Analysis of clustered data in multicentre trauma studies

Bahman Sayyar Roudsari, Avery Nathens, Thomas Koepsell, Charles Mock, Frederick Rivara

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

Introduction: In order to design multicentre studies an estimate of the correlation of the observations within each centre is necessary. A standard measure of the correlation between observations within each centre is the Intraclass Correlation Coefficient (ICC). Method: We used the National Trauma Data Bank (NTDB). By 2004, 448 trauma centres (including 110 level I and 123 level II trauma centres) from 43 states and US territories contributed over 1.2 million records to the NTDB. Data of patients directly transported from the scene of injury to level I or II trauma centres were used to calculate the ICC of in-hospital trauma fatality and emergency department (ED) shock rate. Results: The ICCs of ED shock and in-hospital fatality rate were 0.010 (95% confidence interval (CI): 0.003-0.018) and 0.039 (95% CI: 0.028-0.050), respectively. The ICC of shock in the ED was the highest for penetrating injuries (0.017, 95% CI: 0.003-0.032) and the lowest for women (0.008, 95% CI: 0.002-0.013) although the observed difference between men and women was not statistically significant. The ICC of trauma fatality was the highest for penetrating injuries (0.073, 95% CI: 0.047-0.098), and the lowest for blunt injuries (0.029, 95% CI: 0.020-0.037). Discussion: Although the calculated ICCs might seem so small as to be ignored, the required sample size in studies with exclusively exposed or non-exposed clusters depends on the ICC and the average number of subjects within clusters. Therefore, investigators should be aware of the influence that these ICCs might have on sample size and power of their studies.

Original languageEnglish (US)
Pages (from-to)614-621
Number of pages8
JournalInjury
Volume37
Issue number7
DOIs
StatePublished - Jul 1 2006
Externally publishedYes

Fingerprint

Multicenter Studies
Confidence Intervals
Wounds and Injuries
Trauma Centers
Hospital Emergency Service
Shock
Sample Size
Databases
Nonpenetrating Wounds
Research Personnel

Keywords

  • Cluster design
  • Injury
  • Intraclass correlation coefficient (ICC)
  • Mortality
  • Multicentre studies
  • Shock
  • Trauma

ASJC Scopus subject areas

  • Emergency Medicine
  • Orthopedics and Sports Medicine

Cite this

Analysis of clustered data in multicentre trauma studies. / Sayyar Roudsari, Bahman; Nathens, Avery; Koepsell, Thomas; Mock, Charles; Rivara, Frederick.

In: Injury, Vol. 37, No. 7, 01.07.2006, p. 614-621.

Research output: Contribution to journalArticle

Sayyar Roudsari, B, Nathens, A, Koepsell, T, Mock, C & Rivara, F 2006, 'Analysis of clustered data in multicentre trauma studies', Injury, vol. 37, no. 7, pp. 614-621. https://doi.org/10.1016/j.injury.2006.02.010
Sayyar Roudsari, Bahman ; Nathens, Avery ; Koepsell, Thomas ; Mock, Charles ; Rivara, Frederick. / Analysis of clustered data in multicentre trauma studies. In: Injury. 2006 ; Vol. 37, No. 7. pp. 614-621.
@article{f0439e6dd0d948dcbb6b0db4ea4f68a5,
title = "Analysis of clustered data in multicentre trauma studies",
abstract = "Introduction: In order to design multicentre studies an estimate of the correlation of the observations within each centre is necessary. A standard measure of the correlation between observations within each centre is the Intraclass Correlation Coefficient (ICC). Method: We used the National Trauma Data Bank (NTDB). By 2004, 448 trauma centres (including 110 level I and 123 level II trauma centres) from 43 states and US territories contributed over 1.2 million records to the NTDB. Data of patients directly transported from the scene of injury to level I or II trauma centres were used to calculate the ICC of in-hospital trauma fatality and emergency department (ED) shock rate. Results: The ICCs of ED shock and in-hospital fatality rate were 0.010 (95{\%} confidence interval (CI): 0.003-0.018) and 0.039 (95{\%} CI: 0.028-0.050), respectively. The ICC of shock in the ED was the highest for penetrating injuries (0.017, 95{\%} CI: 0.003-0.032) and the lowest for women (0.008, 95{\%} CI: 0.002-0.013) although the observed difference between men and women was not statistically significant. The ICC of trauma fatality was the highest for penetrating injuries (0.073, 95{\%} CI: 0.047-0.098), and the lowest for blunt injuries (0.029, 95{\%} CI: 0.020-0.037). Discussion: Although the calculated ICCs might seem so small as to be ignored, the required sample size in studies with exclusively exposed or non-exposed clusters depends on the ICC and the average number of subjects within clusters. Therefore, investigators should be aware of the influence that these ICCs might have on sample size and power of their studies.",
keywords = "Cluster design, Injury, Intraclass correlation coefficient (ICC), Mortality, Multicentre studies, Shock, Trauma",
author = "{Sayyar Roudsari}, Bahman and Avery Nathens and Thomas Koepsell and Charles Mock and Frederick Rivara",
year = "2006",
month = "7",
day = "1",
doi = "10.1016/j.injury.2006.02.010",
language = "English (US)",
volume = "37",
pages = "614--621",
journal = "Injury",
issn = "0020-1383",
publisher = "Elsevier Limited",
number = "7",

