The Feasibility of Using Large-Scale Text Mining to Detect Adverse Childhood Experiences in a VA-Treated Population

Kenric W. Hammond, Alon Ben-Ari, Ryan J. Laundry, Edward J. Boyko, Matthew H. Samore

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Free text in electronic health records resists large-scale analysis. Text records facts of interest not found in encoded data, and text mining enables their retrieval and quantification. The U.S. Department of Veterans Affairs (VA) clinical data repository affords an opportunity to apply text-mining methodology to study clinical questions in large populations. To assess the feasibility of text mining, investigation of the relationship between exposure to adverse childhood experiences (ACEs) and recorded diagnoses was conducted among all VA-treated Gulf war veterans, utilizing all progress notes recorded from 2000-2011. Text processing extracted ACE exposures recorded among 44.7 million clinical notes belonging to 243,973 veterans. The relationship of ACE exposure to adult illnesses was analyzed using logistic regression. Bias considerations were assessed. ACE score was strongly associated with suicide attempts and serious mental disorders (ORs = 1.84 to 1.97), and less so with behaviorally mediated and somatic conditions (ORs = 1.02 to 1.36) per unit. Bias adjustments did not remove persistent associations between ACE score and most illnesses. Text mining to detect ACE exposure in a large population was feasible. Analysis of the relationship between ACE score and adult health conditions yielded patterns of association consistent with prior research.

Original languageEnglish (US)
Pages (from-to)505-514
Number of pages10
JournalJournal of Traumatic Stress
Volume28
Issue number6
DOIs
StatePublished - Dec 1 2015
Externally publishedYes

Fingerprint

Data Mining
Veterans
Population
Gulf War
Social Adjustment
United States Department of Veterans Affairs
Electronic Health Records
Mental Disorders
Suicide
Logistic Models
Health
Research

ASJC Scopus subject areas

  • Clinical Psychology
  • Psychiatry and Mental health

Cite this

The Feasibility of Using Large-Scale Text Mining to Detect Adverse Childhood Experiences in a VA-Treated Population. / Hammond, Kenric W.; Ben-Ari, Alon; Laundry, Ryan J.; Boyko, Edward J.; Samore, Matthew H.

In: Journal of Traumatic Stress, Vol. 28, No. 6, 01.12.2015, p. 505-514.

Research output: Contribution to journalArticle

Hammond, Kenric W. ; Ben-Ari, Alon ; Laundry, Ryan J. ; Boyko, Edward J. ; Samore, Matthew H. / The Feasibility of Using Large-Scale Text Mining to Detect Adverse Childhood Experiences in a VA-Treated Population. In: Journal of Traumatic Stress. 2015 ; Vol. 28, No. 6. pp. 505-514.
@article{a6d1fdbe4790453fb294d180bdb93dae,
title = "The Feasibility of Using Large-Scale Text Mining to Detect Adverse Childhood Experiences in a VA-Treated Population",
abstract = "Free text in electronic health records resists large-scale analysis. Text records facts of interest not found in encoded data, and text mining enables their retrieval and quantification. The U.S. Department of Veterans Affairs (VA) clinical data repository affords an opportunity to apply text-mining methodology to study clinical questions in large populations. To assess the feasibility of text mining, investigation of the relationship between exposure to adverse childhood experiences (ACEs) and recorded diagnoses was conducted among all VA-treated Gulf war veterans, utilizing all progress notes recorded from 2000-2011. Text processing extracted ACE exposures recorded among 44.7 million clinical notes belonging to 243,973 veterans. The relationship of ACE exposure to adult illnesses was analyzed using logistic regression. Bias considerations were assessed. ACE score was strongly associated with suicide attempts and serious mental disorders (ORs = 1.84 to 1.97), and less so with behaviorally mediated and somatic conditions (ORs = 1.02 to 1.36) per unit. Bias adjustments did not remove persistent associations between ACE score and most illnesses. Text mining to detect ACE exposure in a large population was feasible. Analysis of the relationship between ACE score and adult health conditions yielded patterns of association consistent with prior research.",
author = "Hammond, {Kenric W.} and Alon Ben-Ari and Laundry, {Ryan J.} and Boyko, {Edward J.} and Samore, {Matthew H.}",
year = "2015",
month = "12",
day = "1",
doi = "10.1002/jts.22058",
language = "English (US)",
volume = "28",
pages = "505--514",
journal = "Journal of Traumatic Stress",
issn = "0894-9867",
publisher = "John Wiley and Sons Inc.",
number = "6",

}

TY - JOUR

T1 - The Feasibility of Using Large-Scale Text Mining to Detect Adverse Childhood Experiences in a VA-Treated Population

AU - Hammond, Kenric W.

AU - Ben-Ari, Alon

AU - Laundry, Ryan J.

AU - Boyko, Edward J.

AU - Samore, Matthew H.

PY - 2015/12/1

Y1 - 2015/12/1

N2 - Free text in electronic health records resists large-scale analysis. Text records facts of interest not found in encoded data, and text mining enables their retrieval and quantification. The U.S. Department of Veterans Affairs (VA) clinical data repository affords an opportunity to apply text-mining methodology to study clinical questions in large populations. To assess the feasibility of text mining, investigation of the relationship between exposure to adverse childhood experiences (ACEs) and recorded diagnoses was conducted among all VA-treated Gulf war veterans, utilizing all progress notes recorded from 2000-2011. Text processing extracted ACE exposures recorded among 44.7 million clinical notes belonging to 243,973 veterans. The relationship of ACE exposure to adult illnesses was analyzed using logistic regression. Bias considerations were assessed. ACE score was strongly associated with suicide attempts and serious mental disorders (ORs = 1.84 to 1.97), and less so with behaviorally mediated and somatic conditions (ORs = 1.02 to 1.36) per unit. Bias adjustments did not remove persistent associations between ACE score and most illnesses. Text mining to detect ACE exposure in a large population was feasible. Analysis of the relationship between ACE score and adult health conditions yielded patterns of association consistent with prior research.

AB - Free text in electronic health records resists large-scale analysis. Text records facts of interest not found in encoded data, and text mining enables their retrieval and quantification. The U.S. Department of Veterans Affairs (VA) clinical data repository affords an opportunity to apply text-mining methodology to study clinical questions in large populations. To assess the feasibility of text mining, investigation of the relationship between exposure to adverse childhood experiences (ACEs) and recorded diagnoses was conducted among all VA-treated Gulf war veterans, utilizing all progress notes recorded from 2000-2011. Text processing extracted ACE exposures recorded among 44.7 million clinical notes belonging to 243,973 veterans. The relationship of ACE exposure to adult illnesses was analyzed using logistic regression. Bias considerations were assessed. ACE score was strongly associated with suicide attempts and serious mental disorders (ORs = 1.84 to 1.97), and less so with behaviorally mediated and somatic conditions (ORs = 1.02 to 1.36) per unit. Bias adjustments did not remove persistent associations between ACE score and most illnesses. Text mining to detect ACE exposure in a large population was feasible. Analysis of the relationship between ACE score and adult health conditions yielded patterns of association consistent with prior research.

UR - http://www.scopus.com/inward/record.url?scp=84949317161&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949317161&partnerID=8YFLogxK

U2 - 10.1002/jts.22058

DO - 10.1002/jts.22058

M3 - Article

C2 - 26579624

AN - SCOPUS:84949317161

VL - 28

SP - 505

EP - 514

JO - Journal of Traumatic Stress

JF - Journal of Traumatic Stress

SN - 0894-9867

IS - 6

ER -