Regensburg 2010 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Downloads | Hilfe
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 13: Networks: From Topology to Dynamics I (with BP, DY)
SOE 13.3: Vortrag
Mittwoch, 24. März 2010, 10:45–11:00, H44
Contact networks and the spread of MRSA in hospitals — Lisa Brouwers1, •Andrzej Jarynowski1,2,3, Fredrik Liljeros1, and Xin Lu1 — 1Stockholm University, S106 91 Stockholm, Sweden — 2Department of Physics, Cologne University, Zülpicher Str. 77. 50937 Köln, Germany — 3The UNESCO Chair of interdisciplinary studies, Wroclaw University, pl. M. Borna 9 50-204 Wrocław, Poland
The bacterium meticillin resistant Staphylococcus aureus(MRSA)is known to be the largest care related the infection problem. We investigated the Common Care Registry containing information about all patient visits within Stockholm County during the outbreak period with registry over diagnosed MRSA cases. Methods to analyze the contact network of persons visiting the same care unit is developed within the project as well as methods to analyze in what way network structure affects the transmission of MRSA. We study matrixes of disease transition in hospitals population(infected versus people, who could sent infection). In stationary case:(a)We have matrixes of estimators of that probabilities and other statisical properties of contact networks. In time evolution case:(b)We divided outbreak in smaller, periodical intervals and looked at how MRSA was spreading in time. Quasi-MCMC(Markov chain Monte Carlo) method and artificial networks(main parameter is number of contacts during specific time interval)help us to understand real- and simulated-paths of disease transition. Matrixes of probabilities(b)were used to find mechanism of change states(vectors of all population 0-health or 1-ill)and we can run quasi-MCMC to get most likely paths.