacologic strategies and medications. These data will be supplemented with personal characteristics of the caregivers obtained through an enrollment survey, clinical characteristics of care recipients found in their EHR and combined with vital records data on the care recipient’s mortality. This proposed study will implement an innovative algorithm to classify medication administration as medically indicated, potentially inappropriate and chemical restraint in consideration of non-pharmacologic strategies tried, safety concerns, and distress of the care recipient (Aim 1). This study will also characterize patterns of chemical restraint to create typologies and determine characteristics of chemical restraint users (Aim 2), and determine the relationship between chemical restraint use and 1-year mortality of the care recipient (Aim 3). This study is the first step towards creating sensitive clinical assessments to identify situations at high risk of adverse outcomes related to caregiving practices. This foundational work will support a long-term research goal to evaluate if randomizing caregivers to screening for chemical restraint results in greater reduced morbidity and mortality for care recipients than the observed harms.