Effect of a Machine Learning-based Severe Sepsis Prediction Algorithm on Patient Survival and Hospital Length of Stay: a Randomised Clinical Trial

Overview

Journal BMJ Open Respir Res

Date 2018 Feb 14

PMID 29435343

Citations 141

Authors

David W Shimabukuro

Christopher W Barton

Mitchell D Feldman

Samson J Mataraso

Ritankar Das

Affiliations

Soon will be listed here.

Abstract

Introduction: Several methods have been developed to electronically monitor patients for severe sepsis, but few provide predictive capabilities to enable early intervention; furthermore, no severe sepsis prediction systems have been previously validated in a randomised study. We tested the use of a machine learning-based severe sepsis prediction system for reductions in average length of stay and in-hospital mortality rate.

Methods: We conducted a randomised controlled clinical trial at two medical-surgical intensive care units at the University of California, San Francisco Medical Center, evaluating the primary outcome of average length of stay, and secondary outcome of in-hospital mortality rate from December 2016 to February 2017. Adult patients (18+) admitted to participating units were eligible for this factorial, open-label study. Enrolled patients were assigned to a trial arm by a random allocation sequence. In the control group, only the current severe sepsis detector was used; in the experimental group, the machine learning algorithm (MLA) was also used. On receiving an alert, the care team evaluated the patient and initiated the severe sepsis bundle, if appropriate. Although participants were randomly assigned to a trial arm, group assignments were automatically revealed for any patients who received MLA alerts.

Results: Outcomes from 75 patients in the control and 67 patients in the experimental group were analysed. Average length of stay decreased from 13.0 days in the control to 10.3 days in the experimental group (p=0.042). In-hospital mortality decreased by 12.4 percentage points when using the MLA (p=0.018), a relative reduction of 58.0%. No adverse events were reported during this trial.

Conclusion: The MLA was associated with improved patient outcomes. This is the first randomised controlled trial of a sepsis surveillance system to demonstrate statistically significant differences in length of stay and in-hospital mortality.

Trial Registration: NCT03015454.

Citing Articles

The Japanese Clinical Practice Guidelines for Management of Sepsis and Septic Shock 2024.

Shime N, Nakada T, Yatabe T, Yamakawa K, Aoki Y, Inoue S Acute Med Surg. 2025; 12(1):e70037.

PMID: 39996161 PMC: 11848044. DOI: 10.1002/ams2.70037.

Development and Validation of a Machine Learning Model for Early Prediction of Sepsis Onset in Hospital Inpatients from All Departments.

Thiboud P, Francois Q, Faure C, Chaufferin G, Arribe B, Ettahar N Diagnostics (Basel). 2025; 15(3).

PMID: 39941233 PMC: 11817331. DOI: 10.3390/diagnostics15030302.

User-Oriented Requirements for Artificial Intelligence-Based Clinical Decision Support Systems in Sepsis: Protocol for a Multimethod Research Project.

Raszke P, Giebel G, Abels C, Wasem J, Adamzik M, Nowak H JMIR Res Protoc. 2025; 14:e62704.

PMID: 39883929 PMC: 11826947. DOI: 10.2196/62704.

Machine Learning Models for the Early Real-Time Prediction of Deterioration in Intensive Care Units-A Novel Approach to the Early Identification of High-Risk Patients.

Thiele D, Rodseth R, Friedland R, Berger F, Mathew C, Maslo C J Clin Med. 2025; 14(2).

PMID: 39860355 PMC: 11766095. DOI: 10.3390/jcm14020350.

Understanding ICU Nursing Knowledge, Perceived Barriers, and Facilitators of Sepsis Recognition and Management: A Cross-Sectional Study.

Kissel K, Krewulak K, Poulin T, Parhar K, Niven D, Doiron V Crit Care Explor. 2025; 7(1):e1200.

PMID: 39804021 PMC: 11732647. DOI: 10.1097/CCE.0000000000001200.

References

Balk R . Severe sepsis and septic shock. Definitions, epidemiology, and clinical manifestations. Crit Care Clin. 2000; 16(2):179-92. DOI: 10.1016/s0749-0704(05)70106-8. View

Nguyen H, Corbett S, Steele R, Banta J, Clark R, Hayes S . Implementation of a bundle of quality indicators for the early management of severe sepsis and septic shock is associated with decreased mortality. Crit Care Med. 2007; 35(4):1105-12. DOI: 10.1097/01.CCM.0000259463.33848.3D. View

Calvert J, Mao Q, Hoffman J, Jay M, Desautels T, Mohamadlou H . Using electronic health record collected clinical variables to predict medical intensive care unit mortality. Ann Med Surg (Lond). 2016; 11:52-57. PMC: 5037117. DOI: 10.1016/j.amsu.2016.09.002. View

