Cardiac arrest (CA) is considered a major public health issue. Few studies have focused on geographic variations in incidence and socioeconomic characteristics. The aim of this study is to identify clusters of municipalities with high or low CA incidence, and find socioeconomic factors associated with them.
CA data from three Parisian counties, representing 123 municipalities, were extracted from the French CA registry. Socioeconomic data for each municipality were collected from the French national institute of statistics. We used a statistical approach combining Bayesian methods to study geographical CA incidence variations, and scan statistics, to identify CA incidence clusters of municipalities. Finally, we compared clusters of municipalities in terms of socioeconomic factors.
Strong geographical variations were found among 123 municipalities: 34 presented a significantly increased risk of incidence and 37 presented a significantly low risk. Scan statistics identified seven significant spatial clusters of CA incidence, including three clusters with low incidence (the relative risk varied from 0.23 to 0.54) and four clusters with high incidence (the relative risk varied from 1.43 to 2). Clusters of municipalities with a high CA incidence are characterized by a lower socioeconomic status than the others (low and normal CA incidence clusters). Analysis showed a statistically significant relationship between social deprivation factors and high incidence.
This study shows strong geographical variations in CA incidence and a statistically significant relationship between over-incidence and social deprivation variables.
aDepartment of Public Health, EA2694, University of Lille
bFrench National Out-of-Hospital Cardiac Arrest Registry Research Team, RéAC
cGR-RéAC, French national out-of-hospital cardiac arrest registry research group, Lille
eAP-HP, Department of Emergency Medicine, Avicenne Hospital, Inserm U942, Paris 13 University
fAPHP, Department of Emergency Medicine, Henri Mondor University Hospital
gAPHP, Department of Emergency Medicine, Raymond Poincaré University Hospital, Paris
hHCL, Department of Emergency Medicine, Lyon University Hospital, Claude Bernard University, Lyon, France
Received 7 July 2017 Accepted 30 October 2017
Correspondence to Hervé Hubert, PhD, Department of Public Health, EA2694, Institute of Health Engineering of Lille, University of Lille, 42, rue Ambroise Paré 59120 Loos, France Tel: +33 320 623 741; fax: +33 320 623 738; e-mail: email@example.com