The effect of picture archiving and communications systems on the accuracy of diagnostic interpretation of pediatric emergency physicians

Serge Gouin, Hema Patel, Sylvie Bergeron, Devendra Amre, Ronald Guérin

Research output: Contribution to journalArticle

Abstract

Objectives: To compare the accuracy of diagnostic interpretation of radiographs by pediatric emergency physicians (EPs) before and after the introduction of a Picture Archiving and Communications System (PACS). Methods: The pre-PACS study period included results from September 2001, when patients were evaluated by using only conventional radiographs. The post-PACS study period consisted of results from September 2002, when patients were evaluated by using only digital radiographic studies. During these periods, consecutive medical records of all patients who underwent radiological studies when attending the pediatric emergency department (ED) were reviewed. The radiographic interpretation by the pediatric EP, documented at the time of the ED visit, was compared with that made by the pediatric radiologist. Results: Data were available from 1,644/1,651 sets of conventional radiographs ordered for the pre-PACS study period and from 1,430/1,431 sets of digital radiographic studies for the post-PACS study period. The prevalence of positive radiological studies as per the radiologists was 32.2% (pre-PACS study period) vs. 28.7% (post-PACS study period). Diagnostic performance of the pediatric EPs for the two time periods was as follows: overall accuracy, 98.1% (95% confidence interval [CI] = 94.5% to 100%) vs. 98.5% (95% CI = 87.5% to 100%); sensitivity, 96.4% (95% CI = 94.5% to 97.8%) vs. 98.1% (95% CI = 96.2% to 99.2%); specificity, 98.9% (95% CI = 98.1% to 99.4%) vs. 98.6% (95% CI = 97.7% to 99.3%); negative predictive value, 98.3% (95% CI = 97.4% to 99.0%) vs. 99.2% (95% CI = 98.5% to 99.7%); and positive predictive value, 97.7% (95% CI = 96.0% to 98.8%) vs. 96.6% (95% CI = 94.4% to 98.2%). The proportion of false negatives (FN) was 1.2% (19/1,644) vs. 0.6% (8/1,430). Only one FN patient for each time period required immediate follow-up for a missed diagnosis. Conclusions: Radiograph interpretations by pediatric EPs with digital studies remain as accurate in comparison with assessments performed by using conventional radiographs.

Original languageEnglish (US)
Pages (from-to)186-190
Number of pages5
JournalAcademic Emergency Medicine
Volume13
Issue number2
DOIs
StatePublished - Feb 2006
Externally publishedYes

Fingerprint

Radiology Information Systems
Emergencies
Confidence Intervals
Pediatrics
Physicians
Hospital Emergency Service
Medical Records

Keywords

  • Diagnostic radiology
  • Emergency medicine

ASJC Scopus subject areas

  • Emergency Medicine

Cite this

The effect of picture archiving and communications systems on the accuracy of diagnostic interpretation of pediatric emergency physicians. / Gouin, Serge; Patel, Hema; Bergeron, Sylvie; Amre, Devendra; Guérin, Ronald.

In: Academic Emergency Medicine, Vol. 13, No. 2, 02.2006, p. 186-190.

Research output: Contribution to journalArticle

Gouin, Serge ; Patel, Hema ; Bergeron, Sylvie ; Amre, Devendra ; Guérin, Ronald. / The effect of picture archiving and communications systems on the accuracy of diagnostic interpretation of pediatric emergency physicians. In: Academic Emergency Medicine. 2006 ; Vol. 13, No. 2. pp. 186-190.
@article{af09369e16264e03a1f1c9a80d8a7a3d,
title = "The effect of picture archiving and communications systems on the accuracy of diagnostic interpretation of pediatric emergency physicians",
abstract = "Objectives: To compare the accuracy of diagnostic interpretation of radiographs by pediatric emergency physicians (EPs) before and after the introduction of a Picture Archiving and Communications System (PACS). Methods: The pre-PACS study period included results from September 2001, when patients were evaluated by using only conventional radiographs. The post-PACS study period consisted of results from September 2002, when patients were evaluated by using only digital radiographic studies. During these periods, consecutive medical records of all patients who underwent radiological studies when attending the pediatric emergency department (ED) were reviewed. The radiographic interpretation by the pediatric EP, documented at the time of the ED visit, was compared with that made by the pediatric radiologist. Results: Data were available from 1,644/1,651 sets of conventional radiographs ordered for the pre-PACS study period and from 1,430/1,431 sets of digital radiographic studies for the post-PACS study period. The prevalence of positive radiological studies as per the radiologists was 32.2{\%} (pre-PACS study period) vs. 28.7{\%} (post-PACS study period). Diagnostic performance of the pediatric EPs for the two time periods was as follows: overall accuracy, 98.1{\%} (95{\%} confidence interval [CI] = 94.5{\%} to 100{\%}) vs. 98.5{\%} (95{\%} CI = 87.5{\%} to 100{\%}); sensitivity, 96.4{\%} (95{\%} CI = 94.5{\%} to 97.8{\%}) vs. 98.1{\%} (95{\%} CI = 96.2{\%} to 99.2{\%}); specificity, 98.9{\%} (95{\%} CI = 98.1{\%} to 99.4{\%}) vs. 98.6{\%} (95{\%} CI = 97.7{\%} to 99.3{\%}); negative predictive value, 98.3{\%} (95{\%} CI = 97.4{\%} to 99.0{\%}) vs. 99.2{\%} (95{\%} CI = 98.5{\%} to 99.7{\%}); and positive predictive value, 97.7{\%} (95{\%} CI = 96.0{\%} to 98.8{\%}) vs. 96.6{\%} (95{\%} CI = 94.4{\%} to 98.2{\%}). The proportion of false negatives (FN) was 1.2{\%} (19/1,644) vs. 0.6{\%} (8/1,430). Only one FN patient for each time period required immediate follow-up for a missed diagnosis. Conclusions: Radiograph interpretations by pediatric EPs with digital studies remain as accurate in comparison with assessments performed by using conventional radiographs.",
keywords = "Diagnostic radiology, Emergency medicine",
author = "Serge Gouin and Hema Patel and Sylvie Bergeron and Devendra Amre and Ronald Gu{\'e}rin",
year = "2006",
month = "2",
doi = "10.1197/j.aem.2005.08.006",
language = "English (US)",
volume = "13",
pages = "186--190",
journal = "Academic Emergency Medicine",
issn = "1069-6563",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - The effect of picture archiving and communications systems on the accuracy of diagnostic interpretation of pediatric emergency physicians

