Prediction of Sepsis and In-Hospital Mortality Using Electronic Health Records

Anahita Khojandi, Varisara Tansakul, Xueping Li, Rebecca S. Koszalinski, William Paiva

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Summary Objectives: Our goal was to develop predictive models for sepsis and in-hospital mortality using electronic health records (EHRs). We showcased the efficiency of these algorithms in patients diagnosed with pneumonia, a group that is highly susceptible to sepsis. Methods: We retrospectively analyzed the Health Facts ® (HF) dataset to develop models to predict mortality and sepsis using the data from the first few hours after admission. In addition, we developed models to predict sepsis using the data collected in the last few hours leading to sepsis onset. We used the random forest classifier to develop the models. Results: The data collected in the EHR system is generally sporadic, making feature extraction and selection difficult, affecting the accuracies of the models. Despite this fact, the developed models can predict sepsis and in-hospital mortality with accuracies of up to 65.26±0.33% and 68.64±0.48%, and sensitivities of up to 67.24±0.36% and 74.00±1.22%, respectively, using only the data from the first 12 hours after admission. The accuracies generally remain consistent for similar models developed using the data from the first 24 and 48 hours after admission. Lastly, the developed models can accurately predict sepsis patients (with up to 98.63±0.17% accuracy and 99.74%±0.13% sensitivity) using the data collected within the last 12 hours before sepsis onset. The results suggest that if such algorithms continuously monitor patients, they can identify sepsis patients in a manner comparable to current screening tools, such as the rulebased Systemic Inflammatory Response Syndrome (SIRS) criteria, while often allowing for early detection of sepsis shortly after admission. Conclusions: The developed models showed promise in early prediction of sepsis, providing an opportunity for directing early intervention efforts to prevent/treat sepsis.

Original languageEnglish
Pages (from-to)185-193
Number of pages9
JournalMethods of Information in Medicine
Volume57
Issue number4
DOIs
StatePublished - 1 Jan 2018

Fingerprint

Electronic Health Records
Hospital Mortality
Sepsis
Systemic Inflammatory Response Syndrome
Pneumonia

Keywords

  • Predictive analytics
  • electronic health records
  • in-hospital mortality
  • sepsis

Cite this

Khojandi, Anahita ; Tansakul, Varisara ; Li, Xueping ; Koszalinski, Rebecca S. ; Paiva, William. / Prediction of Sepsis and In-Hospital Mortality Using Electronic Health Records. In: Methods of Information in Medicine. 2018 ; Vol. 57, No. 4. pp. 185-193.
@article{71ea34dd3b284cedaa59fe9b74f9efd5,
title = "Prediction of Sepsis and In-Hospital Mortality Using Electronic Health Records",
abstract = "Summary Objectives: Our goal was to develop predictive models for sepsis and in-hospital mortality using electronic health records (EHRs). We showcased the efficiency of these algorithms in patients diagnosed with pneumonia, a group that is highly susceptible to sepsis. Methods: We retrospectively analyzed the Health Facts {\circledR} (HF) dataset to develop models to predict mortality and sepsis using the data from the first few hours after admission. In addition, we developed models to predict sepsis using the data collected in the last few hours leading to sepsis onset. We used the random forest classifier to develop the models. Results: The data collected in the EHR system is generally sporadic, making feature extraction and selection difficult, affecting the accuracies of the models. Despite this fact, the developed models can predict sepsis and in-hospital mortality with accuracies of up to 65.26±0.33{\%} and 68.64±0.48{\%}, and sensitivities of up to 67.24±0.36{\%} and 74.00±1.22{\%}, respectively, using only the data from the first 12 hours after admission. The accuracies generally remain consistent for similar models developed using the data from the first 24 and 48 hours after admission. Lastly, the developed models can accurately predict sepsis patients (with up to 98.63±0.17{\%} accuracy and 99.74{\%}±0.13{\%} sensitivity) using the data collected within the last 12 hours before sepsis onset. The results suggest that if such algorithms continuously monitor patients, they can identify sepsis patients in a manner comparable to current screening tools, such as the rulebased Systemic Inflammatory Response Syndrome (SIRS) criteria, while often allowing for early detection of sepsis shortly after admission. Conclusions: The developed models showed promise in early prediction of sepsis, providing an opportunity for directing early intervention efforts to prevent/treat sepsis.",
keywords = "Predictive analytics, electronic health records, in-hospital mortality, sepsis",
author = "Anahita Khojandi and Varisara Tansakul and Xueping Li and Koszalinski, {Rebecca S.} and William Paiva",
year = "2018",
month = "1",
day = "1",
doi = "10.3414/ME18-01-0014",
language = "English",
volume = "57",
pages = "185--193",
journal = "Methods of Information in Medicine",
issn = "0026-1270",
publisher = "Schattauer GmbH",
number = "4",

}

Prediction of Sepsis and In-Hospital Mortality Using Electronic Health Records. / Khojandi, Anahita; Tansakul, Varisara; Li, Xueping; Koszalinski, Rebecca S.; Paiva, William.

