Aviator Certification Period Health Forecasts Using Claims Data

This research project addresses the question of how the Office of Aerospace Medicine can better use medical data for timely, risk-based airman medical certification decision making in an environment of rapid change in both healthcare and aerospace operations. The research seeks to develop and validate tools, techniques, and procedures, particularly in the areas of big data and machine learning, which will form the technological foundations to implement a next generation airman medical certification safety management system. This research seeks to maximize value and avoid duplication by transferring relevant big data analytics and techniques for use with agency medical certification data. It also leverages very large healthcare datasets assembled by private actors and current big data analytics and techniques to enable precision-based (i.e., more individualized vs. population-based) aeromedical risk assessments, which cannot be developed from existing agency medical certification data because of limitations in data quantity and quality. Additionally, calculating aeromedical risk estimates using commercial healthcare datasets provides a mechanism to synchronize aeromedical certification decision making with the current state of the art in clinical medicine, pharmaco-therapeutics, medical devices, etc.