Modelling MRI enhancing lesion counts in multiple sclerosis using a negative binomial model: Implications for clinical trials

M. P. Sormani, P. Bruzzi, D. H. Miller, C. Gasperini, F. Barkhof, M. Filippi

Research output: Contribution to journalArticle

59 Citations (Scopus)

Abstract

In multiple sclerosis (MS) the number of new enhancing lesions seen on monthly magnetic resonance imaging (MRI) scans is the most widely used response variable in MRI-monitored studies of experimental treatments. However, no statistical model has been proposed to describe the distribution of the number of such lesions across MS patients. This article briefly summarizes the statistical models for counted data. The negative binomial (NB) model is proposed to fit the number of new enhancing lesions counted in a set of 56 untreated MS patients followed for 9 months. It is shown that the large variability present in this data set is better addressed by the NB model (residual deviance=66.6, 54 degrees of freedom) than by the Poisson model (residual deviance=1830.1, 55 degrees of freedom). Applications of the parametrization of lesion counts are discussed, and an example related to computer simulations for the sample size estimation is presented.

Original languageEnglish
Pages (from-to)74-80
Number of pages7
JournalJournal of the Neurological Sciences
Volume163
Issue number1
DOIs
Publication statusPublished - Feb 1 1999

Fingerprint

Statistical Models
Multiple Sclerosis
Magnetic Resonance Imaging
Clinical Trials
Computer Simulation
Sample Size
Therapeutics

Keywords

  • Magnetic resonance imaging
  • Multiple sclerosis
  • Negative binomial distribution

ASJC Scopus subject areas

  • Ageing
  • Clinical Neurology
  • Surgery
  • Developmental Neuroscience
  • Neurology
  • Neuroscience(all)

Cite this

Modelling MRI enhancing lesion counts in multiple sclerosis using a negative binomial model : Implications for clinical trials. / Sormani, M. P.; Bruzzi, P.; Miller, D. H.; Gasperini, C.; Barkhof, F.; Filippi, M.

In: Journal of the Neurological Sciences, Vol. 163, No. 1, 01.02.1999, p. 74-80.

Research output: Contribution to journalArticle

@article{9ea67a26a9f5447ebb499bf3916c962c,
title = "Modelling MRI enhancing lesion counts in multiple sclerosis using a negative binomial model: Implications for clinical trials",
abstract = "In multiple sclerosis (MS) the number of new enhancing lesions seen on monthly magnetic resonance imaging (MRI) scans is the most widely used response variable in MRI-monitored studies of experimental treatments. However, no statistical model has been proposed to describe the distribution of the number of such lesions across MS patients. This article briefly summarizes the statistical models for counted data. The negative binomial (NB) model is proposed to fit the number of new enhancing lesions counted in a set of 56 untreated MS patients followed for 9 months. It is shown that the large variability present in this data set is better addressed by the NB model (residual deviance=66.6, 54 degrees of freedom) than by the Poisson model (residual deviance=1830.1, 55 degrees of freedom). Applications of the parametrization of lesion counts are discussed, and an example related to computer simulations for the sample size estimation is presented.",
keywords = "Magnetic resonance imaging, Multiple sclerosis, Negative binomial distribution",
author = "Sormani, {M. P.} and P. Bruzzi and Miller, {D. H.} and C. Gasperini and F. Barkhof and M. Filippi",
year = "1999",
month = "2",
day = "1",
doi = "10.1016/S0022-510X(99)00015-5",
language = "English",
volume = "163",
pages = "74--80",
journal = "Journal of the Neurological Sciences",
issn = "0022-510X",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - Modelling MRI enhancing lesion counts in multiple sclerosis using a negative binomial model

T2 - Implications for clinical trials

AU - Sormani, M. P.

AU - Bruzzi, P.

AU - Miller, D. H.

AU - Gasperini, C.

AU - Barkhof, F.

AU - Filippi, M.

PY - 1999/2/1

Y1 - 1999/2/1

N2 - In multiple sclerosis (MS) the number of new enhancing lesions seen on monthly magnetic resonance imaging (MRI) scans is the most widely used response variable in MRI-monitored studies of experimental treatments. However, no statistical model has been proposed to describe the distribution of the number of such lesions across MS patients. This article briefly summarizes the statistical models for counted data. The negative binomial (NB) model is proposed to fit the number of new enhancing lesions counted in a set of 56 untreated MS patients followed for 9 months. It is shown that the large variability present in this data set is better addressed by the NB model (residual deviance=66.6, 54 degrees of freedom) than by the Poisson model (residual deviance=1830.1, 55 degrees of freedom). Applications of the parametrization of lesion counts are discussed, and an example related to computer simulations for the sample size estimation is presented.

AB - In multiple sclerosis (MS) the number of new enhancing lesions seen on monthly magnetic resonance imaging (MRI) scans is the most widely used response variable in MRI-monitored studies of experimental treatments. However, no statistical model has been proposed to describe the distribution of the number of such lesions across MS patients. This article briefly summarizes the statistical models for counted data. The negative binomial (NB) model is proposed to fit the number of new enhancing lesions counted in a set of 56 untreated MS patients followed for 9 months. It is shown that the large variability present in this data set is better addressed by the NB model (residual deviance=66.6, 54 degrees of freedom) than by the Poisson model (residual deviance=1830.1, 55 degrees of freedom). Applications of the parametrization of lesion counts are discussed, and an example related to computer simulations for the sample size estimation is presented.

KW - Magnetic resonance imaging

KW - Multiple sclerosis

KW - Negative binomial distribution

UR - http://www.scopus.com/inward/record.url?scp=0033081690&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033081690&partnerID=8YFLogxK

U2 - 10.1016/S0022-510X(99)00015-5

DO - 10.1016/S0022-510X(99)00015-5

M3 - Article

C2 - 10223415

AN - SCOPUS:0033081690

VL - 163

SP - 74

EP - 80

JO - Journal of the Neurological Sciences

JF - Journal of the Neurological Sciences

SN - 0022-510X

IS - 1

ER -