One of the reasons that rapid response teams have been less successful than we all once anticipated is due to weakness on the afferent limb of the rapid response system i.e. that we dont identify clinically deteriorating patients soon enough to make a difference. Therefore, multiple attempts have been made to develop scoring systems like MEWS (the modified Early Warning Score) that will help in that earlier identification. Many such systems have been automated, using data readily available from electronic medical records and computerized monitoring devices.
One recent study used real-time automated continuous sampling of electronic medical record data to enable early identification of patients at risk for death (Khurana 2016). An alert would trigger when at least 2 of 4 systemic inflammatory response syndrome (SIRS) criteria plus at least one of 14 acute organ dysfunction parameters was detected. 5.2% of patients for whom the alert triggered died compared to only 0.2% of those without the alert. Those for whom alerts triggered also had more hospital days and ventilator days. In the validation phase, the sensitivity, specificity, and positive and negative likelihood ratios for predicting mortality were quite good.
Its, of course, interesting in that we just recently applauded the proposed removal of the SIRS criteria from the definition of sepsis (see our March 2016 What's New in the Patient Safety World column ). However, the current study would certainly suggest that the SIRS criteria may still be valuable when part of a broader score in predicting mortality in hospitalized patients.
Some of our other columns on MEWS or recognition of clinical deterioration:
Our other columns on rapid response teams:
Our other columns on sepsis:
Khurana HS, Groves RH, Simons MP, et al. Real-Time Automated Sampling of Electronic Medical Records Predicts Hospital Mortality. The American Journal of Medicine 2016; published online 17 May 2016