Automated chart review utilizing natural language processing algorithm for asthma predictive index

Harsheen Kaur, Sunghwan Sohn, Chung Il Wi, Euijung Ryu, Miguel Park, Kay Bachman, Hirohito Kita, Ivana T Croghan, Jose A. Castro-Rodriguez, Gretchen A. Voge, Hongfang D Liu, Young J Juhn

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Background: Thus far, no algorithms have been developed to automatically extract patients who meet Asthma Predictive Index (API) criteria from the Electronic health records (EHR) yet. Our objective is to develop and validate a natural language processing (NLP) algorithm to identify patients that meet API criteria. Methods: This is a cross-sectional study nested in a birth cohort study in Olmsted County, MN. Asthma status ascertained by manual chart review based on API criteria served as gold standard. NLP-API was developed on a training cohort (n=87) and validated on a test cohort (n=427). Criterion validity was measured by sensitivity, specificity, positive predictive value and negative predictive value of the NLP algorithm against manual chart review for asthma status. Construct validity was determined by associations of asthma status defined by NLP-API with known risk factors for asthma. Results: Among the eligible 427 subjects of the test cohort, 48% were males and 74% were White. Median age was 5.3years (interquartile range 3.6-6.8). 35 (8%) had a history of asthma by NLP-API vs. 36 (8%) by abstractor with 31 by both approaches. NLP-API predicted asthma status with sensitivity 86%, specificity 98%, positive predictive value 88%, negative predictive value 98%. Asthma status by both NLP and manual chart review were significantly associated with the known asthma risk factors, such as history of allergic rhinitis, eczema, family history of asthma, and maternal history of smoking during pregnancy (p value <0.05). Maternal smoking [odds ratio: 4.4, 95% confidence interval 1.8-10.7] was associated with asthma status determined by NLP-API and abstractor, and the effect sizes were similar between the reviews with 4.4 vs 4.2 respectively. Conclusion: NLP-API was able to ascertain asthma status in children mining from EHR and has a potential to enhance asthma care and research through population management and large-scale studies when identifying children who meet API criteria.

Original languageEnglish (US)
Article number34
JournalBMC Pulmonary Medicine
Volume18
Issue number1
DOIs
StatePublished - Feb 13 2018

Fingerprint

Natural Language Processing
Asthma
Electronic Health Records

Keywords

  • API
  • Asthma
  • Epidemiology
  • Informatics
  • NLP

ASJC Scopus subject areas

  • Pulmonary and Respiratory Medicine

Cite this

Automated chart review utilizing natural language processing algorithm for asthma predictive index. / Kaur, Harsheen; Sohn, Sunghwan; Wi, Chung Il; Ryu, Euijung; Park, Miguel; Bachman, Kay; Kita, Hirohito; Croghan, Ivana T; Castro-Rodriguez, Jose A.; Voge, Gretchen A.; Liu, Hongfang D; Juhn, Young J.

In: BMC Pulmonary Medicine, Vol. 18, No. 1, 34, 13.02.2018.

Research output: Contribution to journalArticle

@article{496482797db240b692a8694466c87bb9,
title = "Automated chart review utilizing natural language processing algorithm for asthma predictive index",
abstract = "Background: Thus far, no algorithms have been developed to automatically extract patients who meet Asthma Predictive Index (API) criteria from the Electronic health records (EHR) yet. Our objective is to develop and validate a natural language processing (NLP) algorithm to identify patients that meet API criteria. Methods: This is a cross-sectional study nested in a birth cohort study in Olmsted County, MN. Asthma status ascertained by manual chart review based on API criteria served as gold standard. NLP-API was developed on a training cohort (n=87) and validated on a test cohort (n=427). Criterion validity was measured by sensitivity, specificity, positive predictive value and negative predictive value of the NLP algorithm against manual chart review for asthma status. Construct validity was determined by associations of asthma status defined by NLP-API with known risk factors for asthma. Results: Among the eligible 427 subjects of the test cohort, 48{\%} were males and 74{\%} were White. Median age was 5.3years (interquartile range 3.6-6.8). 35 (8{\%}) had a history of asthma by NLP-API vs. 36 (8{\%}) by abstractor with 31 by both approaches. NLP-API predicted asthma status with sensitivity 86{\%}, specificity 98{\%}, positive predictive value 88{\%}, negative predictive value 98{\%}. Asthma status by both NLP and manual chart review were significantly associated with the known asthma risk factors, such as history of allergic rhinitis, eczema, family history of asthma, and maternal history of smoking during pregnancy (p value <0.05). Maternal smoking [odds ratio: 4.4, 95{\%} confidence interval 1.8-10.7] was associated with asthma status determined by NLP-API and abstractor, and the effect sizes were similar between the reviews with 4.4 vs 4.2 respectively. Conclusion: NLP-API was able to ascertain asthma status in children mining from EHR and has a potential to enhance asthma care and research through population management and large-scale studies when identifying children who meet API criteria.",
keywords = "API, Asthma, Epidemiology, Informatics, NLP",
author = "Harsheen Kaur and Sunghwan Sohn and Wi, {Chung Il} and Euijung Ryu and Miguel Park and Kay Bachman and Hirohito Kita and Croghan, {Ivana T} and Castro-Rodriguez, {Jose A.} and Voge, {Gretchen A.} and Liu, {Hongfang D} and Juhn, {Young J}",
year = "2018",
month = "2",
day = "13",
doi = "10.1186/s12890-018-0593-9",
language = "English (US)",
volume = "18",
journal = "BMC Pulmonary Medicine",
issn = "1471-2466",
publisher = "BioMed Central",
number = "1",

