NOV 03, 2016 09:00 AM PDT
KEYNOTE: Big data and the learning healthcare system: practical applications for the clinical laboratory
Presented at the Clinical Diagnostics & Research Virtual Event
CONTINUING EDUCATION (CME/CE/CEU) CREDITS: P.A.C.E. CE | Florida CE
2 8 159

Speakers:
  • Assistant Professor of Pathology Director, Point of Care Testing Associate Director, Chemical Pathology; University of Michigan Hospital
    Biography
      Dr. Lee Schroeder is currently Assistant Professor at the University of Michigan where he is Director of Point-of-Care Testing and Associate Director of Chemical Pathology. His academic focus is at the interface of clinical informatics and health services research, using decision analytic approaches to model and improve the impact of laboratory medicine. This has included research involving laboratory capacity and quality improvement of sub-Saharan African laboratories, modeling the impact of different testing algorithms on health outcomes, and data mining the electronic medical record to better understand the performance of point-of-care testing.

      Lee Schroeder completed his BA in physics at the University of Pennsylvania in 1995, his MD/PhD with an NIH Medical Scientist Training Program scholarship award at the University of California, San Diego in 2011, and his pathology residency at Stanford University in 2014. Lee currently serves as Co-Director of a Blue Cross Blue Shield funded pilot program to improve utilization of genetic testing for hereditary cancer syndromes and is member of the chemistry resource committee at the College of American Pathologists.

    Abstract:

    The aim of the learning healthcare system is to leverage data stored in the electronic health record (EHR) to gain insights into and improve healthcare delivery. Laboratory testing represents the largest source of routinely formatted data in the EHR and is therefore readily accessible for mining. In this presentation I will be describing a variety of practical applications of EHR data analytics relevant to clinical laboratory practice and patient care. These will include automatic generation of institutional reference ranges, synthesis of test panel results with machine learning models, fine-grained characterization of point-of-care instrument accuracy, effects of in-vitro hemolysis on central laboratory testing results, and the use of in-silico simulations to put the findings in context. Taking care to account for implicit confounders present in retrospective studies, a wealth of data exists in the EHR that can be unlocked to aid in day-to-day management of the clinical laboratory as well as guide healthcare policy-making in general.


    Show Resources
    Loading Comments...