Clinician vs mathematical statistical models: which is better at predicting an abnormal chest radiograph finding in injured patients?

Elizabeth Dillard, Fred A. Luchette, Benjamin W. Sears, John Norton, Carol R. Schermer, R. Lawrence Reed, Richard L. Gamelli, Thomas J. Esposito

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Objective: The purpose of this study was to determine if statistical models for prediction of chest injuries would outperform the clinician's (MD) ability to identify injured patients at risk for a thoracic injury diagnosed by chest radiograph (CXR). Design: A prospective observational study was done during a 12-month period. Setting: The study was conducted in a level I trauma center. Patients: Injured patients meeting trauma team activation criteria were enrolled to the study. Interventions: Physical examination findings by a clinician were interpreted and CXR was performed. Outcome measures: The accuracy of 2 mathematical models is compared against the accuracy of clinician's clinical judgment in predicting an injury by CXR. Two newly constructed multivariate models, binary logistic regression (LR) and classification and regression tree (CaRT) analysis, are compared to previously published data of clinician clinical assessment of probability of thoracic injury identified by CXR. Results: Data for 757 patients were analyzed. Classification and regression tree analysis developed a stepwise decision tree to determine which signs/symptoms were indicative of an abnormal CXR finding. The sensitivity (CaRT, 36.6%; LR, 36.3%; MD, 58.7%), specificity (CaRT, 98.3%; LR, 98.2%; MD, 96.4%), and error rates (CaRT, 0.93; LR, 0.94; MD, 0.82) show that the mathematical decision aids are less sensitive and risk more misclassification compared to clinician judgment in predicting an injury by CXR. Conclusion: Clinician judgment was superior to mathematical decision aids for predicting an abnormal CXR finding in injured patients with chest trauma.

Original languageEnglish (US)
Pages (from-to)823-830
Number of pages8
JournalAmerican Journal of Emergency Medicine
Volume25
Issue number7
DOIs
StatePublished - Sep 2007
Externally publishedYes

Fingerprint

Statistical Models
Theoretical Models
Thorax
Thoracic Injuries
Logistic Models
Decision Support Techniques
Wounds and Injuries
Regression Analysis
Decision Trees
Aptitude
Trauma Centers
Signs and Symptoms
Physical Examination
Observational Studies
Outcome Assessment (Health Care)
Prospective Studies

ASJC Scopus subject areas

  • Emergency Medicine

Cite this

Clinician vs mathematical statistical models : which is better at predicting an abnormal chest radiograph finding in injured patients? / Dillard, Elizabeth; Luchette, Fred A.; Sears, Benjamin W.; Norton, John; Schermer, Carol R.; Reed, R. Lawrence; Gamelli, Richard L.; Esposito, Thomas J.

In: American Journal of Emergency Medicine, Vol. 25, No. 7, 09.2007, p. 823-830.

Research output: Contribution to journalArticle

Dillard, Elizabeth ; Luchette, Fred A. ; Sears, Benjamin W. ; Norton, John ; Schermer, Carol R. ; Reed, R. Lawrence ; Gamelli, Richard L. ; Esposito, Thomas J. / Clinician vs mathematical statistical models : which is better at predicting an abnormal chest radiograph finding in injured patients?. In: American Journal of Emergency Medicine. 2007 ; Vol. 25, No. 7. pp. 823-830.
@article{12ca02f5a1a94c8aa4720a90d33c3160,
title = "Clinician vs mathematical statistical models: which is better at predicting an abnormal chest radiograph finding in injured patients?",
abstract = "Objective: The purpose of this study was to determine if statistical models for prediction of chest injuries would outperform the clinician's (MD) ability to identify injured patients at risk for a thoracic injury diagnosed by chest radiograph (CXR). Design: A prospective observational study was done during a 12-month period. Setting: The study was conducted in a level I trauma center. Patients: Injured patients meeting trauma team activation criteria were enrolled to the study. Interventions: Physical examination findings by a clinician were interpreted and CXR was performed. Outcome measures: The accuracy of 2 mathematical models is compared against the accuracy of clinician's clinical judgment in predicting an injury by CXR. Two newly constructed multivariate models, binary logistic regression (LR) and classification and regression tree (CaRT) analysis, are compared to previously published data of clinician clinical assessment of probability of thoracic injury identified by CXR. Results: Data for 757 patients were analyzed. Classification and regression tree analysis developed a stepwise decision tree to determine which signs/symptoms were indicative of an abnormal CXR finding. The sensitivity (CaRT, 36.6{\%}; LR, 36.3{\%}; MD, 58.7{\%}), specificity (CaRT, 98.3{\%}; LR, 98.2{\%}; MD, 96.4{\%}), and error rates (CaRT, 0.93; LR, 0.94; MD, 0.82) show that the mathematical decision aids are less sensitive and risk more misclassification compared to clinician judgment in predicting an injury by CXR. Conclusion: Clinician judgment was superior to mathematical decision aids for predicting an abnormal CXR finding in injured patients with chest trauma.",
author = "Elizabeth Dillard and Luchette, {Fred A.} and Sears, {Benjamin W.} and John Norton and Schermer, {Carol R.} and Reed, {R. Lawrence} and Gamelli, {Richard L.} and Esposito, {Thomas J.}",
year = "2007",
month = "9",
doi = "10.1016/j.ajem.2006.12.009",
language = "English (US)",
volume = "25",
pages = "823--830",
journal = "American Journal of Emergency Medicine",
issn = "0735-6757",
publisher = "W.B. Saunders Ltd",
number = "7",

