Skip to Main content Skip to Navigation

Comprendre et contrôler la transmission des bactéries multirésistantes par l'analyse et la modélisation des réseaux d’interactions interindividuelles en milieu hospitalier

Abstract : Healthcare-associated infections represent a huge public health issue worldwide. Multidrug resistant bacteria (MDR) are a major cause of these infections. Hence, better understanding their transmission routes in hospital settings is crucial to design efficient control measures.The purpose of this thesis is to use detailed data on interindividual contact networks, associated with mathematical modelling methods, to study MDR spread in hospitals and improve their control. To this end, data collected during the i-Bird study was used. This longitudinal prospective study took place at the Berck-sur-Mer hospital during 4 months in 2009. Close proximity interactions were recorded by the use of RFID (Radio Frequency Identification Devices) sensors everyday. Meanwhile, microbiological swabs were collected weekly.In a first part, interindividual contact patterns within and between each individual categories (patients, nurses, hospital porters, etc.) were analyzed. This first study notably underlined the importance of patient-to-patient contacts in long-term care facilities (LTCF). Moreover, some hospital staff categories, such as hospital porters and physicians, were identified as potential superspreaders based on their contact patterns.In a second part, we investigated the impact of the contact network on the spread of two species of Extended-spectrum beta-lactamases (ESBL) Enterobacteriaceae (E. coli and K. pneumoniae). This work showed that the contact network was an important driver of ESBL-K. pneumoniae dynamics, but not of ESBL-E. coli dynamics over the i-Bird study.The last part of the thesis was dedicated to the development of an agent-based model of MDR spread in hospital settings that explicitly formalizes detailed interindividual contacts. This model allows to assess control measures focused on contact patterns. The model was applied to the i-Bird data; we simulated methicillin-resistant Staphylococcus aureus (MRSA) transmission during the 4-month study over the reported contact network. Using our simations, we evaluated measures associated with hospital staff cohorting and showed it can lead to reduce the MRSA acquisition=.This thesis combines network analysis, epidemiology of infectious diseases and dynamic modeling. It allows a better understanding of MDR spread and control in LTCF. Moreover, it brings an innovative tool, intended to be developed, to understand and control BMR spread through contact networks in hospital settings.
Document type :
Complete list of metadata

Cited literature [251 references]  Display  Hide  Download
Contributor : ABES STAR :  Contact
Submitted on : Wednesday, January 22, 2020 - 11:19:10 AM
Last modification on : Tuesday, April 19, 2022 - 10:14:51 AM
Long-term archiving on: : Thursday, April 23, 2020 - 1:40:23 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02448298, version 1



Audrey Duval. Comprendre et contrôler la transmission des bactéries multirésistantes par l'analyse et la modélisation des réseaux d’interactions interindividuelles en milieu hospitalier. Médecine humaine et pathologie. Université Paris-Saclay, 2019. Français. ⟨NNT : 2019SACLV075⟩. ⟨tel-02448298⟩



Record views


Files downloads