Past Meetings

Focus on Predictive Analytics

7/20/18  Back

Our July meeting of Health TechNet was held on Friday, July 20 at Nelson Mullins’ offices in Washington, D.C.

This month, we focused on predictive analytics in healthcare, with a special focus on three critical tasks: 1) getting the data, 2) finding actionable opportunities, and 3) taking action.

Our first presenter was Dr. Nick van Terheyden, MD.  "Dr. Nick” is a leader in Digital Healthcare and Innovation,  and is CEO and founder of Incremental Healthcare, where he provides strategic insights and guidance to support healthcare organizations, medical professionals and patients through information-enabled healthcare. He brings an incremental approach to developing successful strategies and applies his expertise to achieve a technology environment that is interconnected, efficient and patient-focused. Dr. van Terheyden’s topic was "Minority Report Comes to Healthcare.”  Welcome to the age of data – the revolution has arrived made possible by new sources of data available in every industry and every walk of life.  Can we use this data to identify the patients, through predictive analytics, that are sick before their illness manifests in our healthcare system offering an opportunity to intervene and mitigate or even prevent disease? This presentation explored this exciting new potential.

Our second presenter was Dr. James Sorace.  Dr. Sorace is a pathologist with over 10 years of prior experience in the Office of the Assistant Secretary for Planning and Evaluation and 3 years prior experience at the Centers for Medicare and Medicaid Services (CMS) both within the Department of Health and Human Services. Dr. Sorace discussed the concept of Disease-Space (DS), which is the frequency distribution of individual combinations of co-morbidities at the population level. This presentation presented the results of two studies based on Medicare claims data that estimate the complexity of DS in the Medicare population and the ability of heredity to predict it.  This provides clarity on realistic expectations of predictive analytics’ capability to understand the Medicare and Medicaid populations from available data.

The meeting's organizer Dr. Joe Bormel, wrapped up the program.   He presented a case study of the FDA’s Sentinel system demonstrating what can be done with predictive analytics at the largest possible scale.  The FDA's prime use case is active post-market surveillance of medical products.  Given it’s unique public health authority and innovative distributed research network of Data Partners, Sentinel highlights a tremendous model and capability that has proven its ability to answer outcome relationship questions scientifically at a scale previously not possible.