Inference and control of the nosocomial transmission of methicillin-resistant Staphylococcus aureus

Pei, Sen; Morone, Flaviano; Liljeros, Fredrik; Makse, Hernán; Shaman, Jeffrey L

Distributions of some key statistics of the patient flow in infected wards differ from those in uninfected ones ( Figure 1C ). Infected wards tend to have a higher number of inpatients, a longer average length of stay as well as a larger ward size. Intuitively, the number of MRSA infections in a ward should increase as patient-days within the ward increase. However, the average number of infections is not observed to increase linearly with patient-days, indicating that patient-days per ward alone cannot explain the observed patterns of infection (Figure 1-figure supplement 1). While these raw features provide a general understanding of MRSA transmission, they cannot be effectively employed to assess infection or colonization risk in a specific ward due to their largely overlapping distributions, which prevent a clear classification of risk. Instead, a quantitative analysis using mathematical modeling is needed.