Informing and Personalizing the Response to Inpatient Physiological Deterioration Using Early Warning Scores

The Department of Decision Sciences invites you to attend Informing and Personalizing the Response to Inpatient Physiological Deterioration Using Early Warning Scores Speaker: Muge Capan, Ph.D., Research Investigator at Value Institute Thursday, Oct 23rd, 2014, 11am – 12noon. Gerri C. LeBow Hall, Room 722 For more information, please contact Wenjing Shen at wenjings@drexel.edu.

Abstract Hospitalized patients are at risk of unexpected clinical deterioration, usually characterized by a disturbance in physiology. Early detection with appropriate response can reduce the risk of undesired outcomes. Early Warning Scores are widely used in hospitals to structure the early recognition and provide guidance to identify the patients who may need additional attention. However, there are challenges associated with the current use of Early Warning Scores. There is a lack of standardization in clinical guidelines for selecting the physiological measures to be included in a score and their cut-off points. Most published scores include expert opinion based cut-off points to assess risk for a specific event. As a result, the performance of any score depends on the outcome of interest. In addition, the currently used cut-off points are not patient-specific. Our research involves understanding the heterogeneity of the patient population in a large health care system, and developing decision models for patient-centered rapid response. We focus on two types of uncertainty: (i) uncertainty in a patient’s condition, and (ii) provider perception-based uncertainty in model parameters. This talk will discuss the results of electronic medical records-derived Markovian models for identifying optimal rapid response team activation policies.

Biography Muge Capan, PhD, is a Research Investigator at the Value Institute at Christiana Care Health System specializing in health systems engineering and patient outcomes research. She earned her PhD in Industrial and Systems Engineering from North Carolina State University. Her areas of concentration include stochastic processes, decision analysis, and mathematical modeling with applications to medical decision making and optimization of hospital operations using large-scale electronic medical records.