Validation of probabilistic linkage to match de-identified ambulance records to a state trauma registry

Research output: Contribution to journalArticle

68 Citations (Scopus)

Abstract

Objectives: To validate the accuracy of using probabilistic linkage for matching de-identified ambulance records to a state trauma registry. Methods: This was a retrospective cohort analysis. Three thousand nine hundred nineteen true matches between ambulance and state trauma registry data from 1998 to 2003 were identified by deterministic matching on trauma identification number and verified by human review. Two thousand thirty-eight ambulance records from trauma patients not meeting criteria for a true match, and an identical number of trauma registry records randomly selected from the one local county served by a different EMS provider, were included as nonmatches. There were 17 variables considered for linkage, which included the following: age, gender, race, county, hospital, date, rural setting, call and arrival times, mechanism, penetrating injury, vital signs, intubation, and intoxication. Probabilistic linkage was used to link the two data sets, using seven different combinations of common variables (maximum, 17; minimum, 4). The sensitivity and specificity of identifying true matches and nonmatches (95% confidence intervals [95% CI]) were calculated for each combination of variables. Results: Using all 17 available variables, 3,766 of 3,919 true matches were appropriately linked (sensitivity, 96.1%; 95% CI = 95.4% to 96.7%), with eight mismatches (specificity, 99.6%; 95% CI = 99.2% to 99.8%). Sensitivity fell below 95% with 98% regardless of the number of variables included. Conclusions: Probabilistic linkage is a valid method for matching ambulance records to a trauma registry without the use of patient identifiers; however, the sensitivity of identifying true matches is critically dependent on the number and type of common variables included in the analysis.

Original languageEnglish (US)
Pages (from-to)69-75
Number of pages7
JournalAcademic Emergency Medicine
Volume13
Issue number1
DOIs
StatePublished - Jan 2006

Fingerprint

Ambulances
Registries
Wounds and Injuries
Confidence Intervals
County Hospitals
Vital Signs
Intubation
Cohort Studies
Sensitivity and Specificity

Keywords

  • Emergency medical services
  • Probabilistic linkage
  • Trauma

ASJC Scopus subject areas

  • Emergency Medicine

Cite this

Validation of probabilistic linkage to match de-identified ambulance records to a state trauma registry. / Newgard, Craig.

In: Academic Emergency Medicine, Vol. 13, No. 1, 01.2006, p. 69-75.

Research output: Contribution to journalArticle

@article{81a530db837b4f039d719328f2d131e7,
title = "Validation of probabilistic linkage to match de-identified ambulance records to a state trauma registry",
abstract = "Objectives: To validate the accuracy of using probabilistic linkage for matching de-identified ambulance records to a state trauma registry. Methods: This was a retrospective cohort analysis. Three thousand nine hundred nineteen true matches between ambulance and state trauma registry data from 1998 to 2003 were identified by deterministic matching on trauma identification number and verified by human review. Two thousand thirty-eight ambulance records from trauma patients not meeting criteria for a true match, and an identical number of trauma registry records randomly selected from the one local county served by a different EMS provider, were included as nonmatches. There were 17 variables considered for linkage, which included the following: age, gender, race, county, hospital, date, rural setting, call and arrival times, mechanism, penetrating injury, vital signs, intubation, and intoxication. Probabilistic linkage was used to link the two data sets, using seven different combinations of common variables (maximum, 17; minimum, 4). The sensitivity and specificity of identifying true matches and nonmatches (95{\%} confidence intervals [95{\%} CI]) were calculated for each combination of variables. Results: Using all 17 available variables, 3,766 of 3,919 true matches were appropriately linked (sensitivity, 96.1{\%}; 95{\%} CI = 95.4{\%} to 96.7{\%}), with eight mismatches (specificity, 99.6{\%}; 95{\%} CI = 99.2{\%} to 99.8{\%}). Sensitivity fell below 95{\%} with 98{\%} regardless of the number of variables included. Conclusions: Probabilistic linkage is a valid method for matching ambulance records to a trauma registry without the use of patient identifiers; however, the sensitivity of identifying true matches is critically dependent on the number and type of common variables included in the analysis.",
keywords = "Emergency medical services, Probabilistic linkage, Trauma",
author = "Craig Newgard",
year = "2006",
month = "1",
doi = "10.1197/j.aem.2005.07.029",
language = "English (US)",
volume = "13",
pages = "69--75",
journal = "Academic Emergency Medicine",
issn = "1069-6563",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - Validation of probabilistic linkage to match de-identified ambulance records to a state trauma registry

