Automated Deterioration Detection Using Electronic Medical Record Data in Intensive Care Unit Patients: A Systematic Review

Timely detection of deterioration in status for intensive care unit patients can be problematic due to variation in data availability and the necessity of integrating data from multiple sources. This can lead to opaqueness of clinical trends and failure to rescue. Automated deterioration detection using electronic medical record data can reduce the risk of failure to rescue. This review describes the automated use of electronic medical record data in identifying deterioration in intensive care unit patients. PubMed and Google Scholar were used to retrieve publications between January 1, 2006, and March 31, 2016. Six studies met inclusion criteria: intensive care unit patient focus, description of electronic medical record data use in automated patient deterioration detection, and presence of predictive, sensitivity, and/or specificity values. Detection focused on specific clinical events such as infection; data sources were electronic medical record–populated databases. Detection algorithms incorporated laboratory results, vital signs, medication orders, and respiratory therapy and radiology documentation. Positive and negative predictive values and sensitivity and specificity measures varied across studies. Three systems generated clinician alerts. Automated deterioration detection using electronic medical record data may be an important aid in caring for intensive care unit patients, but its usefulness is limited by variable electronic medical record detection approaches and performance. The author has disclosed that she has no significant relationships with, or financial interest in, any commercial companies pertaining to this article. Corresponding author: Laurel A. Despins, PhD, RN, Sinclair School of Nursing, University of Missouri, 1 Hospital Drive, Columbia, MO 65211 (despinsl@missouri.edu). Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.

https://ift.tt/2pKrqNq