Real-time forecasting of pediatric intensive care unit length of stay using computerized provider orders

Scott R. Levin, Eric T. Harley, James C. Fackler, Christoph U. Lehmann, Jason W. Custer, Daniel France, Scott L. Zeger

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

Objective: To develop a model to produce real-time, updated forecasts of patients' intensive care unit length of stay using naturally generated provider orders. The model was designed to be integrated within a computerized decision support system to improve patient flow management. Design: Retrospective cohort study. Setting: Twenty-six bed pediatric intensive care unit within an urban, academic children's hospital using a computerized order entry system. Patients: A total of 2,178 consecutive pediatric intensive care unit admissions during a 16-month time period. Measurements and Main Results: We obtained unit length of stay measurements, time-stamped provider orders, age, admission source, and readmission status. A joint discrete-time logistic regression model was developed to produce probabilistic length of stay forecasts from continuously updated provider orders. Accuracy was assessed by comparing forecasted expected discharge time with observed discharge time, rank probability scoring, and calibration curves. Cross-validation procedures were conducted. The distribution of length of stay was heavily right-skewed with a mean of 3.5 days (95% confidence interval 0.3-19.1). Provider orders were predictive of length of stay in real-time accurately forecasting discharge within a 12-hr window: 46% for patients within 1 day of discharge, 34% for patients within 2 days of discharge, and 27% for patients within 3 days of discharge. The forecast model incorporating predictive orders demonstrated significant improvements in accuracy compared with forecasts based solely on empirical and temporal information. Seventeen predictive orders were found, grouped by medication, ventilation, laboratory, diet, activity, foreign body, and extracorporeal membrane oxygenation. Conclusions: Provider orders reflect dynamic changes in patients' conditions, making them useful for real-time length of stay prediction and patient flow management. Patients' length of stay represent a major source of variability in intensive care unit resource utilization and if accurately predicted and communicated, may lead to proactive bed management with more efficient patient flow.

Original languageEnglish (US)
Pages (from-to)3058-3064
Number of pages7
JournalCritical care medicine
Volume40
Issue number11
DOIs
StatePublished - Nov 1 2012
Externally publishedYes

Fingerprint

Pediatric Intensive Care Units
Length of Stay
Patient Discharge
Intensive Care Units
Logistic Models
Extracorporeal Membrane Oxygenation
Foreign Bodies
Calibration
Ventilation
Patient Care
Cohort Studies
Retrospective Studies
Joints
Confidence Intervals
Diet

Keywords

  • efficiency organizational
  • forecasting
  • length of stay
  • logistic model
  • medical order entry systems
  • pediatric intensive care units

ASJC Scopus subject areas

  • Critical Care and Intensive Care Medicine

Cite this

Real-time forecasting of pediatric intensive care unit length of stay using computerized provider orders. / Levin, Scott R.; Harley, Eric T.; Fackler, James C.; Lehmann, Christoph U.; Custer, Jason W.; France, Daniel; Zeger, Scott L.

In: Critical care medicine, Vol. 40, No. 11, 01.11.2012, p. 3058-3064.

Research output: Contribution to journalArticle

Levin, Scott R. ; Harley, Eric T. ; Fackler, James C. ; Lehmann, Christoph U. ; Custer, Jason W. ; France, Daniel ; Zeger, Scott L. / Real-time forecasting of pediatric intensive care unit length of stay using computerized provider orders. In: Critical care medicine. 2012 ; Vol. 40, No. 11. pp. 3058-3064.
@article{f95cbe473ecd4f9c95ffcb846cfc7145,
title = "Real-time forecasting of pediatric intensive care unit length of stay using computerized provider orders",
abstract = "Objective: To develop a model to produce real-time, updated forecasts of patients' intensive care unit length of stay using naturally generated provider orders. The model was designed to be integrated within a computerized decision support system to improve patient flow management. Design: Retrospective cohort study. Setting: Twenty-six bed pediatric intensive care unit within an urban, academic children's hospital using a computerized order entry system. Patients: A total of 2,178 consecutive pediatric intensive care unit admissions during a 16-month time period. Measurements and Main Results: We obtained unit length of stay measurements, time-stamped provider orders, age, admission source, and readmission status. A joint discrete-time logistic regression model was developed to produce probabilistic length of stay forecasts from continuously updated provider orders. Accuracy was assessed by comparing forecasted expected discharge time with observed discharge time, rank probability scoring, and calibration curves. Cross-validation procedures were conducted. The distribution of length of stay was heavily right-skewed with a mean of 3.5 days (95{\%} confidence interval 0.3-19.1). Provider orders were predictive of length of stay in real-time accurately forecasting discharge within a 12-hr window: 46{\%} for patients within 1 day of discharge, 34{\%} for patients within 2 days of discharge, and 27{\%} for patients within 3 days of discharge. The forecast model incorporating predictive orders demonstrated significant improvements in accuracy compared with forecasts based solely on empirical and temporal information. Seventeen predictive orders were found, grouped by medication, ventilation, laboratory, diet, activity, foreign body, and extracorporeal membrane oxygenation. Conclusions: Provider orders reflect dynamic changes in patients' conditions, making them useful for real-time length of stay prediction and patient flow management. Patients' length of stay represent a major source of variability in intensive care unit resource utilization and if accurately predicted and communicated, may lead to proactive bed management with more efficient patient flow.",
keywords = "efficiency organizational, forecasting, length of stay, logistic model, medical order entry systems, pediatric intensive care units",
author = "Levin, {Scott R.} and Harley, {Eric T.} and Fackler, {James C.} and Lehmann, {Christoph U.} and Custer, {Jason W.} and Daniel France and Zeger, {Scott L.}",
year = "2012",
month = "11",
day = "1",
doi = "10.1097/CCM.0b013e31825bc399",
language = "English (US)",
volume = "40",
pages = "3058--3064",
journal = "Critical Care Medicine",
issn = "0090-3493",
publisher = "Lippincott Williams and Wilkins",
number = "11",

