Using Machine Learning to Shorten and Adapt Fall Risk Assessments for Older Adults

Abstract

Falls are a leading cause of injury and mortality among older adults, placing significant physical, emotional, and financial burdens on individuals, families, and healthcare systems. The early identification of fall risk and frequent reassessments during rehabilitation are essential for prevention and recovery. However, conventional assessments are time-intensive, rely on multiple motor tasks, and are typically conducted in specialized facilities, limiting their accessibility. This study introduces a novel machine learning-based computerized adaptive testing algorithm that personalizes testing to individual capabilities. The adaptive approach reduces task sequences by over 50% while maintaining high predictive accuracy. It also enables remote testing, predicting performance on complex tasks using as few as 2–3 simpler, accessible tasks. This innovation supports scalable online fall risk screening and frequent balance assessments during rehabilitation, offering a practical and efficient solution for both personalized and community-wide healthcare needs.