Use of advanced text analytics and automated reporting to improve CDC emergency preparedness through the Division of Emergency Operations

Objective/Purpose: Through the partnership between the Division of Emergency Operations (DEO), Office of Public Health Preparedness and Response (OPHPR) of the Centers for Disease Control and Prevention (CDC) and Booz Allen Hamilton, a two pronged advanced analytics and IT methodology was developed to address the needs for data standardization and trend analysis of reports and documentation used for planning, responding to, and recovering from public health emergencies. Methods: Booz Allen responded to the need by 1) implementing advanced text analytics to extract useful and high-value information from unstructured historical data, and 2) organizing, standardizing, and structuring reporting tools for future data extraction through a web-based, automated reporting system. This two pronged methodology supports the analysis and identification of data, trends, and rapid information retrieval that is needed in public health emergency preparedness and response. Results: By utilizing natural language processing algorithms to mine unstructured legacy data and by providing tools to standardize new data collection and retrieval, Booz Allen Hamilton was able to significantly decrease man hours needed for manual and repetitive data entry and provide quantitative and qualitative answers to the questions of how effective and prepared is the CDC in responding to public health emergencies.