Automated intensive care unit risk adjustment: Results from a National Veterans Affairs study

Marta L. Render, H. Myra Kim, Deborah E. Welsh, Stephen Timmons, Joseph Johnston, Siu Hui, Alfred F. Connors, Douglas Wagner, Jennifer Daley, Timothy P. Hofer

Research output: Contribution to journalArticle

59 Citations (Scopus)

Abstract

Context: Comparison of outcome among intensive care units (ICUs) requires risk adjustment for differences in severity of illness and risk of death at admission to the ICU, historically obtained by costly chart review and manual data entry. Objective: To accurately estimate patient risk of death in the ICU using data easily available in hospital electronic databases to permit automation. Design and Setting: Cohort study to develop and validate a model to predict mortality at hospital discharge using multivariate logistic regression with a split derivation (17,731) and validation (11,646) sample formed from 29,377 consecutive first ICU admissions to medical, cardiac, and surgical ICUs in 17 Veterans' Health Administration hospitals between February 1996 and July 1997. Main Outcome Measures: Mortality at hospital discharge adjusted for age, laboratory data, diagnosis, source of ICU admission, and comorbid illness. Results: The overall hospital death rate was 11.3%. In the validation sample, the model separated well between survivors and nonsurvivors (area under the receiver operating characteristic curve = 0.885). Examination of the observed vs. the predicted mortality across the range of mortality showed the model was well calibrated. Conclusions: Automation could broaden access to risk adjustment of ICU outcomes with only a small trade-off in discrimination. Broader use might promote valid evaluation of ICU outcomes, encouraging effective practices and improving ICU quality.

Original languageEnglish (US)
Pages (from-to)1638-1646
Number of pages9
JournalCritical Care Medicine
Volume31
Issue number6
DOIs
StatePublished - Jun 1 2003
Externally publishedYes

Fingerprint

Risk Adjustment
Veterans
Intensive Care Units
Automation
Hospital Mortality
Mortality
Veterans Health
United States Department of Veterans Affairs
Clinical Laboratory Techniques
Information Storage and Retrieval
Critical Care
ROC Curve
Survivors
Cohort Studies
Logistic Models
Outcome Assessment (Health Care)
Databases

Keywords

  • Informatics
  • Intensive care unit
  • Mortality
  • Outcome
  • Severity adjustment

ASJC Scopus subject areas

  • Critical Care and Intensive Care Medicine

Cite this

Automated intensive care unit risk adjustment : Results from a National Veterans Affairs study. / Render, Marta L.; Kim, H. Myra; Welsh, Deborah E.; Timmons, Stephen; Johnston, Joseph; Hui, Siu; Connors, Alfred F.; Wagner, Douglas; Daley, Jennifer; Hofer, Timothy P.

In: Critical Care Medicine, Vol. 31, No. 6, 01.06.2003, p. 1638-1646.

Research output: Contribution to journalArticle

Render, ML, Kim, HM, Welsh, DE, Timmons, S, Johnston, J, Hui, S, Connors, AF, Wagner, D, Daley, J & Hofer, TP 2003, 'Automated intensive care unit risk adjustment: Results from a National Veterans Affairs study', Critical Care Medicine, vol. 31, no. 6, pp. 1638-1646. https://doi.org/10.1097/01.CCM.0000055372.08235.09
Render, Marta L. ; Kim, H. Myra ; Welsh, Deborah E. ; Timmons, Stephen ; Johnston, Joseph ; Hui, Siu ; Connors, Alfred F. ; Wagner, Douglas ; Daley, Jennifer ; Hofer, Timothy P. / Automated intensive care unit risk adjustment : Results from a National Veterans Affairs study. In: Critical Care Medicine. 2003 ; Vol. 31, No. 6. pp. 1638-1646.
@article{e9b157e3b5d643f3bf4c65d2c6e8207e,
title = "Automated intensive care unit risk adjustment: Results from a National Veterans Affairs study",
abstract = "Context: Comparison of outcome among intensive care units (ICUs) requires risk adjustment for differences in severity of illness and risk of death at admission to the ICU, historically obtained by costly chart review and manual data entry. Objective: To accurately estimate patient risk of death in the ICU using data easily available in hospital electronic databases to permit automation. Design and Setting: Cohort study to develop and validate a model to predict mortality at hospital discharge using multivariate logistic regression with a split derivation (17,731) and validation (11,646) sample formed from 29,377 consecutive first ICU admissions to medical, cardiac, and surgical ICUs in 17 Veterans' Health Administration hospitals between February 1996 and July 1997. Main Outcome Measures: Mortality at hospital discharge adjusted for age, laboratory data, diagnosis, source of ICU admission, and comorbid illness. Results: The overall hospital death rate was 11.3{\%}. In the validation sample, the model separated well between survivors and nonsurvivors (area under the receiver operating characteristic curve = 0.885). Examination of the observed vs. the predicted mortality across the range of mortality showed the model was well calibrated. Conclusions: Automation could broaden access to risk adjustment of ICU outcomes with only a small trade-off in discrimination. Broader use might promote valid evaluation of ICU outcomes, encouraging effective practices and improving ICU quality.",
keywords = "Informatics, Intensive care unit, Mortality, Outcome, Severity adjustment",
author = "Render, {Marta L.} and Kim, {H. Myra} and Welsh, {Deborah E.} and Stephen Timmons and Joseph Johnston and Siu Hui and Connors, {Alfred F.} and Douglas Wagner and Jennifer Daley and Hofer, {Timothy P.}",
year = "2003",
month = "6",
day = "1",
doi = "10.1097/01.CCM.0000055372.08235.09",
language = "English (US)",
volume = "31",
pages = "1638--1646",
journal = "Critical Care Medicine",
issn = "0090-3493",
publisher = "Lippincott Williams and Wilkins",
number = "6",

