Three-Component Mixture Model-Based Adverse Drug Event Signal Detection for the Adverse Event Reporting System

Pengyue Zhang, Meng Li, Chien Wei Chiang, Lei Wang, Yang Xiang, Lijun Cheng, Weixing Feng, Titus Schleyer, Sara Quinney, Heng Yi Wu, Donglin Zeng, Lang Li

Research output: Contribution to journalArticle

Abstract

The US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) is an important source for detecting adverse drug event (ADE) signals. In this article, we propose a three-component mixture model (3CMM) for FAERS signal detection. In 3CMM, a drug-ADE pair is assumed to have either a zero relative risk (RR), or a background RR (mean RR = 1), or an increased RR (mean RR >1). By clearly defining the second component (mean RR = 1) as the null distribution, 3CMM estimates local false discovery rates (FDRs) for ADE signals under the empirical Bayes framework. Compared with existing approaches, the local FDR's top signals have noninferior or better sensitivities to detect true signals in both FAERS analysis and simulation studies. Additionally, we identify that the top signals of different approaches have different patterns, and they are complementary to each other.

Original languageEnglish (US)
Pages (from-to)499-506
Number of pages8
JournalCPT: Pharmacometrics and Systems Pharmacology
Volume7
Issue number8
DOIs
StatePublished - Aug 1 2018

Fingerprint

Event Detection
Relative Risk
Signal Detection
Signal detection
Component Model
Drug-Related Side Effects and Adverse Reactions
Mixture Model
Drugs
Model-based
Empirical Bayes
Null Distribution
System Simulation
Systems Analysis
Systems analysis
Simulation Study
Food
Zero
Pharmaceutical Preparations
Estimate

ASJC Scopus subject areas

  • Modeling and Simulation
  • Pharmacology (medical)

Cite this

Three-Component Mixture Model-Based Adverse Drug Event Signal Detection for the Adverse Event Reporting System. / Zhang, Pengyue; Li, Meng; Chiang, Chien Wei; Wang, Lei; Xiang, Yang; Cheng, Lijun; Feng, Weixing; Schleyer, Titus; Quinney, Sara; Wu, Heng Yi; Zeng, Donglin; Li, Lang.

In: CPT: Pharmacometrics and Systems Pharmacology, Vol. 7, No. 8, 01.08.2018, p. 499-506.

Research output: Contribution to journalArticle

Zhang, Pengyue ; Li, Meng ; Chiang, Chien Wei ; Wang, Lei ; Xiang, Yang ; Cheng, Lijun ; Feng, Weixing ; Schleyer, Titus ; Quinney, Sara ; Wu, Heng Yi ; Zeng, Donglin ; Li, Lang. / Three-Component Mixture Model-Based Adverse Drug Event Signal Detection for the Adverse Event Reporting System. In: CPT: Pharmacometrics and Systems Pharmacology. 2018 ; Vol. 7, No. 8. pp. 499-506.
@article{75db7efc300746eaa01f6693bb62fe3a,
title = "Three-Component Mixture Model-Based Adverse Drug Event Signal Detection for the Adverse Event Reporting System",
abstract = "The US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) is an important source for detecting adverse drug event (ADE) signals. In this article, we propose a three-component mixture model (3CMM) for FAERS signal detection. In 3CMM, a drug-ADE pair is assumed to have either a zero relative risk (RR), or a background RR (mean RR = 1), or an increased RR (mean RR >1). By clearly defining the second component (mean RR = 1) as the null distribution, 3CMM estimates local false discovery rates (FDRs) for ADE signals under the empirical Bayes framework. Compared with existing approaches, the local FDR's top signals have noninferior or better sensitivities to detect true signals in both FAERS analysis and simulation studies. Additionally, we identify that the top signals of different approaches have different patterns, and they are complementary to each other.",
author = "Pengyue Zhang and Meng Li and Chiang, {Chien Wei} and Lei Wang and Yang Xiang and Lijun Cheng and Weixing Feng and Titus Schleyer and Sara Quinney and Wu, {Heng Yi} and Donglin Zeng and Lang Li",
year = "2018",
month = "8",
day = "1",
doi = "10.1002/psp4.12294",
language = "English (US)",
volume = "7",
pages = "499--506",
journal = "CPT: Pharmacometrics and Systems Pharmacology",
issn = "2163-8306",
publisher = "Nature Publishing Group",
number = "8",

}

TY - JOUR

T1 - Three-Component Mixture Model-Based Adverse Drug Event Signal Detection for the Adverse Event Reporting System

AU - Zhang, Pengyue

AU - Li, Meng

AU - Chiang, Chien Wei

AU - Wang, Lei

AU - Xiang, Yang

AU - Cheng, Lijun

AU - Feng, Weixing

AU - Schleyer, Titus

AU - Quinney, Sara

AU - Wu, Heng Yi

AU - Zeng, Donglin

AU - Li, Lang

PY - 2018/8/1

Y1 - 2018/8/1

N2 - The US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) is an important source for detecting adverse drug event (ADE) signals. In this article, we propose a three-component mixture model (3CMM) for FAERS signal detection. In 3CMM, a drug-ADE pair is assumed to have either a zero relative risk (RR), or a background RR (mean RR = 1), or an increased RR (mean RR >1). By clearly defining the second component (mean RR = 1) as the null distribution, 3CMM estimates local false discovery rates (FDRs) for ADE signals under the empirical Bayes framework. Compared with existing approaches, the local FDR's top signals have noninferior or better sensitivities to detect true signals in both FAERS analysis and simulation studies. Additionally, we identify that the top signals of different approaches have different patterns, and they are complementary to each other.

AB - The US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) is an important source for detecting adverse drug event (ADE) signals. In this article, we propose a three-component mixture model (3CMM) for FAERS signal detection. In 3CMM, a drug-ADE pair is assumed to have either a zero relative risk (RR), or a background RR (mean RR = 1), or an increased RR (mean RR >1). By clearly defining the second component (mean RR = 1) as the null distribution, 3CMM estimates local false discovery rates (FDRs) for ADE signals under the empirical Bayes framework. Compared with existing approaches, the local FDR's top signals have noninferior or better sensitivities to detect true signals in both FAERS analysis and simulation studies. Additionally, we identify that the top signals of different approaches have different patterns, and they are complementary to each other.

UR - http://www.scopus.com/inward/record.url?scp=85052518175&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052518175&partnerID=8YFLogxK

U2 - 10.1002/psp4.12294

DO - 10.1002/psp4.12294

M3 - Article

C2 - 30091855

AN - SCOPUS:85052518175

VL - 7

SP - 499

EP - 506

JO - CPT: Pharmacometrics and Systems Pharmacology

JF - CPT: Pharmacometrics and Systems Pharmacology

SN - 2163-8306

IS - 8

ER -