AU - Gouin, Serge

AU - Patel, Hema

AU - Bergeron, Sylvie

AU - Amre, Devendra

AU - Guérin, Ronald

PY - 2006/2

Y1 - 2006/2

N2 - Objectives: To compare the accuracy of diagnostic interpretation of radiographs by pediatric emergency physicians (EPs) before and after the introduction of a Picture Archiving and Communications System (PACS). Methods: The pre-PACS study period included results from September 2001, when patients were evaluated by using only conventional radiographs. The post-PACS study period consisted of results from September 2002, when patients were evaluated by using only digital radiographic studies. During these periods, consecutive medical records of all patients who underwent radiological studies when attending the pediatric emergency department (ED) were reviewed. The radiographic interpretation by the pediatric EP, documented at the time of the ED visit, was compared with that made by the pediatric radiologist. Results: Data were available from 1,644/1,651 sets of conventional radiographs ordered for the pre-PACS study period and from 1,430/1,431 sets of digital radiographic studies for the post-PACS study period. The prevalence of positive radiological studies as per the radiologists was 32.2% (pre-PACS study period) vs. 28.7% (post-PACS study period). Diagnostic performance of the pediatric EPs for the two time periods was as follows: overall accuracy, 98.1% (95% confidence interval [CI] = 94.5% to 100%) vs. 98.5% (95% CI = 87.5% to 100%); sensitivity, 96.4% (95% CI = 94.5% to 97.8%) vs. 98.1% (95% CI = 96.2% to 99.2%); specificity, 98.9% (95% CI = 98.1% to 99.4%) vs. 98.6% (95% CI = 97.7% to 99.3%); negative predictive value, 98.3% (95% CI = 97.4% to 99.0%) vs. 99.2% (95% CI = 98.5% to 99.7%); and positive predictive value, 97.7% (95% CI = 96.0% to 98.8%) vs. 96.6% (95% CI = 94.4% to 98.2%). The proportion of false negatives (FN) was 1.2% (19/1,644) vs. 0.6% (8/1,430). Only one FN patient for each time period required immediate follow-up for a missed diagnosis. Conclusions: Radiograph interpretations by pediatric EPs with digital studies remain as accurate in comparison with assessments performed by using conventional radiographs.

AB - Objectives: To compare the accuracy of diagnostic interpretation of radiographs by pediatric emergency physicians (EPs) before and after the introduction of a Picture Archiving and Communications System (PACS). Methods: The pre-PACS study period included results from September 2001, when patients were evaluated by using only conventional radiographs. The post-PACS study period consisted of results from September 2002, when patients were evaluated by using only digital radiographic studies. During these periods, consecutive medical records of all patients who underwent radiological studies when attending the pediatric emergency department (ED) were reviewed. The radiographic interpretation by the pediatric EP, documented at the time of the ED visit, was compared with that made by the pediatric radiologist. Results: Data were available from 1,644/1,651 sets of conventional radiographs ordered for the pre-PACS study period and from 1,430/1,431 sets of digital radiographic studies for the post-PACS study period. The prevalence of positive radiological studies as per the radiologists was 32.2% (pre-PACS study period) vs. 28.7% (post-PACS study period). Diagnostic performance of the pediatric EPs for the two time periods was as follows: overall accuracy, 98.1% (95% confidence interval [CI] = 94.5% to 100%) vs. 98.5% (95% CI = 87.5% to 100%); sensitivity, 96.4% (95% CI = 94.5% to 97.8%) vs. 98.1% (95% CI = 96.2% to 99.2%); specificity, 98.9% (95% CI = 98.1% to 99.4%) vs. 98.6% (95% CI = 97.7% to 99.3%); negative predictive value, 98.3% (95% CI = 97.4% to 99.0%) vs. 99.2% (95% CI = 98.5% to 99.7%); and positive predictive value, 97.7% (95% CI = 96.0% to 98.8%) vs. 96.6% (95% CI = 94.4% to 98.2%). The proportion of false negatives (FN) was 1.2% (19/1,644) vs. 0.6% (8/1,430). Only one FN patient for each time period required immediate follow-up for a missed diagnosis. Conclusions: Radiograph interpretations by pediatric EPs with digital studies remain as accurate in comparison with assessments performed by using conventional radiographs.

KW - Diagnostic radiology

KW - Emergency medicine

UR - http://www.scopus.com/inward/record.url?scp=31544474619&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=31544474619&partnerID=8YFLogxK

U2 - 10.1197/j.aem.2005.08.006

DO - 10.1197/j.aem.2005.08.006

M3 - Article

C2 - 16436791

AN - SCOPUS:31544474619

VL - 13

SP - 186

EP - 190

JO - Academic Emergency Medicine

JF - Academic Emergency Medicine

SN - 1069-6563

IS - 2

ER -