}

TY - JOUR

T1 - Automated intensive care unit risk adjustment

T2 - Results from a National Veterans Affairs study

AU - Render, Marta L.

AU - Kim, H. Myra

AU - Welsh, Deborah E.

AU - Timmons, Stephen

AU - Johnston, Joseph

AU - Hui, Siu

AU - Connors, Alfred F.

AU - Wagner, Douglas

AU - Daley, Jennifer

AU - Hofer, Timothy P.

PY - 2003/6/1

Y1 - 2003/6/1

N2 - Context: Comparison of outcome among intensive care units (ICUs) requires risk adjustment for differences in severity of illness and risk of death at admission to the ICU, historically obtained by costly chart review and manual data entry. Objective: To accurately estimate patient risk of death in the ICU using data easily available in hospital electronic databases to permit automation. Design and Setting: Cohort study to develop and validate a model to predict mortality at hospital discharge using multivariate logistic regression with a split derivation (17,731) and validation (11,646) sample formed from 29,377 consecutive first ICU admissions to medical, cardiac, and surgical ICUs in 17 Veterans' Health Administration hospitals between February 1996 and July 1997. Main Outcome Measures: Mortality at hospital discharge adjusted for age, laboratory data, diagnosis, source of ICU admission, and comorbid illness. Results: The overall hospital death rate was 11.3%. In the validation sample, the model separated well between survivors and nonsurvivors (area under the receiver operating characteristic curve = 0.885). Examination of the observed vs. the predicted mortality across the range of mortality showed the model was well calibrated. Conclusions: Automation could broaden access to risk adjustment of ICU outcomes with only a small trade-off in discrimination. Broader use might promote valid evaluation of ICU outcomes, encouraging effective practices and improving ICU quality.

AB - Context: Comparison of outcome among intensive care units (ICUs) requires risk adjustment for differences in severity of illness and risk of death at admission to the ICU, historically obtained by costly chart review and manual data entry. Objective: To accurately estimate patient risk of death in the ICU using data easily available in hospital electronic databases to permit automation. Design and Setting: Cohort study to develop and validate a model to predict mortality at hospital discharge using multivariate logistic regression with a split derivation (17,731) and validation (11,646) sample formed from 29,377 consecutive first ICU admissions to medical, cardiac, and surgical ICUs in 17 Veterans' Health Administration hospitals between February 1996 and July 1997. Main Outcome Measures: Mortality at hospital discharge adjusted for age, laboratory data, diagnosis, source of ICU admission, and comorbid illness. Results: The overall hospital death rate was 11.3%. In the validation sample, the model separated well between survivors and nonsurvivors (area under the receiver operating characteristic curve = 0.885). Examination of the observed vs. the predicted mortality across the range of mortality showed the model was well calibrated. Conclusions: Automation could broaden access to risk adjustment of ICU outcomes with only a small trade-off in discrimination. Broader use might promote valid evaluation of ICU outcomes, encouraging effective practices and improving ICU quality.

KW - Informatics

KW - Intensive care unit

KW - Mortality

KW - Outcome

KW - Severity adjustment

UR - http://www.scopus.com/inward/record.url?scp=0037791848&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037791848&partnerID=8YFLogxK

U2 - 10.1097/01.CCM.0000055372.08235.09

DO - 10.1097/01.CCM.0000055372.08235.09

M3 - Article

C2 - 12794398

AN - SCOPUS:0037791848

VL - 31

SP - 1638

EP - 1646

JO - Critical Care Medicine

JF - Critical Care Medicine

SN - 0090-3493

IS - 6

ER -