Lulla Opatowski - Institute Pasteur
Healthcare-associated infections with Multidrug resistant bacteria (MDR) represent a huge public health issue worldwide. Implementation of effective control measures is crucial for hospitals but can generate non- negligible disorganisation and costs. However, designing efficient control measures is complex, due to the strong stochasticity of events in hospitals and our still limited understanding transmission routes of these pathogens. In this context, hypothesis driven mathematical models are particularly useful. They can help analyse MRB epidemics, estimate unknown parameters and simulate counterfactual scenarios to help implementing optimized control measure. In this talk I will present a series of studies in which we developed specific stochastic individual based models of MRB spread in hospitals to investigate the role of between- human contacts within and between wards on pathogen transmission. The first study focuses on the analysis of the between-ward dissemination of vancomycin-resistant Enterococci (VRE) in hospitals. Based on specific data extracted from a real outbreak in a French hospital and Bayesian inference, a spatially explicit model was built and used to assess the economic and human costs associated with a series of hypothetical scenarios of control measure implementation. The second study focuses on the analysis of methicillin-resistant Staphylococcus aureus (MRSA) transmission over a network of patients and health-care workers. Based on detailed contact- data collected using RFID log-censors in a French rehabilitation hospital, an agent-based model explicitly formalizing inter-individual contacts was built and used to analyse how contact-based control measures can help reducing the spread of MRB in hospitals.