A comparison of methods to detect urinary tract infections using electronic data

Timothy Landers, Mandar Apte, Sandra Hyman, Yoko Furuya, Sharon Glied, Elaine Larson

Research output: Contribution to journalArticle

Abstract

Background: The use of electronic medical records to identify common health care-associated infections (HAIs), including pneumonia, surgical site infections, bloodstream infections, and urinary tract infections (UTIs), has been proposed to help perform HAI surveillance and guide infection prevention efforts. Increased attention on HAIs has led to public health reporting requirements and a focus on quality improvement activities around HAIs. Traditional surveillance to detect HAIs and focus prevention efforts is labor intensive, and computer algorithms could be useful to screen electronic data and provide actionable information. Methods: Seven computer-based decision rules to identify UTIs were compared in a sample of 33,834 admissions to an urban academic health center. These decision rules included combinations of laboratory data, patient clinical data, and administrative data (for example, International Statistical Classification of Diseases and Related Health Problems, Ninth Revision [ICD-9] codes). Results: Of 33,834 hospital admissions, 3,870 UTIs were identified by at least one of the decision rules. The use of ICD-9 codes alone identified 2,614 UTIs. Laboratorybased definitions identified 2,773 infections, but when the presence of fever was included, only 1,125 UTIs were identified. The estimated sensitivity of ICD-9 codes was 55.6% (95% confidence interval [CI], 52.5%-58.5%) when compared with a culture-and symptom-based definition. Of the UTIs identified by ICD-9 codes, 167/1,125 (14.8%) also met two urine-culture decision rules. Discussion: Use of the example of UTI identification shows how different algorithms may be appropriate, depending on the goal of case identification. Electronic surveillance methods may be beneficial for mandatory reporting, process improvement, and economic analysis.

Original languageEnglish (US)
Pages (from-to)411-417
Number of pages7
JournalJoint Commission Journal on Quality and Patient Safety
Volume36
Issue number9
StatePublished - Sep 2010

Fingerprint

International Classification of Diseases
Urinary Tract Infections
Cross Infection
Infection
Mandatory Reporting
Urban Health
Surgical Wound Infection
Electronic Health Records
Quality Improvement
Pneumonia
Fever
Public Health
Economics
Urine
Confidence Intervals

ASJC Scopus subject areas

  • Leadership and Management

Cite this

A comparison of methods to detect urinary tract infections using electronic data. / Landers, Timothy; Apte, Mandar; Hyman, Sandra; Furuya, Yoko; Glied, Sharon; Larson, Elaine.

In: Joint Commission Journal on Quality and Patient Safety, Vol. 36, No. 9, 09.2010, p. 411-417.

Research output: Contribution to journalArticle

Landers, Timothy ; Apte, Mandar ; Hyman, Sandra ; Furuya, Yoko ; Glied, Sharon ; Larson, Elaine. / A comparison of methods to detect urinary tract infections using electronic data. In: Joint Commission Journal on Quality and Patient Safety. 2010 ; Vol. 36, No. 9. pp. 411-417.
@article{70ae16a15275496b803dfd4fe873548c,
title = "A comparison of methods to detect urinary tract infections using electronic data",
abstract = "Background: The use of electronic medical records to identify common health care-associated infections (HAIs), including pneumonia, surgical site infections, bloodstream infections, and urinary tract infections (UTIs), has been proposed to help perform HAI surveillance and guide infection prevention efforts. Increased attention on HAIs has led to public health reporting requirements and a focus on quality improvement activities around HAIs. Traditional surveillance to detect HAIs and focus prevention efforts is labor intensive, and computer algorithms could be useful to screen electronic data and provide actionable information. Methods: Seven computer-based decision rules to identify UTIs were compared in a sample of 33,834 admissions to an urban academic health center. These decision rules included combinations of laboratory data, patient clinical data, and administrative data (for example, International Statistical Classification of Diseases and Related Health Problems, Ninth Revision [ICD-9] codes). Results: Of 33,834 hospital admissions, 3,870 UTIs were identified by at least one of the decision rules. The use of ICD-9 codes alone identified 2,614 UTIs. Laboratorybased definitions identified 2,773 infections, but when the presence of fever was included, only 1,125 UTIs were identified. The estimated sensitivity of ICD-9 codes was 55.6{\%} (95{\%} confidence interval [CI], 52.5{\%}-58.5{\%}) when compared with a culture-and symptom-based definition. Of the UTIs identified by ICD-9 codes, 167/1,125 (14.8{\%}) also met two urine-culture decision rules. Discussion: Use of the example of UTI identification shows how different algorithms may be appropriate, depending on the goal of case identification. Electronic surveillance methods may be beneficial for mandatory reporting, process improvement, and economic analysis.",
author = "Timothy Landers and Mandar Apte and Sandra Hyman and Yoko Furuya and Sharon Glied and Elaine Larson",
year = "2010",
month = "9",
language = "English (US)",
volume = "36",
pages = "411--417",
journal = "Joint Commission Journal on Quality and Patient Safety",
issn = "1553-7250",
publisher = "Joint Commission Resources, Inc.",
number = "9",

