Detecting pharmacovigilance signals combining electronic medical records with spontaneous reports: A case study of conventional disease-modifying antirheumatic drugs for rheumatoid arthritis

Liwei Wang, Majid Rastegar-Mojarad, Zhiliang Ji, Sijia Liu, Ke Liu, Sungrim Moon, Feichen Shen, Yanshan Wang, Lixia Yao, John Manley III Davis, Hongfang D Liu

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Multiple data sources are preferred in adverse drug event (ADEs) surveillance owing to inadequacies of single source. However, analytic methods to monitor potential ADEs after prolonged drug exposure are still lacking. In this study we propose a method aiming to screen potential ADEs by combining FDA Adverse Event Reporting System (FAERS) and Electronic Medical Record (EMR). The proposed method uses natural language processing (NLP) techniques to extract treatment outcome information captured in unstructured text and adopts case-crossover design in EMR. Performances were evaluated using two ADE knowledge bases: Adverse Drug Reaction Classification System (ADReCS) and SIDER. We tested our method in ADE signal detection of conventional disease-modifying antirheumatic drugs (DMARDs) in rheumatoid arthritis patients. Findings showed that recall greatly increased when combining FAERS with EMR compared with FAERS alone and EMR alone, especially for flexible mapping strategy. Precision (FAERS + EMR) in detecting ADEs improved using ADReCS as gold standard compared with SIDER. In addition, signals detected from EMR have considerably overlapped with signals detected from FAERS or ADE knowledge bases, implying the importance of EMR for pharmacovigilance. ADE signals detected from EMR and/or FAERS but not in existing knowledge bases provide hypothesis for future study.

Original languageEnglish (US)
Article number875
JournalFrontiers in Pharmacology
Volume9
Issue numberAUG
DOIs
StatePublished - Aug 7 2018

Fingerprint

Pharmacovigilance
Antirheumatic Agents
Electronic Health Records
Drug-Related Side Effects and Adverse Reactions
Rheumatoid Arthritis
Knowledge Bases
Natural Language Processing
Information Storage and Retrieval
Cross-Over Studies

Keywords

  • Adverse drug event
  • Disease-modifying antirheumatic drug (DMARD)
  • Electronic Medical Records (EMR)
  • FDA Adverse Event Reporting System (FAERS)
  • Natural language processing
  • Pharmacovigilance

ASJC Scopus subject areas

  • Pharmacology
  • Pharmacology (medical)

Cite this

Detecting pharmacovigilance signals combining electronic medical records with spontaneous reports : A case study of conventional disease-modifying antirheumatic drugs for rheumatoid arthritis. / Wang, Liwei; Rastegar-Mojarad, Majid; Ji, Zhiliang; Liu, Sijia; Liu, Ke; Moon, Sungrim; Shen, Feichen; Wang, Yanshan; Yao, Lixia; Davis, John Manley III; Liu, Hongfang D.

In: Frontiers in Pharmacology, Vol. 9, No. AUG, 875, 07.08.2018.

Research output: Contribution to journalArticle

@article{c406c28cf3a74087beebb0b9a8ac17fd,
title = "Detecting pharmacovigilance signals combining electronic medical records with spontaneous reports: A case study of conventional disease-modifying antirheumatic drugs for rheumatoid arthritis",
abstract = "Multiple data sources are preferred in adverse drug event (ADEs) surveillance owing to inadequacies of single source. However, analytic methods to monitor potential ADEs after prolonged drug exposure are still lacking. In this study we propose a method aiming to screen potential ADEs by combining FDA Adverse Event Reporting System (FAERS) and Electronic Medical Record (EMR). The proposed method uses natural language processing (NLP) techniques to extract treatment outcome information captured in unstructured text and adopts case-crossover design in EMR. Performances were evaluated using two ADE knowledge bases: Adverse Drug Reaction Classification System (ADReCS) and SIDER. We tested our method in ADE signal detection of conventional disease-modifying antirheumatic drugs (DMARDs) in rheumatoid arthritis patients. Findings showed that recall greatly increased when combining FAERS with EMR compared with FAERS alone and EMR alone, especially for flexible mapping strategy. Precision (FAERS + EMR) in detecting ADEs improved using ADReCS as gold standard compared with SIDER. In addition, signals detected from EMR have considerably overlapped with signals detected from FAERS or ADE knowledge bases, implying the importance of EMR for pharmacovigilance. ADE signals detected from EMR and/or FAERS but not in existing knowledge bases provide hypothesis for future study.",
keywords = "Adverse drug event, Disease-modifying antirheumatic drug (DMARD), Electronic Medical Records (EMR), FDA Adverse Event Reporting System (FAERS), Natural language processing, Pharmacovigilance",
author = "Liwei Wang and Majid Rastegar-Mojarad and Zhiliang Ji and Sijia Liu and Ke Liu and Sungrim Moon and Feichen Shen and Yanshan Wang and Lixia Yao and Davis, {John Manley III} and Liu, {Hongfang D}",
year = "2018",
month = "8",
day = "7",
doi = "10.3389/fphar.2018.00875",
language = "English (US)",
volume = "9",
journal = "Frontiers in Pharmacology",
issn = "1663-9812",
publisher = "Frontiers Media S. A.",
number = "AUG",

