tetano
Editor, Senior Moderator
Sci Rep. 2019 Feb 5;9(1):1422. doi: 10.1038/s41598-018-38292-x.
[h=1]A statistical tool for comparing seasonal ILI surveillance data.[/h] Ferland R[SUP]1[/SUP], Froda S[SUP]2[/SUP].
[h=3]Author information[/h]
[h=3]Abstract[/h] In this paper, we consider the yearly influenza epidemic, as reflected in the seasonal surveillance data compiled by the CDC (Center for Disease Control and Prevention, USA) and we explore a new methodology for comparing specific features of these data. In particular, we focus on the ten HHS (Health and Human Services) regions, and how the incidence data evolves in these regions. In order to perform the comparisons, we consider the relative distribution of weekly new cases over one season and replace the crude data with predicted values. These predictions are obtained after fitting a negative binomial regression model that controls for important covariates. The prediction is computed on a 'generic' set of covariate values that takes into account the relative size (population wise) of the regions to be compared. The main results are presented in graphical form, that quickly emphasizes relevant features of the seasonal data and facilitates the comparisons.
PMID: 30723245 PMCID: PMC6363783 DOI: 10.1038/s41598-018-38292-x
Free PMC Article
[h=1]A statistical tool for comparing seasonal ILI surveillance data.[/h] Ferland R[SUP]1[/SUP], Froda S[SUP]2[/SUP].
[h=3]Author information[/h]
[h=3]Abstract[/h] In this paper, we consider the yearly influenza epidemic, as reflected in the seasonal surveillance data compiled by the CDC (Center for Disease Control and Prevention, USA) and we explore a new methodology for comparing specific features of these data. In particular, we focus on the ten HHS (Health and Human Services) regions, and how the incidence data evolves in these regions. In order to perform the comparisons, we consider the relative distribution of weekly new cases over one season and replace the crude data with predicted values. These predictions are obtained after fitting a negative binomial regression model that controls for important covariates. The prediction is computed on a 'generic' set of covariate values that takes into account the relative size (population wise) of the regions to be compared. The main results are presented in graphical form, that quickly emphasizes relevant features of the seasonal data and facilitates the comparisons.
PMID: 30723245 PMCID: PMC6363783 DOI: 10.1038/s41598-018-38292-x
Free PMC Article