Research Project
10339728
PROJECT SUMMARY There is a global upsurge of falls in older adults that impacts nearly every family across the world. Millions of older adults fall each year in the United States, leading to catastrophic injuries, deaths and soaring healthcare costs. Over the last decade, 911 fall calls have tripled while transport rates to the hospital after a fall have significantly decreased. Instead, 911 is increasingly used for lift assists (falls that do not result in transport). Deployment of emergency medical services for lift assists diverts care from higher acuity emergencies and costs more than 200 million dollars annually in the United States. There is a potentially powerful yet underutilized solution if we leverage the hidden opportunities of fall events, such as lift assists that do not result in catastrophic consequences, to activate prevention strategies. This study aims to develop a scalable strategy for early identification of individuals at high risk of falls and activate prevention solutions. We hypothesize that a systematic 911 fall call intake which has a broader concept of frailty, Frailty And Cognition+Environment (FaCE), will better account for the compounding and cascading nature of fall risks in older adults. At the completion of this project a scalable machine learning model which incorporates FaCE factors to predict high utilization of 911 for falls will be developed. In addition, we will characterize barriers and facilitators for adoption, implementation, and maintenance of fall prevention strategies in the home for patients with FaCE risk factors. This project will utilize a blend of systems science and community-based participatory research approaches and state of the art predictive analytics to elucidate the FaCE of falls, develop a scalable fall prevention solution that can be implemented nationwide and inform a larger-scale implementation trial for using 911 fall calls to activate effective fall prevention strategies in homes.
