Detailed contact data and the dissemination of Staphylococcus aureus in hospitals

Abstract

Close proximity interactions (CPIs) measured by wireless electronic devices are increasingly used in epidemiological models. However, no evidence supports that electronically collected CPIs inform on the contacts leading to transmission. Here, we analyzed Staphylococcus aureus carriage and CPIs recorded simultaneously in a long-term care facility for 4 months in 329 patients and 261 healthcare workers to test this hypothesis. In the broad diversity of isolated S. aureus strains, 173 transmission events were observed between participants. The joint analysis of carriage and CPIs showed that CPI paths linking incident cases to other individuals carrying the same strain (i.e. possible infectors) had fewer intermediaries than predicted by chance (P < 0.001), a feature that simulations showed to be the signature of transmission along CPIs. Additional analyses revealed a higher dissemination risk between patients via healthcare workers than via other patients. In conclusion, S. aureus transmission was consistent with contacts defined by electronically collected CPIs, illustrating their potential as a tool to control hospital-acquired infections and help direct surveillance.

Fig 1. CPI-based network and S. aureus carriage in the hospital.

The network shown corresponds to interactions occurring during 1 d. Patients are shown as triangles and HCWs as diamonds. Color-coding corresponds to the spa type of the last known colonization status (i.e. the preceding week). Because of the large S. aureus spa types diversity, only the 3 most common are reported in the legend. We used Fruchterman-Reingold force-directed algorithm for the layout (individuals in the network are closer together as the density of links among them increases). Force-directed edge bundling was used to accommodate their high density.

Fig 2. Mean daily numbers and durations of patients’ and HCWs’ CPIs.

Bars illustrate CPIs with patients (black) and HCWs (white).

Fig 3. CPI supported transmitters.

(Top) CPI-supported transmitters were selected among carriers of the same strain (green) as the incident case (yellow) who were the closest in the CPI network. Here, P1 and H1 are two CPI-supported transmitters in the incident case’s 2-hop neighborhood, but not P2 who is further away (3-hop neighborhood). Patients are shown as triangles and HCWs as diamonds. (Bottom left) Comparison of the distance distribution between CPI-supported transmitters and incident cases in the data (black) and with random permutations of carriage data (white) in the simulation study. (Bottom right) Comparison of the distance between CPI-supported transmitters and incident cases in the original data.