TY - JOUR
T1 - Customized reference ranges for laboratory values decrease false positive alerts in intensive care unit patients
AU - Kilickaya, Oguz
AU - Schmickl, Christopher
AU - Ahmed, Adil
AU - Pulido, Juan
AU - Onigkeit, James
AU - Kashani, Kianoush
AU - Gajic, Ognjen
AU - Herasevich, Vitaly
AU - Pickering, Brian
N1 - Funding Information:
This research has been reviewed by the Mayo Clinic Conflict of Interest Review Board and is being conducted in compliance with Mayo Clinic Conflict of Interest policies. This publication was supported by Grant Number UL1 TR000135 from the National Center for Advancing Translational Sciences (NCATS). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. Mayo Clinic and authors (OG, VH, BWP) hold the patent application on AWARE technology referenced in this paper (US 2010/0198622, 12/697861, PCT/US2010/022750). AWARE is licensed to Ambient Clinical Analytics Inc. Dr Pickering additionally serves as member on the Board of Directors of Ambient Clinical Analytics Inc. Dr. Kilickaya reports grant support (UL1 TR000135) from the National Center for Advancing Translational Sciences (NCATS, http://www.ncats.nih.gov/ ) during the conduct of the study. This does not alter the authors’ adherence to PLOS ONE Editorial policies and criteria. The other authors have nothing to disclose.
Publisher Copyright:
© 2014 Kilickaya et al.
PY - 2014/9/18
Y1 - 2014/9/18
N2 - Background: Traditional electronic medical record (EMR) interfaces mark laboratory tests as abnormal based on standard reference ranges derived from healthy, middle-aged adults. This yields many false positive alerts with subsequent alertfatigue when applied to complex populations like hospitalized, critically ill patients. Novel EMR interfaces using adjusted reference ranges customized for specific patient populations may ameliorate this problem.Objective: To compare accuracy of abnormal laboratory value indicators in a novel vs traditional EMR interface. Methods: Laboratory data from intensive care unit (ICU) patients consecutively admitted during a two-day period were recorded. For each patient, available laboratory results and the problem list were sent to two mutually blinded critical care experts, who marked the values about which they would like to be alerted. All disagreements were resolved by an independent super-reviewer. Based on this gold standard, we calculated and compared the sensitivity, specificity, positive and negative predictive values (PPV, NPV) of customized vs traditional abnormal value indicators.Results: Thirty seven patients with a total of 1341 laboratory results were included. Experts' agreement was fair (kappa = 0.39). Compared to the traditional EMR, custom abnormal laboratory value indicators had similar sensitivity (77% vs 85%, P = 0.22) and NPV (97.1% vs 98.6%, P = 0.06) but higher specificity (79% vs 61%, P<0.001) and PPV (28% vs 11%, P< 0.001).Conclusions: Reference ranges for laboratory values customized for an ICU population decrease false positive alerts. Disagreement among clinicians about which laboratory values should be indicated as abnormal limits the development of customized reference ranges.
AB - Background: Traditional electronic medical record (EMR) interfaces mark laboratory tests as abnormal based on standard reference ranges derived from healthy, middle-aged adults. This yields many false positive alerts with subsequent alertfatigue when applied to complex populations like hospitalized, critically ill patients. Novel EMR interfaces using adjusted reference ranges customized for specific patient populations may ameliorate this problem.Objective: To compare accuracy of abnormal laboratory value indicators in a novel vs traditional EMR interface. Methods: Laboratory data from intensive care unit (ICU) patients consecutively admitted during a two-day period were recorded. For each patient, available laboratory results and the problem list were sent to two mutually blinded critical care experts, who marked the values about which they would like to be alerted. All disagreements were resolved by an independent super-reviewer. Based on this gold standard, we calculated and compared the sensitivity, specificity, positive and negative predictive values (PPV, NPV) of customized vs traditional abnormal value indicators.Results: Thirty seven patients with a total of 1341 laboratory results were included. Experts' agreement was fair (kappa = 0.39). Compared to the traditional EMR, custom abnormal laboratory value indicators had similar sensitivity (77% vs 85%, P = 0.22) and NPV (97.1% vs 98.6%, P = 0.06) but higher specificity (79% vs 61%, P<0.001) and PPV (28% vs 11%, P< 0.001).Conclusions: Reference ranges for laboratory values customized for an ICU population decrease false positive alerts. Disagreement among clinicians about which laboratory values should be indicated as abnormal limits the development of customized reference ranges.
UR - http://www.scopus.com/inward/record.url?scp=84907192504&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84907192504&partnerID=8YFLogxK
U2 - 10.1371/journal.pone.0107930
DO - 10.1371/journal.pone.0107930
M3 - Article
C2 - 25233485
AN - SCOPUS:84907192504
SN - 1932-6203
VL - 9
JO - PloS one
JF - PloS one
IS - 9
M1 - e107930
ER -