Biomedical papers - Ahead of Print

Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. X:X | 10.5507/bp.2017.042

Analysis of the seasonal incidence of acute respiratory infections including influenza (ARI) in the Czech Republic - possible contribution of the functional data boxplot in epidemiology

Ondrej Vencaleka, Jan Kynclb
a Department of Mathematical Analysis and Applications of Mathematics, Faculty of Science, Palacky University Olomouc, Czech Republic
b Department of Infectious Diseases Epidemiology, Centre for Epidemiology and Microbiology, National Institute of Public Health, Prague, Czech Republic

Aims: The detection of an epidemic outbreak is possible only if the baseline incidence level of a given disease is well defined. The determination of the baseline is complicated by the presence of epidemic outbreaks in historical data. The aim of the paper is to provide a new way of determining the baseline.

Methods: The analyzed data containing weekly records on the incidence of acute respiratory infections including influenza (ARI) in the Czech Republic and its regions are taken from the nationwide surveillance system; data on 15 seasons from 2001/02 to 2015/16 are included. Functional boxplots of the data are constructed and five distinct methods (componentwise mean, componentwise median, median, trimmed mean, and adjusted mean) were used for the computation of the baseline level function.

Results: It was shown that the methods based on functional data analysis could successfully overcome the problems that arise when the conventional methods are used for the determination of the baseline function.

Conclusion: The functional boxplot - a new statistical tool - can bring not only a transparent visualisation of comprehensive data, but can also help epidemiologists and other public health experts to determine the baseline incidence level of a given disease as well as to detect unusual epidemic seasons.

Keywords: epidemiology, acute respiratory infections, functional data analysis, boxplot, data depth

Received: February 28, 2017; Accepted: September 20, 2017; Prepublished online: October 17, 2017


References

  1. Sorensen H, Goldsmith J, Sangalli LM. An introduction with medical applications to functional data analysis. Stat Med 2013;32(30):5222-40. Go to original source... Go to PubMed...
  2. Wang J-L, Chiou J-M, Muller H-G. Functional data analysis. Annu Rev Statist 2016;3(1):257-95. Go to original source...
  3. Ramsay JO, Silverman BW. Functional Data Analysis. 2nd ed. New York: Springer; 2005.
  4. Prochazka B, Kyncl J. Estimating the Baseline Incidence of a Seasonal Disease Independently of Epidemic Outbreaks. Cent Eur J Public Health 2016;24(3):199-205. Go to original source... Go to PubMed...
  5. Lopez-Pintado S, Romo J. On the concept of depth for functional data. JASA 2009;104(486):718-34. Go to original source...
  6. Kyncl J, Havlickova M, Nagy A, Jirincova H, Piskova I. Early and unexpectedly severe start of influenza epidemic in the Czech Republic during influenza season 2012-13. Euro Surveill 2013 Feb 7 [cited 2017 Feb 28]; 18(6). Available from: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=20396. Go to PubMed...
  7. European Centre for Disease Prevention and Control (ECDC)/World Health Organization Regional Office for Europe (WHO/Europe). Flu News Europe. System description. ECDC/WHO. [cited 2017 Feb 28]. Available from: http://flunewseurope.org/System.
  8. Kyncl J, Paget WJ, Havlickova M, Kriz B. Harmonisation of the acute respiratory infection reporting system in the Czech Republic with the European community networks. Euro surveill. 2005 Mar 1 [cited 2017 Jun 26]; 10(3). Available from: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=525 Go to original source... Go to PubMed...
  9. Sun, Y, Genton, MG. Functional boxplots. J Comput Graph Statist 2011;20:316-34. Go to original source...
  10. Serfling R, Wijesuriya U. Depth-based nonparametric description of functional data, with emphasis on use of spatial depth. Comput Stat Data Anal 2017;105:24-45. Go to original source...
  11. R Core Team (2016). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available from: https://www.R-project.org/.