A probabilistic model for reducing medication errors: A sensitivity analysis using Electronic Health Records data

Chu Ya Huang, Phung Anh Nguyen, Hsuan Chia Yang, Md Mohaimenul Islam, Chia Wei Liang, Fei Peng Lee, Yu Chuan (Jack) Li

研究成果: 雜誌貢獻文章

摘要

Objectives: Medication-related clinical decision support systems have already been considered as a sophisticated method to improve healthcare quality, however, its importance has not been fully recognized. This paper's aim was to validate an existing probabilistic model that can automatically identify medication errors by performing a sensitivity analysis from electronic medical record data. Methods: We first built a knowledge base that consisted of 2.22 million disease-medication (DM) and 0.78 million medication-medication (MM) associations using Taiwan Health and Welfare data science claims data between January 1st, 2009 and December 31st, 2011. Further, we collected 0.6 million outpatient visit prescriptions from six departments across five different medical centers/hospitals. Afterward, we employed the data to our AESOP model and validated it using a sensitivity analysis of 11 various thresholds (α = [0.5; 1.5]) that were used to identify positive DM and MM associations. We randomly selected 2400 randomly prescriptions and compared them to the gold standard of 18 physicians’ manual review for appropriateness. Results: One hundred twenty-one results of 2400 prescriptions with various thresholds were tested by the AESOP model. Validation against the gold standard showed a high accuracy (over 80%), sensitivity (80–96%), and positive predictive value (over 85%). The negative predictive values ranged from 45 to 75% across three departments, cardiology, neurology, and ophthalmology. Conclusion: We performed a sensitivity analysis and validated the AESOP model in different hospitals. Thus, picking the optimal threshold of the model depended on balancing false negatives with false positives and depending on the specialty and the purpose of the system.
原文英語
頁(從 - 到)31-38
頁數8
期刊Computer Methods and Programs in Biomedicine
170
DOIs
出版狀態已發佈 - 三月 1 2019

指紋

Medication Errors
Electronic Health Records
Statistical Models
Sensitivity analysis
Prescriptions
Health
Regional Health Planning
Clinical Decision Support Systems
Knowledge Bases
Quality of Health Care
Ophthalmology
Neurology
Cardiology
Taiwan
Electronic medical equipment
Outpatients
Decision support systems
Physicians

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Health Informatics

引用此文

A probabilistic model for reducing medication errors : A sensitivity analysis using Electronic Health Records data. / Huang, Chu Ya; Nguyen, Phung Anh; Yang, Hsuan Chia; Islam, Md Mohaimenul; Liang, Chia Wei; Lee, Fei Peng; (Jack) Li, Yu Chuan.

於: Computer Methods and Programs in Biomedicine, 卷 170, 01.03.2019, p. 31-38.

研究成果: 雜誌貢獻文章

@article{91a1940ed39a4191a27c70a2b81f8739,
title = "A probabilistic model for reducing medication errors: A sensitivity analysis using Electronic Health Records data",
abstract = "Objectives: Medication-related clinical decision support systems have already been considered as a sophisticated method to improve healthcare quality, however, its importance has not been fully recognized. This paper's aim was to validate an existing probabilistic model that can automatically identify medication errors by performing a sensitivity analysis from electronic medical record data. Methods: We first built a knowledge base that consisted of 2.22 million disease-medication (DM) and 0.78 million medication-medication (MM) associations using Taiwan Health and Welfare data science claims data between January 1st, 2009 and December 31st, 2011. Further, we collected 0.6 million outpatient visit prescriptions from six departments across five different medical centers/hospitals. Afterward, we employed the data to our AESOP model and validated it using a sensitivity analysis of 11 various thresholds (α = [0.5; 1.5]) that were used to identify positive DM and MM associations. We randomly selected 2400 randomly prescriptions and compared them to the gold standard of 18 physicians’ manual review for appropriateness. Results: One hundred twenty-one results of 2400 prescriptions with various thresholds were tested by the AESOP model. Validation against the gold standard showed a high accuracy (over 80{\%}), sensitivity (80–96{\%}), and positive predictive value (over 85{\%}). The negative predictive values ranged from 45 to 75{\%} across three departments, cardiology, neurology, and ophthalmology. Conclusion: We performed a sensitivity analysis and validated the AESOP model in different hospitals. Thus, picking the optimal threshold of the model depended on balancing false negatives with false positives and depending on the specialty and the purpose of the system.",
keywords = "AESOP, EHR, Medication errors, Probabilistic model, Sensitivity analysis",
author = "Huang, {Chu Ya} and Nguyen, {Phung Anh} and Yang, {Hsuan Chia} and Islam, {Md Mohaimenul} and Liang, {Chia Wei} and Lee, {Fei Peng} and {(Jack) Li}, {Yu Chuan}",
year = "2019",
month = "3",
day = "1",
doi = "10.1016/j.cmpb.2018.12.033",
language = "English",
volume = "170",
pages = "31--38",
journal = "Computer Methods and Programs in Biomedicine",
issn = "0169-2607",
publisher = "Elsevier Ireland Ltd",

