Automated chart review for asthma cohort identification using natural language processing: An exploratory study

Stephen T. Wu, Sunghwan Sohn, K. E. Ravikumar, Kavishwar Wagholikar, Siddhartha R. Jonnalagadda, Hongfang D Liu, Young J Juhn

Research output: Contribution to journalArticle

27 Citations (Scopus)

Abstract

Background A significant proportion of children with asthma have delayed diagnosis of asthma by health care providers. Manual chart review according to established criteria is more accurate than directly using diagnosis codes, which tend to under-identify asthmatics, but chart reviews are more costly and less timely. Objective To evaluate the accuracy of a computational approach to asthma ascertainment, characterizing its utility and feasibility toward large-scale deployment in electronic medical records. Methods A natural language processing (NLP) system was developed for extracting predetermined criteria for asthma from unstructured text in electronic medical records and then inferring asthma status based on these criteria. Using manual chart reviews as a gold standard, asthma status (yes vs no) and identification date (first date of a "yes" asthma status) were determined by the NLP system. Results Patients were a group of children (n = 112, 84% Caucasian, 49% girls) younger than 4 years (mean 2.0 years, standard deviation 1.03 years) who participated in previous studies. The NLP approach to asthma ascertainment showed sensitivity, specificity, positive predictive value, negative predictive value, and median delay in diagnosis of 84.6%, 96.5%, 88.0%, 95.4%, and 0 months, respectively; this compared favorably with diagnosis codes, at 30.8%, 93.2%, 57.1%, 82.2%, and 2.3 months, respectively. Conclusion Automated asthma ascertainment from electronic medical records using NLP is feasible and more accurate than traditional approaches such as diagnosis codes. Considering the difficulty of labor-intensive manual record review, NLP approaches for asthma ascertainment should be considered for improving clinical care and research, especially in large-scale efforts.

Original languageEnglish (US)
Pages (from-to)364-369
Number of pages6
JournalAnnals of Allergy, Asthma and Immunology
Volume111
Issue number5
DOIs
StatePublished - Nov 2013

Fingerprint

Natural Language Processing
Asthma
Electronic Health Records
Delayed Diagnosis
Health Personnel

ASJC Scopus subject areas

  • Immunology and Allergy
  • Pulmonary and Respiratory Medicine

Cite this

Automated chart review for asthma cohort identification using natural language processing : An exploratory study. / Wu, Stephen T.; Sohn, Sunghwan; Ravikumar, K. E.; Wagholikar, Kavishwar; Jonnalagadda, Siddhartha R.; Liu, Hongfang D; Juhn, Young J.

In: Annals of Allergy, Asthma and Immunology, Vol. 111, No. 5, 11.2013, p. 364-369.

Research output: Contribution to journalArticle

@article{9f431bdafe0b41ed87073df0c43960b8,
title = "Automated chart review for asthma cohort identification using natural language processing: An exploratory study",
abstract = "Background A significant proportion of children with asthma have delayed diagnosis of asthma by health care providers. Manual chart review according to established criteria is more accurate than directly using diagnosis codes, which tend to under-identify asthmatics, but chart reviews are more costly and less timely. Objective To evaluate the accuracy of a computational approach to asthma ascertainment, characterizing its utility and feasibility toward large-scale deployment in electronic medical records. Methods A natural language processing (NLP) system was developed for extracting predetermined criteria for asthma from unstructured text in electronic medical records and then inferring asthma status based on these criteria. Using manual chart reviews as a gold standard, asthma status (yes vs no) and identification date (first date of a {"}yes{"} asthma status) were determined by the NLP system. Results Patients were a group of children (n = 112, 84{\%} Caucasian, 49{\%} girls) younger than 4 years (mean 2.0 years, standard deviation 1.03 years) who participated in previous studies. The NLP approach to asthma ascertainment showed sensitivity, specificity, positive predictive value, negative predictive value, and median delay in diagnosis of 84.6{\%}, 96.5{\%}, 88.0{\%}, 95.4{\%}, and 0 months, respectively; this compared favorably with diagnosis codes, at 30.8{\%}, 93.2{\%}, 57.1{\%}, 82.2{\%}, and 2.3 months, respectively. Conclusion Automated asthma ascertainment from electronic medical records using NLP is feasible and more accurate than traditional approaches such as diagnosis codes. Considering the difficulty of labor-intensive manual record review, NLP approaches for asthma ascertainment should be considered for improving clinical care and research, especially in large-scale efforts.",
author = "Wu, {Stephen T.} and Sunghwan Sohn and Ravikumar, {K. E.} and Kavishwar Wagholikar and Jonnalagadda, {Siddhartha R.} and Liu, {Hongfang D} and Juhn, {Young J}",
year = "2013",
month = "11",
doi = "10.1016/j.anai.2013.07.022",
language = "English (US)",
volume = "111",
pages = "364--369",
journal = "Annals of Allergy, Asthma and Immunology",
issn = "1081-1206",
publisher = "American College of Allergy, Asthma and Immunology",
number = "5",

