Effectiveness of infection control strategies against intra- and inter-hospital transmission of MultidruG-resistant Enterobacteriaceae – insights from a multi-level mathematical NeTwork model
Hospital-acquired infections, including Multidrug-resistant Enterobacteriaceae (MDR-E), are recognised to be an important public health issue and cause significant burden on health care systems worldwide. While classic infection control strategies mainly target the spread of these pathogens at intra-hospital level (inside a hospital), inter-hospital (between various hospitals and health care institutions) patient movements have recently been implicated to trigger the spread of MDR-E.
Aim
Within the Emerge-Net project, network models describing the MDR-E transmission both at intra-hospital and inter-hospital levels have been developed. These models allow simulation of the transmission dynamics of MDR-E, specifically E. Coli and Klebsiella, and were used to investigate the effectiveness of infection control strategies at intra- and inter- hospital levels. In particular, these generic network models allowed us to study how interventions applied in one or more hospitals influence the transmission of MDR-E to other hospitals in the network connected through patient transfers.
Methods
As part of the Emerge-Net consortium, researchers developed an intra-hospital model for MDR-E spread, taking into account admissions and discharge of patients and their movements between different hospital wards. The model was used to simulate spread of MDR-E and the impact of various interventions in a hospital. In the model, if there is exposure of patients to the pathogen, patients can become colonised. Such colonised patients can transmit the pathogen to other susceptible patients in the same ward. Wards form a network, where the connections are defined by patients’ movements between the wards. Based on estimates for the duration of colonisation, transmission probabilities in every ward, and patient movement patterns, the model simulates possible outbreaks inside the hospital.
The researchers estimated model parameters and patient movement patterns from data collected in various participating hospitals. They obtained electronic medical record (EMR) of patient admissions from hospitals in Spain, The Netherlands, Germany, Poland, and Israel. They analysed individual patient movements by tracking the wards where patients stayed during the admission and constructed a network of patient movements to identify wards where most of the incoming and outgoing patient flow occurs. Furthermore, stratifying patients into risk groups, they observed that 30.0 % - 37.0 % of admissions were patients were characterised as high-risk in in all participating hospitals. Moreover, the mean length of stay per admission was higher for high-risk patients compared to low-risk patients. Additionally, the mean movement rate per hospital admission was higher for high-risk patients.
Results
Using the simulation model, the researchers analysed the networks of movement patterns in all participating hospitals and simulated possible spread of a pathogen through these networks. They investigated how spread of pathogens depended on various centrality measures of the connected wards. Finally, they simulated effectiveness of different intervention strategies within a hospital to prevent spread of pathogens. As a spin-off of this project, the researchers used the simulation model to study effective interventions for preventing nosocomial infections with SARS-CoV2.
The researchers collaborated within the Emerge-Net consortium to contribute to modelling the inter-hospital transmission of MDR-E and to combine intra- and interhospital transmission models. For the inter-hospital modelling they use data on patient transfers between hospitals from insurance companies in Germany.
Products
Author: Xia H, Horn J, Piotrowska MJ, Sakowski K, Karch A, Tahir H, Kretzschmar M, Mikolajczyk R
Magazine: PLoS Computational Biology
Author: Piotrowska MJ, Sakowski K, Lonc A, Tahir H, Kretzschmar ME.
Magazine: Epidemics
Author: Tahir H, López-Cortés LE, Kola A, Yahav D, Karch A, Xia H, Horn J, Sakowski K, Piotrowska MJ, Leibovici L, Mikolajczyk RT, Kretzschmar ME.
Magazine: PLoS Computational Biology
Author: Pham TM, Tahir H, van de Wijgert JHHM, Van der Roest BR, Ellerbroek P, Bonten MJM, Bootsma MCJ, Kretzschmar ME
Magazine: BMC Medicine
Author: Piotrowska MJ, Sakowski K, Karch A, Tahir H, Horn J, Kretzschmar ME, Mikolajczyk RT
Magazine: PLoS Computational Biology