}

TY - JOUR

T1 - A comparison of methods to detect urinary tract infections using electronic data

AU - Landers, Timothy

AU - Apte, Mandar

AU - Hyman, Sandra

AU - Furuya, Yoko

AU - Glied, Sharon

AU - Larson, Elaine

PY - 2010/9

Y1 - 2010/9

N2 - Background: The use of electronic medical records to identify common health care-associated infections (HAIs), including pneumonia, surgical site infections, bloodstream infections, and urinary tract infections (UTIs), has been proposed to help perform HAI surveillance and guide infection prevention efforts. Increased attention on HAIs has led to public health reporting requirements and a focus on quality improvement activities around HAIs. Traditional surveillance to detect HAIs and focus prevention efforts is labor intensive, and computer algorithms could be useful to screen electronic data and provide actionable information. Methods: Seven computer-based decision rules to identify UTIs were compared in a sample of 33,834 admissions to an urban academic health center. These decision rules included combinations of laboratory data, patient clinical data, and administrative data (for example, International Statistical Classification of Diseases and Related Health Problems, Ninth Revision [ICD-9] codes). Results: Of 33,834 hospital admissions, 3,870 UTIs were identified by at least one of the decision rules. The use of ICD-9 codes alone identified 2,614 UTIs. Laboratorybased definitions identified 2,773 infections, but when the presence of fever was included, only 1,125 UTIs were identified. The estimated sensitivity of ICD-9 codes was 55.6% (95% confidence interval [CI], 52.5%-58.5%) when compared with a culture-and symptom-based definition. Of the UTIs identified by ICD-9 codes, 167/1,125 (14.8%) also met two urine-culture decision rules. Discussion: Use of the example of UTI identification shows how different algorithms may be appropriate, depending on the goal of case identification. Electronic surveillance methods may be beneficial for mandatory reporting, process improvement, and economic analysis.

AB - Background: The use of electronic medical records to identify common health care-associated infections (HAIs), including pneumonia, surgical site infections, bloodstream infections, and urinary tract infections (UTIs), has been proposed to help perform HAI surveillance and guide infection prevention efforts. Increased attention on HAIs has led to public health reporting requirements and a focus on quality improvement activities around HAIs. Traditional surveillance to detect HAIs and focus prevention efforts is labor intensive, and computer algorithms could be useful to screen electronic data and provide actionable information. Methods: Seven computer-based decision rules to identify UTIs were compared in a sample of 33,834 admissions to an urban academic health center. These decision rules included combinations of laboratory data, patient clinical data, and administrative data (for example, International Statistical Classification of Diseases and Related Health Problems, Ninth Revision [ICD-9] codes). Results: Of 33,834 hospital admissions, 3,870 UTIs were identified by at least one of the decision rules. The use of ICD-9 codes alone identified 2,614 UTIs. Laboratorybased definitions identified 2,773 infections, but when the presence of fever was included, only 1,125 UTIs were identified. The estimated sensitivity of ICD-9 codes was 55.6% (95% confidence interval [CI], 52.5%-58.5%) when compared with a culture-and symptom-based definition. Of the UTIs identified by ICD-9 codes, 167/1,125 (14.8%) also met two urine-culture decision rules. Discussion: Use of the example of UTI identification shows how different algorithms may be appropriate, depending on the goal of case identification. Electronic surveillance methods may be beneficial for mandatory reporting, process improvement, and economic analysis.

UR - http://www.scopus.com/inward/record.url?scp=77958177772&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77958177772&partnerID=8YFLogxK

M3 - Article

C2 - 20873674

AN - SCOPUS:77958177772

VL - 36

SP - 411

EP - 417

JO - Joint Commission Journal on Quality and Patient Safety

JF - Joint Commission Journal on Quality and Patient Safety

SN - 1553-7250

IS - 9

ER -