Clustered and missing data in the US National Trauma Data Bank: Implications for analysis

Bahman Sayyar Roudsari, C. Field, R. Caetano

Research output: Contribution to journalArticle

45 Citations (Scopus)

Abstract

Background: Injury researchers are increasingly using the US National Trauma Data Bank (NTDB). However, there are some methodological issues that might threaten the validity of studies that use this database for injury research. Methods: Two methodological issues were evaluated: clustering of patients within trauma centers and missing data. To illustrate how these issues might affect the results of a study, the following four analytical approaches that evaluated the association between patients' blood alcohol concentration (BAC) in the emergency department (ED), patients' resource utilization, and ED or hospital disposition were compared: (A) deleting subjects with missing BAC and ignoring clustering of patients within trauma centers; (B) deleting subjects with missing BAC while taking into account clustering; (C) using imputed values for patients' BAC and ignoring the clustering issue; (D) using the imputed data while taking into account clustering. Results: Adjustment for clustering of patients within trauma centers increased the CIs in models B and D. The results of the analyses based on imputed data showed that estimates based on complete case analysis were biased. For example, the odds ratio for the use of a head CT scan fell from 1.84 (95% Cl 1.49 to 2.28) in approach B to 1.26 (95% Cl 0.98 to 1.64) in approach D. Conclusions: Excluding patients with missing values for BAC in studies that evaluate the association between this variable and patients' resource utilization and ED or hospital disposition, using the NTDB, led to biased estimates. Furthermore, ignoring the clustering design led to artificially narrow CIs.

Original languageEnglish (US)
Pages (from-to)96-100
Number of pages5
JournalInjury Prevention
Volume14
Issue number2
DOIs
StatePublished - Apr 1 2008
Externally publishedYes

Fingerprint

Cluster Analysis
Databases
Wounds and Injuries
Trauma Centers
Hospital Emergency Service
Odds Ratio
Head
Research Personnel
Blood Alcohol Content
Research

ASJC Scopus subject areas

  • Pediatrics, Perinatology, and Child Health

Cite this

Clustered and missing data in the US National Trauma Data Bank : Implications for analysis. / Sayyar Roudsari, Bahman; Field, C.; Caetano, R.

In: Injury Prevention, Vol. 14, No. 2, 01.04.2008, p. 96-100.

Research output: Contribution to journalArticle

@article{2a9211f072774251bf0e3499444cb554,
title = "Clustered and missing data in the US National Trauma Data Bank: Implications for analysis",
abstract = "Background: Injury researchers are increasingly using the US National Trauma Data Bank (NTDB). However, there are some methodological issues that might threaten the validity of studies that use this database for injury research. Methods: Two methodological issues were evaluated: clustering of patients within trauma centers and missing data. To illustrate how these issues might affect the results of a study, the following four analytical approaches that evaluated the association between patients' blood alcohol concentration (BAC) in the emergency department (ED), patients' resource utilization, and ED or hospital disposition were compared: (A) deleting subjects with missing BAC and ignoring clustering of patients within trauma centers; (B) deleting subjects with missing BAC while taking into account clustering; (C) using imputed values for patients' BAC and ignoring the clustering issue; (D) using the imputed data while taking into account clustering. Results: Adjustment for clustering of patients within trauma centers increased the CIs in models B and D. The results of the analyses based on imputed data showed that estimates based on complete case analysis were biased. For example, the odds ratio for the use of a head CT scan fell from 1.84 (95{\%} Cl 1.49 to 2.28) in approach B to 1.26 (95{\%} Cl 0.98 to 1.64) in approach D. Conclusions: Excluding patients with missing values for BAC in studies that evaluate the association between this variable and patients' resource utilization and ED or hospital disposition, using the NTDB, led to biased estimates. Furthermore, ignoring the clustering design led to artificially narrow CIs.",
author = "{Sayyar Roudsari}, Bahman and C. Field and R. Caetano",
year = "2008",
month = "4",
day = "1",
doi = "10.1136/ip.2007.017129",
language = "English (US)",
volume = "14",
pages = "96--100",
journal = "Injury Prevention",
issn = "1353-8047",
publisher = "BMJ Publishing Group",
number = "2",

}

TY - JOUR

T1 - Clustered and missing data in the US National Trauma Data Bank

T2 - Implications for analysis

AU - Sayyar Roudsari, Bahman

AU - Field, C.

AU - Caetano, R.

PY - 2008/4/1

Y1 - 2008/4/1

N2 - Background: Injury researchers are increasingly using the US National Trauma Data Bank (NTDB). However, there are some methodological issues that might threaten the validity of studies that use this database for injury research. Methods: Two methodological issues were evaluated: clustering of patients within trauma centers and missing data. To illustrate how these issues might affect the results of a study, the following four analytical approaches that evaluated the association between patients' blood alcohol concentration (BAC) in the emergency department (ED), patients' resource utilization, and ED or hospital disposition were compared: (A) deleting subjects with missing BAC and ignoring clustering of patients within trauma centers; (B) deleting subjects with missing BAC while taking into account clustering; (C) using imputed values for patients' BAC and ignoring the clustering issue; (D) using the imputed data while taking into account clustering. Results: Adjustment for clustering of patients within trauma centers increased the CIs in models B and D. The results of the analyses based on imputed data showed that estimates based on complete case analysis were biased. For example, the odds ratio for the use of a head CT scan fell from 1.84 (95% Cl 1.49 to 2.28) in approach B to 1.26 (95% Cl 0.98 to 1.64) in approach D. Conclusions: Excluding patients with missing values for BAC in studies that evaluate the association between this variable and patients' resource utilization and ED or hospital disposition, using the NTDB, led to biased estimates. Furthermore, ignoring the clustering design led to artificially narrow CIs.

AB - Background: Injury researchers are increasingly using the US National Trauma Data Bank (NTDB). However, there are some methodological issues that might threaten the validity of studies that use this database for injury research. Methods: Two methodological issues were evaluated: clustering of patients within trauma centers and missing data. To illustrate how these issues might affect the results of a study, the following four analytical approaches that evaluated the association between patients' blood alcohol concentration (BAC) in the emergency department (ED), patients' resource utilization, and ED or hospital disposition were compared: (A) deleting subjects with missing BAC and ignoring clustering of patients within trauma centers; (B) deleting subjects with missing BAC while taking into account clustering; (C) using imputed values for patients' BAC and ignoring the clustering issue; (D) using the imputed data while taking into account clustering. Results: Adjustment for clustering of patients within trauma centers increased the CIs in models B and D. The results of the analyses based on imputed data showed that estimates based on complete case analysis were biased. For example, the odds ratio for the use of a head CT scan fell from 1.84 (95% Cl 1.49 to 2.28) in approach B to 1.26 (95% Cl 0.98 to 1.64) in approach D. Conclusions: Excluding patients with missing values for BAC in studies that evaluate the association between this variable and patients' resource utilization and ED or hospital disposition, using the NTDB, led to biased estimates. Furthermore, ignoring the clustering design led to artificially narrow CIs.

UR - http://www.scopus.com/inward/record.url?scp=42149109802&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=42149109802&partnerID=8YFLogxK

U2 - 10.1136/ip.2007.017129

DO - 10.1136/ip.2007.017129

M3 - Article

C2 - 18388229

AN - SCOPUS:42149109802

VL - 14

SP - 96

EP - 100

JO - Injury Prevention

JF - Injury Prevention

SN - 1353-8047

IS - 2

ER -