Skip to Main Content

Navigating the Minefield of Using Patient Data to Develop Analytics and Decision Support

This Briefing is brought to you by AHLA’s Academic Medical Centers and Teaching Hospitals Practice Group.
  • July 30, 2020
  • Ellen Wright Clayton , Vanderbilt University Medical Center and Vanderbilt University School of Law
  • Kyle J. McKibbin

Many people believe that analyzing large amounts of data about individuals to understand contributions to health and disease and to develop algorithms to individualize clinical interventions promises great social value. Analytic approaches embrace those commonly described as artificial intelligence (AI), the overarching term which includes efforts to imitate intelligent human thought, and machine learning, which develops analytic frameworks from direct examination of data. These analyses provide the foundation for strategies to identify what interventions a particular patient should or should not have. One tool for delivering such information is clinical decision support (CDS), which warns the clinician of a potential problem posed by a proposed course of action. For example, a physician who starts to order a new drug for a patient might receive a warning that that drug will not work in that patient or that it will interact badly with medications the patient is already taking.

Keep reading this article by creating an account or logging in.