}

TY - JOUR

T1 - A probabilistic model for reducing medication errors

T2 - A sensitivity analysis using Electronic Health Records data

AU - Huang, Chu Ya

AU - Nguyen, Phung Anh

AU - Yang, Hsuan Chia

AU - Islam, Md Mohaimenul

AU - Liang, Chia Wei

AU - Lee, Fei Peng

AU - (Jack) Li, Yu Chuan

PY - 2019/3/1

Y1 - 2019/3/1

N2 - Objectives: Medication-related clinical decision support systems have already been considered as a sophisticated method to improve healthcare quality, however, its importance has not been fully recognized. This paper's aim was to validate an existing probabilistic model that can automatically identify medication errors by performing a sensitivity analysis from electronic medical record data. Methods: We first built a knowledge base that consisted of 2.22 million disease-medication (DM) and 0.78 million medication-medication (MM) associations using Taiwan Health and Welfare data science claims data between January 1st, 2009 and December 31st, 2011. Further, we collected 0.6 million outpatient visit prescriptions from six departments across five different medical centers/hospitals. Afterward, we employed the data to our AESOP model and validated it using a sensitivity analysis of 11 various thresholds (α = [0.5; 1.5]) that were used to identify positive DM and MM associations. We randomly selected 2400 randomly prescriptions and compared them to the gold standard of 18 physicians’ manual review for appropriateness. Results: One hundred twenty-one results of 2400 prescriptions with various thresholds were tested by the AESOP model. Validation against the gold standard showed a high accuracy (over 80%), sensitivity (80–96%), and positive predictive value (over 85%). The negative predictive values ranged from 45 to 75% across three departments, cardiology, neurology, and ophthalmology. Conclusion: We performed a sensitivity analysis and validated the AESOP model in different hospitals. Thus, picking the optimal threshold of the model depended on balancing false negatives with false positives and depending on the specialty and the purpose of the system.

AB - Objectives: Medication-related clinical decision support systems have already been considered as a sophisticated method to improve healthcare quality, however, its importance has not been fully recognized. This paper's aim was to validate an existing probabilistic model that can automatically identify medication errors by performing a sensitivity analysis from electronic medical record data. Methods: We first built a knowledge base that consisted of 2.22 million disease-medication (DM) and 0.78 million medication-medication (MM) associations using Taiwan Health and Welfare data science claims data between January 1st, 2009 and December 31st, 2011. Further, we collected 0.6 million outpatient visit prescriptions from six departments across five different medical centers/hospitals. Afterward, we employed the data to our AESOP model and validated it using a sensitivity analysis of 11 various thresholds (α = [0.5; 1.5]) that were used to identify positive DM and MM associations. We randomly selected 2400 randomly prescriptions and compared them to the gold standard of 18 physicians’ manual review for appropriateness. Results: One hundred twenty-one results of 2400 prescriptions with various thresholds were tested by the AESOP model. Validation against the gold standard showed a high accuracy (over 80%), sensitivity (80–96%), and positive predictive value (over 85%). The negative predictive values ranged from 45 to 75% across three departments, cardiology, neurology, and ophthalmology. Conclusion: We performed a sensitivity analysis and validated the AESOP model in different hospitals. Thus, picking the optimal threshold of the model depended on balancing false negatives with false positives and depending on the specialty and the purpose of the system.

KW - AESOP

KW - EHR

KW - Medication errors

KW - Probabilistic model

KW - Sensitivity analysis

UR - http://www.scopus.com/inward/record.url?scp=85059938813&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059938813&partnerID=8YFLogxK

U2 - 10.1016/j.cmpb.2018.12.033

DO - 10.1016/j.cmpb.2018.12.033

M3 - Article

AN - SCOPUS:85059938813

VL - 170

SP - 31

EP - 38

JO - Computer Methods and Programs in Biomedicine

JF - Computer Methods and Programs in Biomedicine

SN - 0169-2607

ER -