Big Data and ICU Scoring Systems

James Todd, Adrian Gepp, Bruce J Vanstone, Brent Richards

Research output: Contribution to conferenceAbstractResearchpeer-review

Abstract

Background: Severity scoring systems are used in
intensive care units for stratifying patients in clinical
research and benchmarking ICU performance. A variant
of the Acute Physiology and Chronic Health Evaluation
(APACHE) system is used in Australia for benchmarking
purposes – the APACHE III-j. This and other major
scoring systems have been developed under an old
paradigm of minimum data collection, while the current
paradigm is to use all useful data. The APACHE III-j
system uses the worst observations in the first 24-hours of a patient’s ICU stay, ignoring the rest of the
distributional information. We hypothesise that scoring
system performance can be improved by adding
variables that capture this ignored distributional
information.
Methods: To test this hypothesis, the APACHE III-j system
will be replicated and compared to a modified version
that adds metrics describing the distribution of an
underlying physiology variable utilising high frequency
data. Data for the study is provided by the Gold Coast
University Hospital.
Results: Results were assessed by comparing the two
models on the basis on discrimination using ROC curves
and calibration through the Hosmer-Lemeshow statistics.
Discussion: Findings suggest future use of high frequency
data to capture additional distributional information to
improve severity scoring systems for use in ICUs.
Original languageEnglish
Pages47-48
Number of pages2
Publication statusPublished - 2017
EventThe First Gold Coast Health Research Week Conference 2017 - Gold Coast University Hospital, Gold Coast, Australia
Duration: 28 Nov 201730 Nov 2017
https://www.goldcoast.health.qld.gov.au/research/research-week

Conference

ConferenceThe First Gold Coast Health Research Week Conference 2017
CountryAustralia
CityGold Coast
Period28/11/1730/11/17
Internet address

Fingerprint

APACHE
Benchmarking
Gold
Calibration

Cite this

Todd, J., Gepp, A., Vanstone, B. J., & Richards, B. (2017). Big Data and ICU Scoring Systems. 47-48. Abstract from The First Gold Coast Health Research Week Conference 2017, Gold Coast, Australia.
Todd, James ; Gepp, Adrian ; Vanstone, Bruce J ; Richards, Brent. / Big Data and ICU Scoring Systems. Abstract from The First Gold Coast Health Research Week Conference 2017, Gold Coast, Australia.2 p.
@conference{9e8f7592136a48199192e7b5b7e1357e,
title = "Big Data and ICU Scoring Systems",
abstract = "Background: Severity scoring systems are used inintensive care units for stratifying patients in clinicalresearch and benchmarking ICU performance. A variantof the Acute Physiology and Chronic Health Evaluation(APACHE) system is used in Australia for benchmarkingpurposes – the APACHE III-j. This and other majorscoring systems have been developed under an oldparadigm of minimum data collection, while the currentparadigm is to use all useful data. The APACHE III-jsystem uses the worst observations in the first 24-hours of a patient’s ICU stay, ignoring the rest of thedistributional information. We hypothesise that scoringsystem performance can be improved by addingvariables that capture this ignored distributionalinformation.Methods: To test this hypothesis, the APACHE III-j systemwill be replicated and compared to a modified versionthat adds metrics describing the distribution of anunderlying physiology variable utilising high frequencydata. Data for the study is provided by the Gold CoastUniversity Hospital.Results: Results were assessed by comparing the twomodels on the basis on discrimination using ROC curvesand calibration through the Hosmer-Lemeshow statistics.Discussion: Findings suggest future use of high frequencydata to capture additional distributional information toimprove severity scoring systems for use in ICUs.",
author = "James Todd and Adrian Gepp and Vanstone, {Bruce J} and Brent Richards",
year = "2017",
language = "English",
pages = "47--48",
note = "The First Gold Coast Health Research Week Conference 2017 ; Conference date: 28-11-2017 Through 30-11-2017",
url = "https://www.goldcoast.health.qld.gov.au/research/research-week",

}

Todd, J, Gepp, A, Vanstone, BJ & Richards, B 2017, 'Big Data and ICU Scoring Systems' The First Gold Coast Health Research Week Conference 2017, Gold Coast, Australia, 28/11/17 - 30/11/17, pp. 47-48.

Big Data and ICU Scoring Systems. / Todd, James; Gepp, Adrian; Vanstone, Bruce J; Richards, Brent.

2017. 47-48 Abstract from The First Gold Coast Health Research Week Conference 2017, Gold Coast, Australia.

Research output: Contribution to conferenceAbstractResearchpeer-review

TY - CONF

T1 - Big Data and ICU Scoring Systems

AU - Todd, James

AU - Gepp, Adrian

AU - Vanstone, Bruce J

AU - Richards, Brent

PY - 2017

Y1 - 2017

N2 - Background: Severity scoring systems are used inintensive care units for stratifying patients in clinicalresearch and benchmarking ICU performance. A variantof the Acute Physiology and Chronic Health Evaluation(APACHE) system is used in Australia for benchmarkingpurposes – the APACHE III-j. This and other majorscoring systems have been developed under an oldparadigm of minimum data collection, while the currentparadigm is to use all useful data. The APACHE III-jsystem uses the worst observations in the first 24-hours of a patient’s ICU stay, ignoring the rest of thedistributional information. We hypothesise that scoringsystem performance can be improved by addingvariables that capture this ignored distributionalinformation.Methods: To test this hypothesis, the APACHE III-j systemwill be replicated and compared to a modified versionthat adds metrics describing the distribution of anunderlying physiology variable utilising high frequencydata. Data for the study is provided by the Gold CoastUniversity Hospital.Results: Results were assessed by comparing the twomodels on the basis on discrimination using ROC curvesand calibration through the Hosmer-Lemeshow statistics.Discussion: Findings suggest future use of high frequencydata to capture additional distributional information toimprove severity scoring systems for use in ICUs.

AB - Background: Severity scoring systems are used inintensive care units for stratifying patients in clinicalresearch and benchmarking ICU performance. A variantof the Acute Physiology and Chronic Health Evaluation(APACHE) system is used in Australia for benchmarkingpurposes – the APACHE III-j. This and other majorscoring systems have been developed under an oldparadigm of minimum data collection, while the currentparadigm is to use all useful data. The APACHE III-jsystem uses the worst observations in the first 24-hours of a patient’s ICU stay, ignoring the rest of thedistributional information. We hypothesise that scoringsystem performance can be improved by addingvariables that capture this ignored distributionalinformation.Methods: To test this hypothesis, the APACHE III-j systemwill be replicated and compared to a modified versionthat adds metrics describing the distribution of anunderlying physiology variable utilising high frequencydata. Data for the study is provided by the Gold CoastUniversity Hospital.Results: Results were assessed by comparing the twomodels on the basis on discrimination using ROC curvesand calibration through the Hosmer-Lemeshow statistics.Discussion: Findings suggest future use of high frequencydata to capture additional distributional information toimprove severity scoring systems for use in ICUs.

UR - https://www.goldcoast.health.qld.gov.au/research/research-week

M3 - Abstract

SP - 47

EP - 48

ER -

Todd J, Gepp A, Vanstone BJ, Richards B. Big Data and ICU Scoring Systems. 2017. Abstract from The First Gold Coast Health Research Week Conference 2017, Gold Coast, Australia.