A regional outbreak of a rare CRE (producing New Delhi metallo-β-lactamase [NDM]) was traced to duodenoscope exposure at Hospital A, a tertiary care hospital in northeastern Illinois (January – October 2013) precipitating changes to scope reprocessing procedures. We can apply social network analysis to construct an ego network, which identifies those hospitals most likely to share patients with a hospital experiencing an outbreak. An ego network that can predict which hospitals are most likely to encounter outbreak patients could guide clinical and public health interventions.
METHODS:
We established Hospital A’s ego network using UCINET software and Illinois’ hospital discharge data, defining a connection as two facilities sharing at least 50 patients over the two-year study period (January 2013 – December 2014). We defined outbreak cases as any NDM case reported to the XDRO registry by Hospital A in 2013, and then matched individual cases by name and date-of-birth to their Illinois hospital discharge database records (acute- and long-term acute care hospitals); we identified which facilities outbreak patients subsequently visited. We then searched the XDRO registry for new NDM cases (not at Hospital A) reported after the initial outbreak assuming that, due to NDM’s rarity at the time, the organisms most likely had disseminated from Hospital A.
RESULTS:
We identified 22 facilities in Hospital A’s ego network (out of 66 possible facilities in the same geographic region). Of 31 outbreak case patients entered into the XDRO registry that had visited hospital A, 19 (61%) were admitted to 13 other hospitals following the outbreak. Of the 13 hospitals, 9 (69%) were in the ego network, and 5 of those 9 hospitals reported at least one additional NDM case (not the transferred patients). Overall, there were 10 new NDM cases in patients who had not been to Hospital A, and 9 of these were reported by facilities in the ego network. The facility with the highest number of patients shared with Hospital A reported the most new, unique NDM cases (N=4).
CONCLUSIONS:
Ascertaining a hospital’s ego network accurately identified hospitals most likely to report new NDM cases; there was rare documented transmission (1 case) outside the ego network. Had public health professionals known Hospitals A’s ego network during the outbreak, these hospitals could have been targeted for early intervention, perhaps preventing regional dissemination of NDM. In the case of future outbreaks, knowing a hospital’s ego network would help prioritize which facilities receive prompt public health intervention.