}

TY - JOUR

T1 - Analysis of clustered data in multicentre trauma studies

AU - Sayyar Roudsari, Bahman

AU - Nathens, Avery

AU - Koepsell, Thomas

AU - Mock, Charles

AU - Rivara, Frederick

PY - 2006/7/1

Y1 - 2006/7/1

N2 - Introduction: In order to design multicentre studies an estimate of the correlation of the observations within each centre is necessary. A standard measure of the correlation between observations within each centre is the Intraclass Correlation Coefficient (ICC). Method: We used the National Trauma Data Bank (NTDB). By 2004, 448 trauma centres (including 110 level I and 123 level II trauma centres) from 43 states and US territories contributed over 1.2 million records to the NTDB. Data of patients directly transported from the scene of injury to level I or II trauma centres were used to calculate the ICC of in-hospital trauma fatality and emergency department (ED) shock rate. Results: The ICCs of ED shock and in-hospital fatality rate were 0.010 (95% confidence interval (CI): 0.003-0.018) and 0.039 (95% CI: 0.028-0.050), respectively. The ICC of shock in the ED was the highest for penetrating injuries (0.017, 95% CI: 0.003-0.032) and the lowest for women (0.008, 95% CI: 0.002-0.013) although the observed difference between men and women was not statistically significant. The ICC of trauma fatality was the highest for penetrating injuries (0.073, 95% CI: 0.047-0.098), and the lowest for blunt injuries (0.029, 95% CI: 0.020-0.037). Discussion: Although the calculated ICCs might seem so small as to be ignored, the required sample size in studies with exclusively exposed or non-exposed clusters depends on the ICC and the average number of subjects within clusters. Therefore, investigators should be aware of the influence that these ICCs might have on sample size and power of their studies.

AB - Introduction: In order to design multicentre studies an estimate of the correlation of the observations within each centre is necessary. A standard measure of the correlation between observations within each centre is the Intraclass Correlation Coefficient (ICC). Method: We used the National Trauma Data Bank (NTDB). By 2004, 448 trauma centres (including 110 level I and 123 level II trauma centres) from 43 states and US territories contributed over 1.2 million records to the NTDB. Data of patients directly transported from the scene of injury to level I or II trauma centres were used to calculate the ICC of in-hospital trauma fatality and emergency department (ED) shock rate. Results: The ICCs of ED shock and in-hospital fatality rate were 0.010 (95% confidence interval (CI): 0.003-0.018) and 0.039 (95% CI: 0.028-0.050), respectively. The ICC of shock in the ED was the highest for penetrating injuries (0.017, 95% CI: 0.003-0.032) and the lowest for women (0.008, 95% CI: 0.002-0.013) although the observed difference between men and women was not statistically significant. The ICC of trauma fatality was the highest for penetrating injuries (0.073, 95% CI: 0.047-0.098), and the lowest for blunt injuries (0.029, 95% CI: 0.020-0.037). Discussion: Although the calculated ICCs might seem so small as to be ignored, the required sample size in studies with exclusively exposed or non-exposed clusters depends on the ICC and the average number of subjects within clusters. Therefore, investigators should be aware of the influence that these ICCs might have on sample size and power of their studies.

KW - Cluster design

KW - Injury

KW - Intraclass correlation coefficient (ICC)

KW - Mortality

KW - Multicentre studies

KW - Shock

KW - Trauma

UR - http://www.scopus.com/inward/record.url?scp=33744924855&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33744924855&partnerID=8YFLogxK

U2 - 10.1016/j.injury.2006.02.010

DO - 10.1016/j.injury.2006.02.010

M3 - Article

C2 - 16769309

AN - SCOPUS:33744924855

VL - 37

SP - 614

EP - 621

JO - Injury

JF - Injury

SN - 0020-1383

IS - 7

ER -