Andrew O. Johnson, PhD is the Manager of Data Science for Mission Health’s Analytics department. In this capacity, he leads a team of data scientists and allied analytics personnel to develop new data science initiatives, enhance organizational data assets, and partner with other Mission teams to achieve data-driven operational improvements in support of Mission Health’s BIG(GER) Aim. Developed in partnership with Mission’s future-minded administrative and clinical leaders, the expanding data science portfolio of Mission Analytics includes: predictive models for readmission risk, length of stay, and workforce retention; enriched patient-level geospatial data; clinical volume forecasting; and social media data mining for clinical inference.
Prior to joining Mission Analytics in 2015, Dr. Johnson served as Senior Analyst for Population Health at the Medical University of South Carolina, and Senior Data Scientist (IT-Advanced Analytics) and Assistant Professor of Health Services Management at the University of Kentucky. He holds degrees in Health Services Policy, Public Health Administration, Biology, and Music from the University of South Carolina; Mathematics from the University of Kentucky; Geographic Information Science from the Pennsylvania State University; and is currently pursuing graduate work in Computer Science from the Georgia Institute of Technology. He also holds an adjunct faculty appointment in the Department of Healthcare Leadership & Management of the Medical University of South Carolina, where he teaches courses in applied statistics, data mining, and research methods in the Master of Healthcare Informatics program.
Predictive analytics is playing an increasingly important role in the care of populations of patients. It identifies patients who need special medical intervention and shows the most effective interventions.
Learn how an MD and chief quality officer, and a PhD and data scientist partnered to achieve improvements. This presentation will describe the organizational assets, team structures, and technical approaches used to add predictive modeling functionality to existing enterprise data warehouses and reporting structures. Learn about the various project management approaches for predictive/data science projects, suggested personnel and data assets required for data science work, and how to avoid technical pitfalls in model development.