AU - Newgard, Craig

PY - 2006/1

Y1 - 2006/1

N2 - Objectives: To validate the accuracy of using probabilistic linkage for matching de-identified ambulance records to a state trauma registry. Methods: This was a retrospective cohort analysis. Three thousand nine hundred nineteen true matches between ambulance and state trauma registry data from 1998 to 2003 were identified by deterministic matching on trauma identification number and verified by human review. Two thousand thirty-eight ambulance records from trauma patients not meeting criteria for a true match, and an identical number of trauma registry records randomly selected from the one local county served by a different EMS provider, were included as nonmatches. There were 17 variables considered for linkage, which included the following: age, gender, race, county, hospital, date, rural setting, call and arrival times, mechanism, penetrating injury, vital signs, intubation, and intoxication. Probabilistic linkage was used to link the two data sets, using seven different combinations of common variables (maximum, 17; minimum, 4). The sensitivity and specificity of identifying true matches and nonmatches (95% confidence intervals [95% CI]) were calculated for each combination of variables. Results: Using all 17 available variables, 3,766 of 3,919 true matches were appropriately linked (sensitivity, 96.1%; 95% CI = 95.4% to 96.7%), with eight mismatches (specificity, 99.6%; 95% CI = 99.2% to 99.8%). Sensitivity fell below 95% with 98% regardless of the number of variables included. Conclusions: Probabilistic linkage is a valid method for matching ambulance records to a trauma registry without the use of patient identifiers; however, the sensitivity of identifying true matches is critically dependent on the number and type of common variables included in the analysis.

AB - Objectives: To validate the accuracy of using probabilistic linkage for matching de-identified ambulance records to a state trauma registry. Methods: This was a retrospective cohort analysis. Three thousand nine hundred nineteen true matches between ambulance and state trauma registry data from 1998 to 2003 were identified by deterministic matching on trauma identification number and verified by human review. Two thousand thirty-eight ambulance records from trauma patients not meeting criteria for a true match, and an identical number of trauma registry records randomly selected from the one local county served by a different EMS provider, were included as nonmatches. There were 17 variables considered for linkage, which included the following: age, gender, race, county, hospital, date, rural setting, call and arrival times, mechanism, penetrating injury, vital signs, intubation, and intoxication. Probabilistic linkage was used to link the two data sets, using seven different combinations of common variables (maximum, 17; minimum, 4). The sensitivity and specificity of identifying true matches and nonmatches (95% confidence intervals [95% CI]) were calculated for each combination of variables. Results: Using all 17 available variables, 3,766 of 3,919 true matches were appropriately linked (sensitivity, 96.1%; 95% CI = 95.4% to 96.7%), with eight mismatches (specificity, 99.6%; 95% CI = 99.2% to 99.8%). Sensitivity fell below 95% with 98% regardless of the number of variables included. Conclusions: Probabilistic linkage is a valid method for matching ambulance records to a trauma registry without the use of patient identifiers; however, the sensitivity of identifying true matches is critically dependent on the number and type of common variables included in the analysis.

KW - Emergency medical services

KW - Probabilistic linkage

KW - Trauma

UR - http://www.scopus.com/inward/record.url?scp=29244488576&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=29244488576&partnerID=8YFLogxK

U2 - 10.1197/j.aem.2005.07.029

DO - 10.1197/j.aem.2005.07.029

M3 - Article

VL - 13

SP - 69

EP - 75

JO - Academic Emergency Medicine

JF - Academic Emergency Medicine

SN - 1069-6563

IS - 1

ER -