Statistical analysis of data from dilution assays with censored correlated counts

Jorge Quiroz, Jeffrey Wilson, Satrajit Roychoudhury

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Frequently, count data obtained from dilution assays are subject to an upper detection limit, and as such, data obtained from these assays are usually censored. Also, counts from the same subject at different dilution levels are correlated. Ignoring the censoring and the correlation may provide unreliable and misleading results. Therefore, any meaningful data modeling requires that the censoring and the correlation be simultaneously addressed. Such comprehensive approaches of modeling censoring and correlation are not widely used in the analysis of dilution assays data. Traditionally, these data are analyzed using a general linear model on a logarithmic-transformed average count per subject. However, this traditional approach ignores the between-subject variability and risks, providing inconsistent results and unreliable conclusions. In this paper, we propose the use of a censored negative binomial model with normal random effects to analyze such data. This model addresses, in addition to the censoring and the correlation, any overdispersion that may be present in count data. The model is shown to be widely accessible through the use of several modern statistical software.

Original languageEnglish (US)
Pages (from-to)63-73
Number of pages11
JournalPharmaceutical Statistics
Volume11
Issue number1
DOIs
StatePublished - Jan 2012

Fingerprint

Statistical Data Interpretation
Statistical Models
Censoring
Statistical Analysis
Limit of Detection
Linear Models
Count
Software
Count Data
Logarithmic Average
Negative Binomial Model
Overdispersion
Statistical Software
Detection Limit
Data Modeling
Random Effects
Inconsistent
Linear Model
Modeling
Model

Keywords

  • censored correlated count data
  • censored mixed negative binomial regression
  • hierarchical likelihood
  • NLMIXED
  • R programming language

ASJC Scopus subject areas

  • Pharmacology (medical)
  • Statistics and Probability
  • Pharmacology

Cite this

Statistical analysis of data from dilution assays with censored correlated counts. / Quiroz, Jorge; Wilson, Jeffrey; Roychoudhury, Satrajit.

In: Pharmaceutical Statistics, Vol. 11, No. 1, 01.2012, p. 63-73.

Research output: Contribution to journalArticle

Quiroz, Jorge ; Wilson, Jeffrey ; Roychoudhury, Satrajit. / Statistical analysis of data from dilution assays with censored correlated counts. In: Pharmaceutical Statistics. 2012 ; Vol. 11, No. 1. pp. 63-73.
@article{01e50c40cfb64c5ea811c23a262ecd5b,
title = "Statistical analysis of data from dilution assays with censored correlated counts",
abstract = "Frequently, count data obtained from dilution assays are subject to an upper detection limit, and as such, data obtained from these assays are usually censored. Also, counts from the same subject at different dilution levels are correlated. Ignoring the censoring and the correlation may provide unreliable and misleading results. Therefore, any meaningful data modeling requires that the censoring and the correlation be simultaneously addressed. Such comprehensive approaches of modeling censoring and correlation are not widely used in the analysis of dilution assays data. Traditionally, these data are analyzed using a general linear model on a logarithmic-transformed average count per subject. However, this traditional approach ignores the between-subject variability and risks, providing inconsistent results and unreliable conclusions. In this paper, we propose the use of a censored negative binomial model with normal random effects to analyze such data. This model addresses, in addition to the censoring and the correlation, any overdispersion that may be present in count data. The model is shown to be widely accessible through the use of several modern statistical software.",
keywords = "censored correlated count data, censored mixed negative binomial regression, hierarchical likelihood, NLMIXED, R programming language",
author = "Jorge Quiroz and Jeffrey Wilson and Satrajit Roychoudhury",
year = "2012",
month = "1",
doi = "10.1002/pst.499",
language = "English (US)",
volume = "11",
pages = "63--73",
journal = "Pharmaceutical Statistics",
issn = "1539-1604",
publisher = "John Wiley and Sons Ltd",
number = "1",

}

TY - JOUR

T1 - Statistical analysis of data from dilution assays with censored correlated counts

AU - Quiroz, Jorge

AU - Wilson, Jeffrey

AU - Roychoudhury, Satrajit

PY - 2012/1

Y1 - 2012/1

N2 - Frequently, count data obtained from dilution assays are subject to an upper detection limit, and as such, data obtained from these assays are usually censored. Also, counts from the same subject at different dilution levels are correlated. Ignoring the censoring and the correlation may provide unreliable and misleading results. Therefore, any meaningful data modeling requires that the censoring and the correlation be simultaneously addressed. Such comprehensive approaches of modeling censoring and correlation are not widely used in the analysis of dilution assays data. Traditionally, these data are analyzed using a general linear model on a logarithmic-transformed average count per subject. However, this traditional approach ignores the between-subject variability and risks, providing inconsistent results and unreliable conclusions. In this paper, we propose the use of a censored negative binomial model with normal random effects to analyze such data. This model addresses, in addition to the censoring and the correlation, any overdispersion that may be present in count data. The model is shown to be widely accessible through the use of several modern statistical software.

AB - Frequently, count data obtained from dilution assays are subject to an upper detection limit, and as such, data obtained from these assays are usually censored. Also, counts from the same subject at different dilution levels are correlated. Ignoring the censoring and the correlation may provide unreliable and misleading results. Therefore, any meaningful data modeling requires that the censoring and the correlation be simultaneously addressed. Such comprehensive approaches of modeling censoring and correlation are not widely used in the analysis of dilution assays data. Traditionally, these data are analyzed using a general linear model on a logarithmic-transformed average count per subject. However, this traditional approach ignores the between-subject variability and risks, providing inconsistent results and unreliable conclusions. In this paper, we propose the use of a censored negative binomial model with normal random effects to analyze such data. This model addresses, in addition to the censoring and the correlation, any overdispersion that may be present in count data. The model is shown to be widely accessible through the use of several modern statistical software.

KW - censored correlated count data

KW - censored mixed negative binomial regression

KW - hierarchical likelihood

KW - NLMIXED

KW - R programming language

UR - http://www.scopus.com/inward/record.url?scp=84862786774&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862786774&partnerID=8YFLogxK

U2 - 10.1002/pst.499

DO - 10.1002/pst.499

M3 - Article

C2 - 22275293

AN - SCOPUS:84862786774

VL - 11

SP - 63

EP - 73

JO - Pharmaceutical Statistics

JF - Pharmaceutical Statistics

SN - 1539-1604

IS - 1

ER -