In: Methods of Information in Medicine, Vol. 57, No. 4, 01.01.2018, p. 185-193.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Prediction of Sepsis and In-Hospital Mortality Using Electronic Health Records

AU - Khojandi, Anahita

AU - Tansakul, Varisara

AU - Li, Xueping

AU - Koszalinski, Rebecca S.

AU - Paiva, William

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Summary Objectives: Our goal was to develop predictive models for sepsis and in-hospital mortality using electronic health records (EHRs). We showcased the efficiency of these algorithms in patients diagnosed with pneumonia, a group that is highly susceptible to sepsis. Methods: We retrospectively analyzed the Health Facts ® (HF) dataset to develop models to predict mortality and sepsis using the data from the first few hours after admission. In addition, we developed models to predict sepsis using the data collected in the last few hours leading to sepsis onset. We used the random forest classifier to develop the models. Results: The data collected in the EHR system is generally sporadic, making feature extraction and selection difficult, affecting the accuracies of the models. Despite this fact, the developed models can predict sepsis and in-hospital mortality with accuracies of up to 65.26±0.33% and 68.64±0.48%, and sensitivities of up to 67.24±0.36% and 74.00±1.22%, respectively, using only the data from the first 12 hours after admission. The accuracies generally remain consistent for similar models developed using the data from the first 24 and 48 hours after admission. Lastly, the developed models can accurately predict sepsis patients (with up to 98.63±0.17% accuracy and 99.74%±0.13% sensitivity) using the data collected within the last 12 hours before sepsis onset. The results suggest that if such algorithms continuously monitor patients, they can identify sepsis patients in a manner comparable to current screening tools, such as the rulebased Systemic Inflammatory Response Syndrome (SIRS) criteria, while often allowing for early detection of sepsis shortly after admission. Conclusions: The developed models showed promise in early prediction of sepsis, providing an opportunity for directing early intervention efforts to prevent/treat sepsis.

AB - Summary Objectives: Our goal was to develop predictive models for sepsis and in-hospital mortality using electronic health records (EHRs). We showcased the efficiency of these algorithms in patients diagnosed with pneumonia, a group that is highly susceptible to sepsis. Methods: We retrospectively analyzed the Health Facts ® (HF) dataset to develop models to predict mortality and sepsis using the data from the first few hours after admission. In addition, we developed models to predict sepsis using the data collected in the last few hours leading to sepsis onset. We used the random forest classifier to develop the models. Results: The data collected in the EHR system is generally sporadic, making feature extraction and selection difficult, affecting the accuracies of the models. Despite this fact, the developed models can predict sepsis and in-hospital mortality with accuracies of up to 65.26±0.33% and 68.64±0.48%, and sensitivities of up to 67.24±0.36% and 74.00±1.22%, respectively, using only the data from the first 12 hours after admission. The accuracies generally remain consistent for similar models developed using the data from the first 24 and 48 hours after admission. Lastly, the developed models can accurately predict sepsis patients (with up to 98.63±0.17% accuracy and 99.74%±0.13% sensitivity) using the data collected within the last 12 hours before sepsis onset. The results suggest that if such algorithms continuously monitor patients, they can identify sepsis patients in a manner comparable to current screening tools, such as the rulebased Systemic Inflammatory Response Syndrome (SIRS) criteria, while often allowing for early detection of sepsis shortly after admission. Conclusions: The developed models showed promise in early prediction of sepsis, providing an opportunity for directing early intervention efforts to prevent/treat sepsis.

KW - Predictive analytics

KW - electronic health records

KW - in-hospital mortality

KW - sepsis

UR - http://www.scopus.com/inward/record.url?scp=85053881052&partnerID=8YFLogxK

U2 - 10.3414/ME18-01-0014

DO - 10.3414/ME18-01-0014

M3 - Article

C2 - 30248708

AN - SCOPUS:85053881052

VL - 57

SP - 185

EP - 193

JO - Methods of Information in Medicine

JF - Methods of Information in Medicine

SN - 0026-1270

IS - 4

ER -