Project Summary Pain management in older adults is more important during the post-acute care period than at any other time after hip fracture hospitalization. Pain control can positively impact function and help avert adverse outcomes like delirium, especially among persons with Alzheimer’s disease and related dementias (ADRD). Annually, over 300,000 older U.S. adults are discharged to receive rehabilitation in post-acute skilled nursing facilities (SNFs) after hip fracture, making SNFs a critical place to study analgesic prescribing practices and their impact following fracture. Pain is often inadequately treated following hip fracture, and there are no evidence-based guidelines to guide post-hospitalization pain management. Large clinical trials have not been conducted in post-acute SNFs and until now, observational studies using drug claims data were impossible because drug dispensing data is not captured by Medicare Part D claims for SNF residents. To effectively and safely treat pain in older adults with hip fracture, we must understand the drivers and downstream effects of prescribing different analgesic regimens. This proposal has two specific aims: (1) To describe the individual- (e.g., ADRD) and facility-level (e.g., staffing) predictors of using different analgesic drug regimens, including those containing opioids, in post-acute SNF care after hip fracture; and (2) To evaluate the comparative effectiveness (i.e., pain, functioning) and safety (i.e., rehospitalization) of different analgesic regimens. The rigorous cohort study we propose will combine novel linked administrative and clinical datasets to identify the medication dispensing data that are “missing” from Medicare Part D for >60% of all U.S. SNFs, including over-the-counter and pro re nata (PRN) use. The study population will comprise community-dwelling older adults who were hospitalized for hip fracture between 2013 and 2016, and then admitted to a nursing home for a SNF stay. Subjects will be followed for outcomes after SNF admission, comparing those who were initiated on different analgesic medication regimens in post-acute care. Data for this study will come from (1) a large long-term care pharmacy, which captures drug use during periods of SNF care where claims data from Part D are unavailable; (2) the Minimum Data Set version 3.0, a comprehensive database of standardized clinical assessment data; (3) the Certification and Survey Providence Enhanced Reporting database containing comprehensive facility- level information collected on facilities; (4) Medicare Part A inpatient claims data; (5) Medicare Part D drug claims; and (6) medication administration record data for 125 SNFs. Propensity score approaches and other causal inference methods will be employed. Knowledge gained from this study could provide the first empirical guidance for the management of post-hip fracture pain in older adults, which could ultimately result in improved treatment decisions and subsequent pati...