}

TY - JOUR

T1 - Automated chart review utilizing natural language processing algorithm for asthma predictive index

AU - Kaur, Harsheen

AU - Sohn, Sunghwan

AU - Wi, Chung Il

AU - Ryu, Euijung

AU - Park, Miguel

AU - Bachman, Kay

AU - Kita, Hirohito

AU - Croghan, Ivana T

AU - Castro-Rodriguez, Jose A.

AU - Voge, Gretchen A.

AU - Liu, Hongfang D

AU - Juhn, Young J

PY - 2018/2/13

Y1 - 2018/2/13

N2 - Background: Thus far, no algorithms have been developed to automatically extract patients who meet Asthma Predictive Index (API) criteria from the Electronic health records (EHR) yet. Our objective is to develop and validate a natural language processing (NLP) algorithm to identify patients that meet API criteria. Methods: This is a cross-sectional study nested in a birth cohort study in Olmsted County, MN. Asthma status ascertained by manual chart review based on API criteria served as gold standard. NLP-API was developed on a training cohort (n=87) and validated on a test cohort (n=427). Criterion validity was measured by sensitivity, specificity, positive predictive value and negative predictive value of the NLP algorithm against manual chart review for asthma status. Construct validity was determined by associations of asthma status defined by NLP-API with known risk factors for asthma. Results: Among the eligible 427 subjects of the test cohort, 48% were males and 74% were White. Median age was 5.3years (interquartile range 3.6-6.8). 35 (8%) had a history of asthma by NLP-API vs. 36 (8%) by abstractor with 31 by both approaches. NLP-API predicted asthma status with sensitivity 86%, specificity 98%, positive predictive value 88%, negative predictive value 98%. Asthma status by both NLP and manual chart review were significantly associated with the known asthma risk factors, such as history of allergic rhinitis, eczema, family history of asthma, and maternal history of smoking during pregnancy (p value <0.05). Maternal smoking [odds ratio: 4.4, 95% confidence interval 1.8-10.7] was associated with asthma status determined by NLP-API and abstractor, and the effect sizes were similar between the reviews with 4.4 vs 4.2 respectively. Conclusion: NLP-API was able to ascertain asthma status in children mining from EHR and has a potential to enhance asthma care and research through population management and large-scale studies when identifying children who meet API criteria.

AB - Background: Thus far, no algorithms have been developed to automatically extract patients who meet Asthma Predictive Index (API) criteria from the Electronic health records (EHR) yet. Our objective is to develop and validate a natural language processing (NLP) algorithm to identify patients that meet API criteria. Methods: This is a cross-sectional study nested in a birth cohort study in Olmsted County, MN. Asthma status ascertained by manual chart review based on API criteria served as gold standard. NLP-API was developed on a training cohort (n=87) and validated on a test cohort (n=427). Criterion validity was measured by sensitivity, specificity, positive predictive value and negative predictive value of the NLP algorithm against manual chart review for asthma status. Construct validity was determined by associations of asthma status defined by NLP-API with known risk factors for asthma. Results: Among the eligible 427 subjects of the test cohort, 48% were males and 74% were White. Median age was 5.3years (interquartile range 3.6-6.8). 35 (8%) had a history of asthma by NLP-API vs. 36 (8%) by abstractor with 31 by both approaches. NLP-API predicted asthma status with sensitivity 86%, specificity 98%, positive predictive value 88%, negative predictive value 98%. Asthma status by both NLP and manual chart review were significantly associated with the known asthma risk factors, such as history of allergic rhinitis, eczema, family history of asthma, and maternal history of smoking during pregnancy (p value <0.05). Maternal smoking [odds ratio: 4.4, 95% confidence interval 1.8-10.7] was associated with asthma status determined by NLP-API and abstractor, and the effect sizes were similar between the reviews with 4.4 vs 4.2 respectively. Conclusion: NLP-API was able to ascertain asthma status in children mining from EHR and has a potential to enhance asthma care and research through population management and large-scale studies when identifying children who meet API criteria.

KW - API

KW - Asthma

KW - Epidemiology

KW - Informatics

KW - NLP

UR - http://www.scopus.com/inward/record.url?scp=85042107573&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85042107573&partnerID=8YFLogxK

U2 - 10.1186/s12890-018-0593-9

DO - 10.1186/s12890-018-0593-9

M3 - Article

VL - 18

JO - BMC Pulmonary Medicine

JF - BMC Pulmonary Medicine

SN - 1471-2466

IS - 1

M1 - 34

ER -