Automating assessment of lifestyle counseling in electronic health records

Brian L. Hazlehurst, Jean M. Lawrence, William T. Donahoo, Nancy E. Sherwood, Stephen E. Kurtz, Stan Xu, John F. Steiner

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Background Numerous population-based surveys indicate that overweight and obese patients can benefit from lifestyle counseling during routine clinical care. Purpose To determine if natural language processing (NLP) could be applied to information in the electronic health record (EHR) to automatically assess delivery of weight management-related counseling in clinical healthcare encounters. Methods The MediClass system with NLP capabilities was used to identify weight-management counseling in EHRs. Knowledge for the NLP application was derived from the 5As framework for behavior counseling: Ask (evaluate weight and related disease), Advise at-risk patients to lose weight, Assess patients' readiness to change behavior, Assist through discussion of weight-loss methods and programs, and Arrange follow-up efforts including referral. Using samples of EHR data between January 1, 2007, and March 31, 2011, from two health systems, the accuracy of the MediClass processor for identifying these counseling elements was evaluated in postpartum visits of 600 women with gestational diabetes mellitus (GDM) compared to manual chart review as the gold standard. Data were analyzed in 2013. Results Mean sensitivity and specificity for each of the 5As compared to the gold standard was at or above 85%, with the exception of sensitivity for Assist, which was 40% and 60% for each of the two health systems. The automated method identified many valid Assist cases not identified in the gold standard. Conclusions The MediClass processor has performance capability sufficiently similar to human abstractors to permit automated assessment of counseling for weight loss in postpartum encounter records.

Original languageEnglish (US)
Pages (from-to)457-464
Number of pages8
JournalAmerican Journal of Preventive Medicine
Volume46
Issue number5
DOIs
StatePublished - 2014
Externally publishedYes

Fingerprint

Electronic Health Records
Life Style
Counseling
Natural Language Processing
Weights and Measures
Postpartum Period
Weight Reduction Programs
Gestational Diabetes
Health
Weight Loss
Referral and Consultation
Delivery of Health Care
Sensitivity and Specificity
Population

ASJC Scopus subject areas

  • Public Health, Environmental and Occupational Health
  • Epidemiology
  • Medicine(all)

Cite this

Hazlehurst, B. L., Lawrence, J. M., Donahoo, W. T., Sherwood, N. E., Kurtz, S. E., Xu, S., & Steiner, J. F. (2014). Automating assessment of lifestyle counseling in electronic health records. American Journal of Preventive Medicine, 46(5), 457-464. https://doi.org/10.1016/j.amepre.2014.01.001

Automating assessment of lifestyle counseling in electronic health records. / Hazlehurst, Brian L.; Lawrence, Jean M.; Donahoo, William T.; Sherwood, Nancy E.; Kurtz, Stephen E.; Xu, Stan; Steiner, John F.

In: American Journal of Preventive Medicine, Vol. 46, No. 5, 2014, p. 457-464.

Research output: Contribution to journalArticle

Hazlehurst, BL, Lawrence, JM, Donahoo, WT, Sherwood, NE, Kurtz, SE, Xu, S & Steiner, JF 2014, 'Automating assessment of lifestyle counseling in electronic health records', American Journal of Preventive Medicine, vol. 46, no. 5, pp. 457-464. https://doi.org/10.1016/j.amepre.2014.01.001
Hazlehurst, Brian L. ; Lawrence, Jean M. ; Donahoo, William T. ; Sherwood, Nancy E. ; Kurtz, Stephen E. ; Xu, Stan ; Steiner, John F. / Automating assessment of lifestyle counseling in electronic health records. In: American Journal of Preventive Medicine. 2014 ; Vol. 46, No. 5. pp. 457-464.
@article{acef11c874334420a6e5de997866dbea,
title = "Automating assessment of lifestyle counseling in electronic health records",
abstract = "Background Numerous population-based surveys indicate that overweight and obese patients can benefit from lifestyle counseling during routine clinical care. Purpose To determine if natural language processing (NLP) could be applied to information in the electronic health record (EHR) to automatically assess delivery of weight management-related counseling in clinical healthcare encounters. Methods The MediClass system with NLP capabilities was used to identify weight-management counseling in EHRs. Knowledge for the NLP application was derived from the 5As framework for behavior counseling: Ask (evaluate weight and related disease), Advise at-risk patients to lose weight, Assess patients' readiness to change behavior, Assist through discussion of weight-loss methods and programs, and Arrange follow-up efforts including referral. Using samples of EHR data between January 1, 2007, and March 31, 2011, from two health systems, the accuracy of the MediClass processor for identifying these counseling elements was evaluated in postpartum visits of 600 women with gestational diabetes mellitus (GDM) compared to manual chart review as the gold standard. Data were analyzed in 2013. Results Mean sensitivity and specificity for each of the 5As compared to the gold standard was at or above 85{\%}, with the exception of sensitivity for Assist, which was 40{\%} and 60{\%} for each of the two health systems. The automated method identified many valid Assist cases not identified in the gold standard. Conclusions The MediClass processor has performance capability sufficiently similar to human abstractors to permit automated assessment of counseling for weight loss in postpartum encounter records.",
author = "Hazlehurst, {Brian L.} and Lawrence, {Jean M.} and Donahoo, {William T.} and Sherwood, {Nancy E.} and Kurtz, {Stephen E.} and Stan Xu and Steiner, {John F.}",
year = "2014",
doi = "10.1016/j.amepre.2014.01.001",
language = "English (US)",
volume = "46",
pages = "457--464",
journal = "American Journal of Preventive Medicine",
issn = "0749-3797",
publisher = "Elsevier Inc.",
number = "5",

