PReventing Injury in Skilled nursing facilities through optimizing Medications (PRISM), a protocol for a cluster randomized trial to reduce injurious falls in post-acute care.

Abstract

Background: Patients who receive post-acute care in a skilled nursing facility (SNF) following a fracture are at high risk for subsequent fall-related injuries. Optimizing medication management may mitigate this risk. This manuscript describes the protocol for a cluster crossover randomized controlled trial titled, PReventing Injury in Skilled Nursing Facilities through optimizing Medications (PRISM), designed to compare the effectiveness of three care models on the rates of injurious falls and other patient-centered outcomes.

Methods: We will enroll 42 SNFs that are sharing electronic health record data with the Long-Term Care Data Cooperative (LTCDC). Matched control facilities will be identified at a ratio of 3:1 based on rural/urban location, profit status, and annual number of post-acute care admissions. Over 6-month periods, in random order, the participating SNFs will sequentially implement three evidence-based care models in a random order: a Deprescribing model, a Bone Health model, and a combined model (referred to as an Injury Prevention model). Patients with recent fractures admitted to participating SNFs for post-acute care during the intervention period will be eligible (target n = 3780). A remote nurse fracture consultant will review medical records, engage in shared decision-making, develop and coordinate a medication optimization plan with SNF and primary care providers, and follow up with the patient and primary care provider upon discharge. The primary outcome is incident injurious falls, measured using Medicare claims data (mean 2-year follow-up). Secondary outcomes include process measures (e.g., adherence with recommendations) and patient-reported outcomes ascertained by telephone survey at 90 days (e.g., medication burden, anxiety, depression, pain, sleep). Safety outcomes will be compared between the three models using Medicare claims data to identify events.

Discussion: This cluster crossover trial aims to compare patient outcomes between each of the three care models and against matched control facilities. Results will inform patients, payors, health systems, and SNF chains of the most effective model to improve outcomes for older adults receiving post-acute care following a fracture.

Trial registration: NCT06304428. Registered on February 25, 2025.