Assessing the role of a patient transfer network in the spread of carbapenemase-producing Enterobacteriaceae: The case of France between 2012 and 2015

The spread of carbapenemase-producing Enterobacteriaceae (CPE) is a major public health threat that has been associated with cross-border and local transfer of patients between healthcare facilities. However, the impact of healthcare transfer networks on CPE spread dynamics may vary in time and betw...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Revue d'épidémiologie et de santé publique 2018-07, Vol.66, p.S265-S266
Hauptverfasser: Nekkab, N., Crépey, P., Temime, L.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The spread of carbapenemase-producing Enterobacteriaceae (CPE) is a major public health threat that has been associated with cross-border and local transfer of patients between healthcare facilities. However, the impact of healthcare transfer networks on CPE spread dynamics may vary in time and between countries. In this context, our study aimed to assess the contribution of the patient transfer network on CPE spread in France from 2012 to 2015. Using the French healthcare network of 2.3 million patients, we extended a previously proposed statistical method and tested the ability of this network to support observed CPE incidence episodes. First, using 2237 CPE episodes that occurred from 2012 to 2015, we identified the most likely infector for the 1251 non-imported episodes using network-supported paths (NSPs). We then compared the observed NSP distances to those expected by chance, using random permutations of the CPE data. The impact of the assumed time window between infector episode and CPE episode was investigated in a sensitivity analysis. More than half of all CPE episodes were linked, either as infectors or incident episodes. The percentage of episodes with identified potential infectors over the network increased with time, from 57% in 2012 to 66% in 2015. NSP distances from 2013 to 2015 were significantly shorter in the observed data than expected by chance, indicating a role of the transfer network in CPE spread dynamics in France. In 2012 however, this result was not found. Over the entire study period, linked episodes tended to occur in the same administrative department or within close geographic distances. The 3-to-4 weeks baseline window between infector and episode was supported by the sensitivity analysis, where the strongest evidence for network-supported CPE transmission was observed for 2014 and 2015 episodes. We observed a transition in 2013 from an epidemic sustained by importation to local transmission events sustaining the epidemic. As a result of a growing contribution of transfers in CPE spread over time, coordinated prevention and infection control strategies in France should focus on at-risk patient transfers to reduce regional and inter-regional transmission of CPE.
ISSN:0398-7620
1773-0627
DOI:10.1016/j.respe.2018.05.080