}

TY - JOUR

T1 - Real-time forecasting of pediatric intensive care unit length of stay using computerized provider orders

AU - Levin, Scott R.

AU - Harley, Eric T.

AU - Fackler, James C.

AU - Lehmann, Christoph U.

AU - Custer, Jason W.

AU - France, Daniel

AU - Zeger, Scott L.

PY - 2012/11/1

Y1 - 2012/11/1

N2 - Objective: To develop a model to produce real-time, updated forecasts of patients' intensive care unit length of stay using naturally generated provider orders. The model was designed to be integrated within a computerized decision support system to improve patient flow management. Design: Retrospective cohort study. Setting: Twenty-six bed pediatric intensive care unit within an urban, academic children's hospital using a computerized order entry system. Patients: A total of 2,178 consecutive pediatric intensive care unit admissions during a 16-month time period. Measurements and Main Results: We obtained unit length of stay measurements, time-stamped provider orders, age, admission source, and readmission status. A joint discrete-time logistic regression model was developed to produce probabilistic length of stay forecasts from continuously updated provider orders. Accuracy was assessed by comparing forecasted expected discharge time with observed discharge time, rank probability scoring, and calibration curves. Cross-validation procedures were conducted. The distribution of length of stay was heavily right-skewed with a mean of 3.5 days (95% confidence interval 0.3-19.1). Provider orders were predictive of length of stay in real-time accurately forecasting discharge within a 12-hr window: 46% for patients within 1 day of discharge, 34% for patients within 2 days of discharge, and 27% for patients within 3 days of discharge. The forecast model incorporating predictive orders demonstrated significant improvements in accuracy compared with forecasts based solely on empirical and temporal information. Seventeen predictive orders were found, grouped by medication, ventilation, laboratory, diet, activity, foreign body, and extracorporeal membrane oxygenation. Conclusions: Provider orders reflect dynamic changes in patients' conditions, making them useful for real-time length of stay prediction and patient flow management. Patients' length of stay represent a major source of variability in intensive care unit resource utilization and if accurately predicted and communicated, may lead to proactive bed management with more efficient patient flow.

AB - Objective: To develop a model to produce real-time, updated forecasts of patients' intensive care unit length of stay using naturally generated provider orders. The model was designed to be integrated within a computerized decision support system to improve patient flow management. Design: Retrospective cohort study. Setting: Twenty-six bed pediatric intensive care unit within an urban, academic children's hospital using a computerized order entry system. Patients: A total of 2,178 consecutive pediatric intensive care unit admissions during a 16-month time period. Measurements and Main Results: We obtained unit length of stay measurements, time-stamped provider orders, age, admission source, and readmission status. A joint discrete-time logistic regression model was developed to produce probabilistic length of stay forecasts from continuously updated provider orders. Accuracy was assessed by comparing forecasted expected discharge time with observed discharge time, rank probability scoring, and calibration curves. Cross-validation procedures were conducted. The distribution of length of stay was heavily right-skewed with a mean of 3.5 days (95% confidence interval 0.3-19.1). Provider orders were predictive of length of stay in real-time accurately forecasting discharge within a 12-hr window: 46% for patients within 1 day of discharge, 34% for patients within 2 days of discharge, and 27% for patients within 3 days of discharge. The forecast model incorporating predictive orders demonstrated significant improvements in accuracy compared with forecasts based solely on empirical and temporal information. Seventeen predictive orders were found, grouped by medication, ventilation, laboratory, diet, activity, foreign body, and extracorporeal membrane oxygenation. Conclusions: Provider orders reflect dynamic changes in patients' conditions, making them useful for real-time length of stay prediction and patient flow management. Patients' length of stay represent a major source of variability in intensive care unit resource utilization and if accurately predicted and communicated, may lead to proactive bed management with more efficient patient flow.

KW - efficiency organizational

KW - forecasting

KW - length of stay

KW - logistic model

KW - medical order entry systems

KW - pediatric intensive care units

UR - http://www.scopus.com/inward/record.url?scp=84868202045&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84868202045&partnerID=8YFLogxK

U2 - 10.1097/CCM.0b013e31825bc399

DO - 10.1097/CCM.0b013e31825bc399

M3 - Article

C2 - 22824935

AN - SCOPUS:84868202045

VL - 40

SP - 3058

EP - 3064

JO - Critical Care Medicine

JF - Critical Care Medicine

SN - 0090-3493

IS - 11

ER -