}

TY - JOUR

T1 - Detecting pharmacovigilance signals combining electronic medical records with spontaneous reports

T2 - A case study of conventional disease-modifying antirheumatic drugs for rheumatoid arthritis

AU - Wang, Liwei

AU - Rastegar-Mojarad, Majid

AU - Ji, Zhiliang

AU - Liu, Sijia

AU - Liu, Ke

AU - Moon, Sungrim

AU - Shen, Feichen

AU - Wang, Yanshan

AU - Yao, Lixia

AU - Davis, John Manley III

AU - Liu, Hongfang D

PY - 2018/8/7

Y1 - 2018/8/7

N2 - Multiple data sources are preferred in adverse drug event (ADEs) surveillance owing to inadequacies of single source. However, analytic methods to monitor potential ADEs after prolonged drug exposure are still lacking. In this study we propose a method aiming to screen potential ADEs by combining FDA Adverse Event Reporting System (FAERS) and Electronic Medical Record (EMR). The proposed method uses natural language processing (NLP) techniques to extract treatment outcome information captured in unstructured text and adopts case-crossover design in EMR. Performances were evaluated using two ADE knowledge bases: Adverse Drug Reaction Classification System (ADReCS) and SIDER. We tested our method in ADE signal detection of conventional disease-modifying antirheumatic drugs (DMARDs) in rheumatoid arthritis patients. Findings showed that recall greatly increased when combining FAERS with EMR compared with FAERS alone and EMR alone, especially for flexible mapping strategy. Precision (FAERS + EMR) in detecting ADEs improved using ADReCS as gold standard compared with SIDER. In addition, signals detected from EMR have considerably overlapped with signals detected from FAERS or ADE knowledge bases, implying the importance of EMR for pharmacovigilance. ADE signals detected from EMR and/or FAERS but not in existing knowledge bases provide hypothesis for future study.

AB - Multiple data sources are preferred in adverse drug event (ADEs) surveillance owing to inadequacies of single source. However, analytic methods to monitor potential ADEs after prolonged drug exposure are still lacking. In this study we propose a method aiming to screen potential ADEs by combining FDA Adverse Event Reporting System (FAERS) and Electronic Medical Record (EMR). The proposed method uses natural language processing (NLP) techniques to extract treatment outcome information captured in unstructured text and adopts case-crossover design in EMR. Performances were evaluated using two ADE knowledge bases: Adverse Drug Reaction Classification System (ADReCS) and SIDER. We tested our method in ADE signal detection of conventional disease-modifying antirheumatic drugs (DMARDs) in rheumatoid arthritis patients. Findings showed that recall greatly increased when combining FAERS with EMR compared with FAERS alone and EMR alone, especially for flexible mapping strategy. Precision (FAERS + EMR) in detecting ADEs improved using ADReCS as gold standard compared with SIDER. In addition, signals detected from EMR have considerably overlapped with signals detected from FAERS or ADE knowledge bases, implying the importance of EMR for pharmacovigilance. ADE signals detected from EMR and/or FAERS but not in existing knowledge bases provide hypothesis for future study.

KW - Adverse drug event

KW - Disease-modifying antirheumatic drug (DMARD)

KW - Electronic Medical Records (EMR)

KW - FDA Adverse Event Reporting System (FAERS)

KW - Natural language processing

KW - Pharmacovigilance

UR - http://www.scopus.com/inward/record.url?scp=85054929485&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054929485&partnerID=8YFLogxK

U2 - 10.3389/fphar.2018.00875

DO - 10.3389/fphar.2018.00875

M3 - Article

AN - SCOPUS:85054929485

VL - 9

JO - Frontiers in Pharmacology

JF - Frontiers in Pharmacology

SN - 1663-9812

IS - AUG

M1 - 875

ER -