search

Leveraging AI to Improve Prioritization of Care Management

  • December 16, 2019
  • 3:00 PM – 4:00 PM ET
  • online
about

Care managers with limited resources must prioritize patients for outreach. Segmenting the population based on the level of intervention required at a point in time helps care managers focus on members of the population with the greatest need. Predictive models add additional information regarding risk at an individual level for specific adverse outcomes, such as hospitalization or increasing cost.

In addition to using machine learning for the predictions, new AI techniques are bringing in focus the relevant drivers that have high impact on local estimates in the model outcomes. This information, coupled with domain expertise, give the practitioners better understanding of the underlying reasons for the patients’ predictions.

Attendees will learn:

  • Advantages of using a time sensitive population classification algorithm in care management
  • How predictive risk models, including risk of hospitalization and risk of rising cost, can be used by care managers to prioritize patients for outreach
  • Application of AI in classification of health care populations
  • Use of local interpretable model agnostic explanations (LIME) in interpreting factors affecting classification

Speakers


Janet Young, MD, MHSA,
Lead Clinical Scientist
IBM Watson Health

Janet Young, M.D., M.H.S.A. is a lead clinical scientist at IBM Watson Health. She has over 30 years of experience in development of healthcare analytics related to measurement and improvement of clinical performance, as well as design of predictive models of healthcare outcomes. She has provided guidance in methodologies used to classify or group claims for greater interpretability, including episodes of care, disease staging and procedure groupers. Recent work has included development of a time sensitive population classification methodology. Dr. Young earned her medical degree from Yale University School of Medicine and MHSA degree from University of Michigan School of Public Health.


George Sirbu, Ph.D.
Lead Data Scientist
IBM Watson Health

George is a lead data scientist at IBM Watson Health. Over the course of more than 10 years, he has developed various methodologies used in Care Management, Provider Performance or Consumer Transparency tools. He is focused on applying new innovative data science algorithms to solving healthcare problems in both payer and provider markets. Recent work includes predictive models for healthcare outcomes and population classification management. George Sirbu earned his Ph.D. degree in statistics from Michigan State University with a dissertation in Adaptive Designs with Covariates.


Kevin Ruane
Analytic Leader
IBM Watson Health

Kevin Ruane is an Analytic Leader for IBM Watson Health. Kevin specializes in working with clients to apply advanced analytic methods to help solve complex healthcare problems, including population heath, performance measurement and financial analysis. Kevin has over 25 years of experience in health care, working in many facets. He currently manages a team of analytic methodology subject matter experts, implementing advanced analytics into client environments. Kevin earned a Bachelor of Science degree from Michigan State University.