Calvert J, Mao Q, Rogers A, Barton C, Jay M, Desautels T . A computational approach to mortality prediction of alcohol use disorder inpatients. Comput Biol Med. 2016; 75:74-9. DOI: 10.1016/j.compbiomed.2016.05.015. View

Desautels T, Calvert J, Hoffman J, Jay M, Kerem Y, Shieh L . Prediction of Sepsis in the Intensive Care Unit With Minimal Electronic Health Record Data: A Machine Learning Approach. JMIR Med Inform. 2016; 4(3):e28. PMC: 5065680. DOI: 10.2196/medinform.5909. View

McLymont N, Glover G . Scoring systems for the characterization of sepsis and associated outcomes. Ann Transl Med. 2017; 4(24):527. PMC: 5233540. DOI: 10.21037/atm.2016.12.53. View

Henry K, Hager D, Pronovost P, Saria S . A targeted real-time early warning score (TREWScore) for septic shock. Sci Transl Med. 2015; 7(299):299ra122. DOI: 10.1126/scitranslmed.aab3719. View

Calvert J, Desautels T, Chettipally U, Barton C, Hoffman J, Jay M . High-performance detection and early prediction of septic shock for alcohol-use disorder patients. Ann Med Surg (Lond). 2016; 8:50-5. PMC: 4960347. DOI: 10.1016/j.amsu.2016.04.023. View

Manaktala S, Claypool S . Evaluating the impact of a computerized surveillance algorithm and decision support system on sepsis mortality. J Am Med Inform Assoc. 2016; 24(1):88-95. PMC: 7654083. DOI: 10.1093/jamia/ocw056. View

10.

Chalupka A, Talmor D . The economics of sepsis. Crit Care Clin. 2011; 28(1):57-76, vi. DOI: 10.1016/j.ccc.2011.09.003. View

11.

Stanculescu I, Williams C, Freer Y . Autoregressive hidden Markov models for the early detection of neonatal sepsis. IEEE J Biomed Health Inform. 2014; 18(5):1560-70. DOI: 10.1109/JBHI.2013.2294692. View

12.

Dellinger R, Levy M, Rhodes A, Annane D, Gerlach H, Opal S . Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Intensive Care Med. 2013; 39(2):165-228. PMC: 7095153. DOI: 10.1007/s00134-012-2769-8. View

13.

Stevenson E, Rubenstein A, Radin G, Wiener R, Walkey A . Two decades of mortality trends among patients with severe sepsis: a comparative meta-analysis*. Crit Care Med. 2013; 42(3):625-31. PMC: 4313930. DOI: 10.1097/CCM.0000000000000026. View

14.

Singer M . Antibiotics for Sepsis: Does Each Hour Really Count, or Is It Incestuous Amplification?. Am J Respir Crit Care Med. 2017; 196(7):800-802. DOI: 10.1164/rccm.201703-0621ED. View

15.

Yealy D, Kellum J, Huang D, Barnato A, Weissfeld L, Pike F . A randomized trial of protocol-based care for early septic shock. N Engl J Med. 2014; 370(18):1683-93. PMC: 4101700. DOI: 10.1056/NEJMoa1401602. View

16.

Lever A, Mackenzie I . Sepsis: definition, epidemiology, and diagnosis. BMJ. 2007; 335(7625):879-83. PMC: 2043413. DOI: 10.1136/bmj.39346.495880.AE. View

17.

Churpek M, Snyder A, Han X, Sokol S, Pettit N, Howell M . Quick Sepsis-related Organ Failure Assessment, Systemic Inflammatory Response Syndrome, and Early Warning Scores for Detecting Clinical Deterioration in Infected Patients outside the Intensive Care Unit. Am J Respir Crit Care Med. 2016; 195(7):906-911. PMC: 5387705. DOI: 10.1164/rccm.201604-0854OC. View

18.

Angus D, Lidicker J, Clermont G, Carcillo J, Pinsky M . Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med. 2001; 29(7):1303-10. DOI: 10.1097/00003246-200107000-00002. View

19.

Berger T, Birnbaum A, Bijur P, Kuperman G, Gennis P . A Computerized Alert Screening for Severe Sepsis in Emergency Department Patients Increases Lactate Testing but does not Improve Inpatient Mortality. Appl Clin Inform. 2013; 1(4):394-407. PMC: 3633314. DOI: 10.4338/ACI-2010-09-RA-0054. View

20.

Calvert J, Price D, Chettipally U, Barton C, Feldman M, Hoffman J . A computational approach to early sepsis detection. Comput Biol Med. 2016; 74:69-73. DOI: 10.1016/j.compbiomed.2016.05.003. View