View source for Automated electronic medical record sepsis detection in the emergency department

== Abstract ==

“Background. While often first treated in the emergency department (ED), identification of sepsis is difficult. [[EMR|Electronic medical record]] (EMR) clinical decision tools offer a novel strategy for identifying patients with sepsis. The objective of this study was to test the accuracy of an EMR-based, automated sepsis identification system. Methods. We tested an EMR-based sepsis identification tool at a major academic, urban ED with 64,000 annual visits. The EMR system collected vital sign and laboratory test information on all ED patients, triggering a “sepsis alert” for those with ≥2 SIRS (systemic inflammatory response syndrome) criteria (fever, tachycardia, tachypnea, leukocytosis) plus ≥1 major organ dysfunction (SBP ≤ 90 mm Hg, lactic acid ≥2.0 mg/dL). We confirmed the presence of sepsis through manual review of physician, nursing, and laboratory records. We also reviewed a random selection of ED cases that did not trigger a sepsis alert. We evaluated the diagnostic accuracy of the sepsis identification tool. Results. From January 1 through March 31, 2012, there were 795 automated sepsis alerts. We randomly selected 300 cases without a sepsis alert from the same period. The true prevalence of sepsis was 355/795 (44.7%) among alerts and 0/300 (0%) among non-alerts. The positive predictive value of the sepsis alert was 44.7% (95% CI [41.2–48.2%]). Pneumonia and respiratory infections (38%) and urinary tract infection (32.7%) were the most common infections among the 355 patients with true sepsis (true positives). Among false-positive sepsis alerts, the most common medical conditions were gastrointestinal (26.1%), traumatic (25.7%), and cardiovascular (20.0%) conditions. Rates of hospital admission were: true-positive sepsis alert 91.0%, false-positive alert 83.0%, no sepsis alert 5.7%. Conclusions. This ED EMR-based automated sepsis identification system was able to detect cases with sepsis. Automated EMR-based detection may provide a viable strategy for identifying sepsis in the ED.” <ref name="Nguyen">Nguyen, S. Q., Mwakalindile, E., Booth, J. S., Hogan, V., Morgan, J., Prickett, C. T., … Wang, H. E. (2014). Automated electronic medical record sepsis detection in the emergency department. PeerJ, 2, e343. http://doi.org/10.7717/peerj.343</ref>

== Purpose ==

The purpose of this retrospective study was to test a [[CDS|clinical decision support (CDS)]] tool aimed at detecting the presence of sepsis in patients in the emergency department.

== Methods ==

The researchers developed the sepsis CDS tool using the Cerner EMR used at the University of Alabama at Birmingham hospital.  The tool returned a “’sepsis alert’ if the EMR identified two or more Systemic Inflammatory Response Syndrome (SIRS) criteria and at least one sign of shock” in a patient.  To determine the accuracy of the tool, the researchers then combed through the positive alerts and manually confirmed the presence (or absence) of sepsis in the patients.  They also randomly checked 300 records of patients that were not flagged by the tool to ensure they did not (or did) have sepsis.

== Results ==

The sepsis CDS tool gave a positive alert in 795 patients, of which 355 were manually confirmed by the researchers.  This gave a PPV of 44.7%.  Review of the 300 randomly checked records showed that none had sepsis which gave an estimated NPV of 100%.

== Discussion ==

The researchers were encouraged by the findings of this study.  They also felt the number of false positives (440) of the tool was not too high due to the difficulty in diagnosing sepsis.  In addition, they noted there are many other diseases that can exhibit SIRS criteria and possibly trigger a false positive alert.  Another important finding is that a large majority of patients that triggered an alert were eventually admitted to the hospital.  Thus, they postulated that the tool could also be used as an indicator for how sick a patient is and the need for hospitalization.

== Thoughts ==

This study provides another example of how a CDS tool can aid a busy ED physician in determining if a patient has sepsis or has the need for hospitalization.  More research needs to be done with regards to this tool though.  A prospective study would be beneficial in addition to reviewing more non-alert patient records.  The sample sizes are rather small.

== Reference ==

<references/>


[[Category:Reviews]]
[[Category:CDS]]
[[Category:HI5313-2015-FALL]]
[[Category:EHR implementation project]]