}

TY - JOUR

T1 - Clinician vs mathematical statistical models

T2 - which is better at predicting an abnormal chest radiograph finding in injured patients?

AU - Dillard, Elizabeth

AU - Luchette, Fred A.

AU - Sears, Benjamin W.

AU - Norton, John

AU - Schermer, Carol R.

AU - Reed, R. Lawrence

AU - Gamelli, Richard L.

AU - Esposito, Thomas J.

PY - 2007/9

Y1 - 2007/9

N2 - Objective: The purpose of this study was to determine if statistical models for prediction of chest injuries would outperform the clinician's (MD) ability to identify injured patients at risk for a thoracic injury diagnosed by chest radiograph (CXR). Design: A prospective observational study was done during a 12-month period. Setting: The study was conducted in a level I trauma center. Patients: Injured patients meeting trauma team activation criteria were enrolled to the study. Interventions: Physical examination findings by a clinician were interpreted and CXR was performed. Outcome measures: The accuracy of 2 mathematical models is compared against the accuracy of clinician's clinical judgment in predicting an injury by CXR. Two newly constructed multivariate models, binary logistic regression (LR) and classification and regression tree (CaRT) analysis, are compared to previously published data of clinician clinical assessment of probability of thoracic injury identified by CXR. Results: Data for 757 patients were analyzed. Classification and regression tree analysis developed a stepwise decision tree to determine which signs/symptoms were indicative of an abnormal CXR finding. The sensitivity (CaRT, 36.6%; LR, 36.3%; MD, 58.7%), specificity (CaRT, 98.3%; LR, 98.2%; MD, 96.4%), and error rates (CaRT, 0.93; LR, 0.94; MD, 0.82) show that the mathematical decision aids are less sensitive and risk more misclassification compared to clinician judgment in predicting an injury by CXR. Conclusion: Clinician judgment was superior to mathematical decision aids for predicting an abnormal CXR finding in injured patients with chest trauma.

AB - Objective: The purpose of this study was to determine if statistical models for prediction of chest injuries would outperform the clinician's (MD) ability to identify injured patients at risk for a thoracic injury diagnosed by chest radiograph (CXR). Design: A prospective observational study was done during a 12-month period. Setting: The study was conducted in a level I trauma center. Patients: Injured patients meeting trauma team activation criteria were enrolled to the study. Interventions: Physical examination findings by a clinician were interpreted and CXR was performed. Outcome measures: The accuracy of 2 mathematical models is compared against the accuracy of clinician's clinical judgment in predicting an injury by CXR. Two newly constructed multivariate models, binary logistic regression (LR) and classification and regression tree (CaRT) analysis, are compared to previously published data of clinician clinical assessment of probability of thoracic injury identified by CXR. Results: Data for 757 patients were analyzed. Classification and regression tree analysis developed a stepwise decision tree to determine which signs/symptoms were indicative of an abnormal CXR finding. The sensitivity (CaRT, 36.6%; LR, 36.3%; MD, 58.7%), specificity (CaRT, 98.3%; LR, 98.2%; MD, 96.4%), and error rates (CaRT, 0.93; LR, 0.94; MD, 0.82) show that the mathematical decision aids are less sensitive and risk more misclassification compared to clinician judgment in predicting an injury by CXR. Conclusion: Clinician judgment was superior to mathematical decision aids for predicting an abnormal CXR finding in injured patients with chest trauma.

UR - http://www.scopus.com/inward/record.url?scp=34548523119&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34548523119&partnerID=8YFLogxK

U2 - 10.1016/j.ajem.2006.12.009

DO - 10.1016/j.ajem.2006.12.009

M3 - Article

C2 - 17870489

AN - SCOPUS:34548523119

VL - 25

SP - 823

EP - 830

JO - American Journal of Emergency Medicine

JF - American Journal of Emergency Medicine

SN - 0735-6757

IS - 7

ER -