Overdispersion models for correlated multinomial data: Applications to blinding assessment

V. Landsman, D. Landsman, C. S. Li, H. Bang

Research output: Contribution to journalArticle

Abstract

Overdispersion models have been extensively studied for correlated normal and binomial data but much less so for correlated multinomial data. In this work, we describe a multinomial overdispersion model that leads to the specification of the first two moments of the outcome and allows the estimation of the global parameters using generalized estimating equations (GEE). We introduce a Global Blinding Index as a target parameter and illustrate the application of the GEE method to its estimation from (1) a clinical trial with clustering by practitioner and (2) a meta-analysis on psychiatric disorders. We examine the impact of a small number of clusters, high variability in cluster sizes, and the magnitude of the intraclass correlation on the performance of the GEE estimators of the Global Blinding Index using the data simulated from different models. We compare these estimators with the inverse-variance weighted estimators and a maximum-likelihood estimator, derived under the Dirichlet-multinomial model. Our results indicate that the performance of the GEE estimators was satisfactory even in situations with a small number of clusters, whereas the inverse-variance weighted estimators performed poorly, especially for larger values of the intraclass correlation coefficient. Our findings and illustrations may be instrumental for practitioners who analyze clustered multinomial data from clinical trials and/or meta-analysis.

Original languageEnglish (US)
JournalStatistics in Medicine
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Overdispersion
Generalized Estimating Equations
Meta-Analysis
Clinical Trials
Estimator
Cluster Analysis
Psychiatry
Number of Clusters
Intraclass Correlation Coefficient
Intraclass Correlation
Multinomial Model
Model
Maximum Likelihood Estimator
Dirichlet
Disorder
Clustering
Specification
Moment
Target

Keywords

  • blinding index
  • Dirichlet-multinomial
  • GEE
  • meta-analysis

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Overdispersion models for correlated multinomial data : Applications to blinding assessment. / Landsman, V.; Landsman, D.; Li, C. S.; Bang, H.

In: Statistics in Medicine, 01.01.2019.

Research output: Contribution to journalArticle

@article{323c1b3672ff448bb2c9d38952565747,
title = "Overdispersion models for correlated multinomial data: Applications to blinding assessment",
abstract = "Overdispersion models have been extensively studied for correlated normal and binomial data but much less so for correlated multinomial data. In this work, we describe a multinomial overdispersion model that leads to the specification of the first two moments of the outcome and allows the estimation of the global parameters using generalized estimating equations (GEE). We introduce a Global Blinding Index as a target parameter and illustrate the application of the GEE method to its estimation from (1) a clinical trial with clustering by practitioner and (2) a meta-analysis on psychiatric disorders. We examine the impact of a small number of clusters, high variability in cluster sizes, and the magnitude of the intraclass correlation on the performance of the GEE estimators of the Global Blinding Index using the data simulated from different models. We compare these estimators with the inverse-variance weighted estimators and a maximum-likelihood estimator, derived under the Dirichlet-multinomial model. Our results indicate that the performance of the GEE estimators was satisfactory even in situations with a small number of clusters, whereas the inverse-variance weighted estimators performed poorly, especially for larger values of the intraclass correlation coefficient. Our findings and illustrations may be instrumental for practitioners who analyze clustered multinomial data from clinical trials and/or meta-analysis.",
keywords = "blinding index, Dirichlet-multinomial, GEE, meta-analysis",
author = "V. Landsman and D. Landsman and Li, {C. S.} and H. Bang",
year = "2019",
month = "1",
day = "1",
doi = "10.1002/sim.8344",
language = "English (US)",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",

}

TY - JOUR

T1 - Overdispersion models for correlated multinomial data

T2 - Applications to blinding assessment

AU - Landsman, V.

AU - Landsman, D.

AU - Li, C. S.

AU - Bang, H.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Overdispersion models have been extensively studied for correlated normal and binomial data but much less so for correlated multinomial data. In this work, we describe a multinomial overdispersion model that leads to the specification of the first two moments of the outcome and allows the estimation of the global parameters using generalized estimating equations (GEE). We introduce a Global Blinding Index as a target parameter and illustrate the application of the GEE method to its estimation from (1) a clinical trial with clustering by practitioner and (2) a meta-analysis on psychiatric disorders. We examine the impact of a small number of clusters, high variability in cluster sizes, and the magnitude of the intraclass correlation on the performance of the GEE estimators of the Global Blinding Index using the data simulated from different models. We compare these estimators with the inverse-variance weighted estimators and a maximum-likelihood estimator, derived under the Dirichlet-multinomial model. Our results indicate that the performance of the GEE estimators was satisfactory even in situations with a small number of clusters, whereas the inverse-variance weighted estimators performed poorly, especially for larger values of the intraclass correlation coefficient. Our findings and illustrations may be instrumental for practitioners who analyze clustered multinomial data from clinical trials and/or meta-analysis.

AB - Overdispersion models have been extensively studied for correlated normal and binomial data but much less so for correlated multinomial data. In this work, we describe a multinomial overdispersion model that leads to the specification of the first two moments of the outcome and allows the estimation of the global parameters using generalized estimating equations (GEE). We introduce a Global Blinding Index as a target parameter and illustrate the application of the GEE method to its estimation from (1) a clinical trial with clustering by practitioner and (2) a meta-analysis on psychiatric disorders. We examine the impact of a small number of clusters, high variability in cluster sizes, and the magnitude of the intraclass correlation on the performance of the GEE estimators of the Global Blinding Index using the data simulated from different models. We compare these estimators with the inverse-variance weighted estimators and a maximum-likelihood estimator, derived under the Dirichlet-multinomial model. Our results indicate that the performance of the GEE estimators was satisfactory even in situations with a small number of clusters, whereas the inverse-variance weighted estimators performed poorly, especially for larger values of the intraclass correlation coefficient. Our findings and illustrations may be instrumental for practitioners who analyze clustered multinomial data from clinical trials and/or meta-analysis.

KW - blinding index

KW - Dirichlet-multinomial

KW - GEE

KW - meta-analysis

UR - http://www.scopus.com/inward/record.url?scp=85071317768&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071317768&partnerID=8YFLogxK

U2 - 10.1002/sim.8344

DO - 10.1002/sim.8344

M3 - Article

C2 - 31460677

AN - SCOPUS:85071317768

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

ER -