Electronic health information quality challenges and interventions to improve public health surveillance data and practice

Brian Dixon, Jason A. Siegel, Tanya V. Oemig, Shaun Grannis

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

Objective. We examined completeness, an attribute of data quality, in the context of electronic laboratory reporting (ELR) of notifable disease information to public health agencies. Methods. We extracted more than seven million ELR messages from multiple clinical information systems in two states. We calculated and compared the completeness of various data fields within the messages that were identifed to be important to public health reporting processes. We compared unaltered, original messages from source systems with similar messages from another state as well as messages enriched by a health information exchange (HIE). Our analysis focused on calculating completeness (i.e., the number of nonmiss-ing values) for fields deemed important for inclusion in notifable disease case reports. Results. The completeness of data fields for laboratory transactions varied across clinical information systems and jurisdictions. Fields identifying the patient and test results were usually complete (97%-100%). Fields containing patient demographics, patient contact information, and provider contact information were suboptimal (6%-89%). Transactions enhanced by the HIE were found to be more complete (increases ranged from 2% to 25%) than the original messages. Conclusion. ELR data from clinical information systems can be of suboptimal quality. Public health monitoring of data sources and augmentation of ELR message content using HIE services can improve data quality.

Original languageEnglish
Pages (from-to)546-553
Number of pages8
JournalPublic Health Reports
Volume128
Issue number6
StatePublished - Nov 2013

Fingerprint

Public Health Surveillance
Public Health Practice
Information Systems
Health
Public Health
Information Services
Information Storage and Retrieval
Research Design
Demography
Health Information Exchange

ASJC Scopus subject areas

  • Public Health, Environmental and Occupational Health

Cite this

Electronic health information quality challenges and interventions to improve public health surveillance data and practice. / Dixon, Brian; Siegel, Jason A.; Oemig, Tanya V.; Grannis, Shaun.

In: Public Health Reports, Vol. 128, No. 6, 11.2013, p. 546-553.

Research output: Contribution to journalArticle

@article{e8e26338d7704d75906fdf855888bb42,
title = "Electronic health information quality challenges and interventions to improve public health surveillance data and practice",
abstract = "Objective. We examined completeness, an attribute of data quality, in the context of electronic laboratory reporting (ELR) of notifable disease information to public health agencies. Methods. We extracted more than seven million ELR messages from multiple clinical information systems in two states. We calculated and compared the completeness of various data fields within the messages that were identifed to be important to public health reporting processes. We compared unaltered, original messages from source systems with similar messages from another state as well as messages enriched by a health information exchange (HIE). Our analysis focused on calculating completeness (i.e., the number of nonmiss-ing values) for fields deemed important for inclusion in notifable disease case reports. Results. The completeness of data fields for laboratory transactions varied across clinical information systems and jurisdictions. Fields identifying the patient and test results were usually complete (97{\%}-100{\%}). Fields containing patient demographics, patient contact information, and provider contact information were suboptimal (6{\%}-89{\%}). Transactions enhanced by the HIE were found to be more complete (increases ranged from 2{\%} to 25{\%}) than the original messages. Conclusion. ELR data from clinical information systems can be of suboptimal quality. Public health monitoring of data sources and augmentation of ELR message content using HIE services can improve data quality.",
author = "Brian Dixon and Siegel, {Jason A.} and Oemig, {Tanya V.} and Shaun Grannis",
year = "2013",
month = "11",
language = "English",
volume = "128",
pages = "546--553",
journal = "Public Health Reports",
issn = "0033-3549",
publisher = "Association of Schools of Public Health",
number = "6",

}

TY - JOUR

T1 - Electronic health information quality challenges and interventions to improve public health surveillance data and practice

AU - Dixon, Brian

AU - Siegel, Jason A.

AU - Oemig, Tanya V.

AU - Grannis, Shaun

PY - 2013/11

Y1 - 2013/11

N2 - Objective. We examined completeness, an attribute of data quality, in the context of electronic laboratory reporting (ELR) of notifable disease information to public health agencies. Methods. We extracted more than seven million ELR messages from multiple clinical information systems in two states. We calculated and compared the completeness of various data fields within the messages that were identifed to be important to public health reporting processes. We compared unaltered, original messages from source systems with similar messages from another state as well as messages enriched by a health information exchange (HIE). Our analysis focused on calculating completeness (i.e., the number of nonmiss-ing values) for fields deemed important for inclusion in notifable disease case reports. Results. The completeness of data fields for laboratory transactions varied across clinical information systems and jurisdictions. Fields identifying the patient and test results were usually complete (97%-100%). Fields containing patient demographics, patient contact information, and provider contact information were suboptimal (6%-89%). Transactions enhanced by the HIE were found to be more complete (increases ranged from 2% to 25%) than the original messages. Conclusion. ELR data from clinical information systems can be of suboptimal quality. Public health monitoring of data sources and augmentation of ELR message content using HIE services can improve data quality.

AB - Objective. We examined completeness, an attribute of data quality, in the context of electronic laboratory reporting (ELR) of notifable disease information to public health agencies. Methods. We extracted more than seven million ELR messages from multiple clinical information systems in two states. We calculated and compared the completeness of various data fields within the messages that were identifed to be important to public health reporting processes. We compared unaltered, original messages from source systems with similar messages from another state as well as messages enriched by a health information exchange (HIE). Our analysis focused on calculating completeness (i.e., the number of nonmiss-ing values) for fields deemed important for inclusion in notifable disease case reports. Results. The completeness of data fields for laboratory transactions varied across clinical information systems and jurisdictions. Fields identifying the patient and test results were usually complete (97%-100%). Fields containing patient demographics, patient contact information, and provider contact information were suboptimal (6%-89%). Transactions enhanced by the HIE were found to be more complete (increases ranged from 2% to 25%) than the original messages. Conclusion. ELR data from clinical information systems can be of suboptimal quality. Public health monitoring of data sources and augmentation of ELR message content using HIE services can improve data quality.

UR - http://www.scopus.com/inward/record.url?scp=84887332915&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84887332915&partnerID=8YFLogxK

M3 - Article

VL - 128

SP - 546

EP - 553

JO - Public Health Reports

JF - Public Health Reports

SN - 0033-3549

IS - 6

ER -