}

TY - JOUR

T1 - Automated chart review for asthma cohort identification using natural language processing

T2 - An exploratory study

AU - Wu, Stephen T.

AU - Sohn, Sunghwan

AU - Ravikumar, K. E.

AU - Wagholikar, Kavishwar

AU - Jonnalagadda, Siddhartha R.

AU - Liu, Hongfang D

AU - Juhn, Young J

PY - 2013/11

Y1 - 2013/11

N2 - Background A significant proportion of children with asthma have delayed diagnosis of asthma by health care providers. Manual chart review according to established criteria is more accurate than directly using diagnosis codes, which tend to under-identify asthmatics, but chart reviews are more costly and less timely. Objective To evaluate the accuracy of a computational approach to asthma ascertainment, characterizing its utility and feasibility toward large-scale deployment in electronic medical records. Methods A natural language processing (NLP) system was developed for extracting predetermined criteria for asthma from unstructured text in electronic medical records and then inferring asthma status based on these criteria. Using manual chart reviews as a gold standard, asthma status (yes vs no) and identification date (first date of a "yes" asthma status) were determined by the NLP system. Results Patients were a group of children (n = 112, 84% Caucasian, 49% girls) younger than 4 years (mean 2.0 years, standard deviation 1.03 years) who participated in previous studies. The NLP approach to asthma ascertainment showed sensitivity, specificity, positive predictive value, negative predictive value, and median delay in diagnosis of 84.6%, 96.5%, 88.0%, 95.4%, and 0 months, respectively; this compared favorably with diagnosis codes, at 30.8%, 93.2%, 57.1%, 82.2%, and 2.3 months, respectively. Conclusion Automated asthma ascertainment from electronic medical records using NLP is feasible and more accurate than traditional approaches such as diagnosis codes. Considering the difficulty of labor-intensive manual record review, NLP approaches for asthma ascertainment should be considered for improving clinical care and research, especially in large-scale efforts.

AB - Background A significant proportion of children with asthma have delayed diagnosis of asthma by health care providers. Manual chart review according to established criteria is more accurate than directly using diagnosis codes, which tend to under-identify asthmatics, but chart reviews are more costly and less timely. Objective To evaluate the accuracy of a computational approach to asthma ascertainment, characterizing its utility and feasibility toward large-scale deployment in electronic medical records. Methods A natural language processing (NLP) system was developed for extracting predetermined criteria for asthma from unstructured text in electronic medical records and then inferring asthma status based on these criteria. Using manual chart reviews as a gold standard, asthma status (yes vs no) and identification date (first date of a "yes" asthma status) were determined by the NLP system. Results Patients were a group of children (n = 112, 84% Caucasian, 49% girls) younger than 4 years (mean 2.0 years, standard deviation 1.03 years) who participated in previous studies. The NLP approach to asthma ascertainment showed sensitivity, specificity, positive predictive value, negative predictive value, and median delay in diagnosis of 84.6%, 96.5%, 88.0%, 95.4%, and 0 months, respectively; this compared favorably with diagnosis codes, at 30.8%, 93.2%, 57.1%, 82.2%, and 2.3 months, respectively. Conclusion Automated asthma ascertainment from electronic medical records using NLP is feasible and more accurate than traditional approaches such as diagnosis codes. Considering the difficulty of labor-intensive manual record review, NLP approaches for asthma ascertainment should be considered for improving clinical care and research, especially in large-scale efforts.

UR - http://www.scopus.com/inward/record.url?scp=84885960018&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885960018&partnerID=8YFLogxK

U2 - 10.1016/j.anai.2013.07.022

DO - 10.1016/j.anai.2013.07.022

M3 - Article

C2 - 24125142

AN - SCOPUS:84885960018

VL - 111

SP - 364

EP - 369

JO - Annals of Allergy, Asthma and Immunology

JF - Annals of Allergy, Asthma and Immunology

SN - 1081-1206

IS - 5

ER -