Skip to Main content Skip to Navigation
Master Thesis

Estimations intégratives des incidences régionales des syndromes grippaux en France par utilisation de données de délivrances médicamenteuses

Abstract : Policies to prevent and fight against epidemics of influenza-like illness (ILI) require epidemiological surveillance continuously, based on the statements in real-time GPs volunteers. The weekly calculation of incidences at national and regional level allows us to estimate the spatial distribution of outbreaks in France. However, the stability of these estimates varies over time and across regions. This work aims to improve regional estimates using external data source, based medication sales. Two areas of study are presented here: one consists in the "spatial disaggregation" of a model fitted on national sales, the other in "spatial extrapolation" of a model fitted on sales in Rhône-Alpes, French region chosen for the stability of its incidences. For this, a preselection procedure of drug classes related to ILI was developed by classification methods. Similarly, different models of periodic regression (log-linear, log-nonlinear and PLS-Poisson) and approaches (unmoving or moving models with fixed or unfixed parameters) were compared to select the best model for predicting the incidences. The spatial "disaggregation" and "extrapolation" allowed us to propose new regional estimates of the incidences, of better stability. Finally, a bootstrap was used to estimate a confidence interval around the new predictions.
Document type :
Master Thesis
Complete list of metadata

Cited literature [18 references]  Display  Hide  Download

https://dumas.ccsd.cnrs.fr/dumas-01105254
Contributor : Agrocampus Ouest <>
Submitted on : Tuesday, January 20, 2015 - 9:51:47 AM
Last modification on : Monday, July 6, 2020 - 3:38:21 PM
Long-term archiving on: : Tuesday, April 21, 2015 - 10:16:27 AM

Identifiers

  • HAL Id : dumas-01105254, version 1

Collections

Citation

Cyril Esnault. Estimations intégratives des incidences régionales des syndromes grippaux en France par utilisation de données de délivrances médicamenteuses. Sciences agricoles. 2014. ⟨dumas-01105254⟩

Share

Metrics

Record views

187

Files downloads

132