}

TY - JOUR

T1 - Automating assessment of lifestyle counseling in electronic health records

AU - Hazlehurst, Brian L.

AU - Lawrence, Jean M.

AU - Donahoo, William T.

AU - Sherwood, Nancy E.

AU - Kurtz, Stephen E.

AU - Xu, Stan

AU - Steiner, John F.

PY - 2014

Y1 - 2014

N2 - Background Numerous population-based surveys indicate that overweight and obese patients can benefit from lifestyle counseling during routine clinical care. Purpose To determine if natural language processing (NLP) could be applied to information in the electronic health record (EHR) to automatically assess delivery of weight management-related counseling in clinical healthcare encounters. Methods The MediClass system with NLP capabilities was used to identify weight-management counseling in EHRs. Knowledge for the NLP application was derived from the 5As framework for behavior counseling: Ask (evaluate weight and related disease), Advise at-risk patients to lose weight, Assess patients' readiness to change behavior, Assist through discussion of weight-loss methods and programs, and Arrange follow-up efforts including referral. Using samples of EHR data between January 1, 2007, and March 31, 2011, from two health systems, the accuracy of the MediClass processor for identifying these counseling elements was evaluated in postpartum visits of 600 women with gestational diabetes mellitus (GDM) compared to manual chart review as the gold standard. Data were analyzed in 2013. Results Mean sensitivity and specificity for each of the 5As compared to the gold standard was at or above 85%, with the exception of sensitivity for Assist, which was 40% and 60% for each of the two health systems. The automated method identified many valid Assist cases not identified in the gold standard. Conclusions The MediClass processor has performance capability sufficiently similar to human abstractors to permit automated assessment of counseling for weight loss in postpartum encounter records.

AB - Background Numerous population-based surveys indicate that overweight and obese patients can benefit from lifestyle counseling during routine clinical care. Purpose To determine if natural language processing (NLP) could be applied to information in the electronic health record (EHR) to automatically assess delivery of weight management-related counseling in clinical healthcare encounters. Methods The MediClass system with NLP capabilities was used to identify weight-management counseling in EHRs. Knowledge for the NLP application was derived from the 5As framework for behavior counseling: Ask (evaluate weight and related disease), Advise at-risk patients to lose weight, Assess patients' readiness to change behavior, Assist through discussion of weight-loss methods and programs, and Arrange follow-up efforts including referral. Using samples of EHR data between January 1, 2007, and March 31, 2011, from two health systems, the accuracy of the MediClass processor for identifying these counseling elements was evaluated in postpartum visits of 600 women with gestational diabetes mellitus (GDM) compared to manual chart review as the gold standard. Data were analyzed in 2013. Results Mean sensitivity and specificity for each of the 5As compared to the gold standard was at or above 85%, with the exception of sensitivity for Assist, which was 40% and 60% for each of the two health systems. The automated method identified many valid Assist cases not identified in the gold standard. Conclusions The MediClass processor has performance capability sufficiently similar to human abstractors to permit automated assessment of counseling for weight loss in postpartum encounter records.

UR - http://www.scopus.com/inward/record.url?scp=84899023520&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899023520&partnerID=8YFLogxK

U2 - 10.1016/j.amepre.2014.01.001

DO - 10.1016/j.amepre.2014.01.001

M3 - Article

C2 - 24745635

AN - SCOPUS:84899023520

VL - 46

SP - 457

EP - 464

JO - American Journal of Preventive Medicine

JF - American Journal of Preventive Medicine

SN - 0749-3797

IS - 5

ER -