Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Nov;47(11):1485-1492.
doi: 10.1097/CCM.0000000000003891.

A Machine Learning Algorithm to Predict Severe Sepsis and Septic Shock: Development, Implementation, and Impact on Clinical Practice

Heather M Giannini  1 , Jennifer C Ginestra  1 , Corey Chivers  2 , Michael Draugelis  2 , Asaf Hanish  2 , William D Schweickert  2   3 , Barry D Fuchs  2   3 , Laurie Meadows  4 , Michael Lynch  4 , Patrick J Donnelly  5 , Kimberly Pavan  6 , Neil O Fishman  2 , C William Hanson 3rd  2 , Craig A Umscheid  2   7   8
Affiliations

A Machine Learning Algorithm to Predict Severe Sepsis and Septic Shock: Development, Implementation, and Impact on Clinical Practice

Heather M Giannini et al. Crit Care Med. 2019 Nov.

Abstract

Objectives: Develop and implement a machine learning algorithm to predict severe sepsis and septic shock and evaluate the impact on clinical practice and patient outcomes.

Design: Retrospective cohort for algorithm derivation and validation, pre-post impact evaluation.

Setting: Tertiary teaching hospital system in Philadelphia, PA.

Patients: All non-ICU admissions; algorithm derivation July 2011 to June 2014 (n = 162,212); algorithm validation October to December 2015 (n = 10,448); silent versus alert comparison January 2016 to February 2017 (silent n = 22,280; alert n = 32,184).

Interventions: A random-forest classifier, derived and validated using electronic health record data, was deployed both silently and later with an alert to notify clinical teams of sepsis prediction.

Measurement and main result: Patients identified for training the algorithm were required to have International Classification of Diseases, 9th Edition codes for severe sepsis or septic shock and a positive blood culture during their hospital encounter with either a lactate greater than 2.2 mmol/L or a systolic blood pressure less than 90 mm Hg. The algorithm demonstrated a sensitivity of 26% and specificity of 98%, with a positive predictive value of 29% and positive likelihood ratio of 13. The alert resulted in a small statistically significant increase in lactate testing and IV fluid administration. There was no significant difference in mortality, discharge disposition, or transfer to ICU, although there was a reduction in time-to-ICU transfer.

Conclusions: Our machine learning algorithm can predict, with low sensitivity but high specificity, the impending occurrence of severe sepsis and septic shock. Algorithm-generated predictive alerts modestly impacted clinical measures. Next steps include describing clinical perception of this tool and optimizing algorithm design and delivery.

PubMed Disclaimer

Conflict of interest statement

The remaining authors have disclosed that they do not have any potential conflicts of interest.

Figures

Figure 1:
Figure 1:
Proportion of Screen Positive Patients Meeting Systemic Inflammatory Response Syndrome (SIRS) Criteria in the Hours Following Algorithm Detection, Compared with Controls SIRS (Systemic Inflammatory Response Syndrome) Criteria include: (1) Temp >38°C (100.4°F) or < 36°C (96.8°F) (2) Heart rate > 90 (3) Respiratory rate > 20 or PaCO2 < 32 mm Hg (4) WBC > 12,000/mm3, < 4,000/mm3, or > 10% bands Criteria for Severe Sepsis: >2 SIRS and positive blood or urine culture and lactate >2.2; Septic Shock: Severe Sepsis AND systolic blood pressure <90 mm Hg
See this image and copyright information in PMC

Comment in

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Rhee C, Dantes R, Epstein L, et al.: Incidence and Trends of Sepsis in US Hospitals Using Clinical vs Claims Data, 2009–2014. JAMA 2017; - PMC - PubMed
    1. Torio CM, Andrews RM: National Inpatient Hospital Costs: The Most Expensive Conditions by Payer, 2011: Statistical Brief #160 [Internet]. In: Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Rockville (MD): Agency for Healthcare Research and Quality (US); 2006. [cited 2017 Oct 3] Available from: http://www.ncbi.nlm.nih.gov/books/NBK169005/ - PubMed
    1. Liu VX, Fielding-Singh V, Greene JD, et al.: The Timing of Early Antibiotics and Hospital Mortality in Sepsis. Am J Respir Crit Care Med 2017; 196:856–863 - PMC - PubMed
    1. Escobar GJ, LaGuardia JC, Turk BJ, et al.: Early detection of impending physiologic deterioration among patients who are not in intensive care: development of predictive models using data from an automated electronic medical record. J Hosp Med 2012; 7:388–395 - PubMed
    1. Bellomo R, Ackerman M, Bailey M, et al.: A controlled trial of electronic automated advisory vital signs monitoring in general hospital wards. Crit Care Med 2012; 40:2349–2361 - PubMed

Publication types