The present chapter will report an overview of the main statistical methods that can be used to provide quantitative estimates of the association between count variables, such as micronuclei (MN) and genotoxic exposures. The description of data is a critical step of statistical analysis, since possible mistakes can be detected and corrected in this phase. Towards this aim, traditional (histograms and box and whisker plots) and advanced [spline and locally weighted scatterplot smoothing (LOWESS)] graphical procedures are described. Other basic procedures, such as sample size evaluation and data transformation, are described with a practical approach. Different regression models that can be used to fit MN data, which take into consideration confounding and detect effect modification, are discussed, including normal, log-normal, Poisson and negative binomial distributions. The chapter briefly illustrates in a non-technical manner the advanced procedures that can be efficiently applied to MN analysis, such as zero-inflated count and multilevel count modelling.

Original languageEnglish
Title of host publicationChemical Health Threats
Subtitle of host publicationAssessing and Alerting
EditorsMichael Fenech, Siegfried Knasmuller
PublisherRoyal Society of Chemistry
Number of pages18
Publication statusPublished - Jan 1 2019

Publication series

NameIssues in Toxicology
ISSN (Print)1757-7179
ISSN (Electronic)1757-7187

ASJC Scopus subject areas

  • Toxicology
  • Pharmacology
  • Health, Toxicology and Mutagenesis


Dive into the research topics of 'CHAPTER 21: Micronucleus Assay: Epidemiological and Statistical Issues'. Together they form a unique fingerprint.

Cite this