Fracture Risk Estimation in Post-menopausal Women: A Comparison of Machine Learning Methods and Software Systems.

Challenge

Fracture risk prediction in post-menopausal women relies on established scoring tools, but whether machine learning methods could outperform or complement traditional software-based tools in a real-world commercial insurance population had not been evaluated.

Solution

Target RWE researchers compared ML algorithms against traditional fracture risk estimation tools in a cohort of commercially insured post-menopausal women, evaluating predictive performance and characterizing which patient characteristics drive differential model performance.