The analysis of the geographical distribution of emergency departments' frequent users: a tool to prioritize public health policies?

BMC Public Health. 2021 Sep 16;21(1):1689. doi: 10.1186/s12889-021-11682-z.

Abstract

Background: The individual factors associated to Frequent Users (FUs) in Emergency Departments are well known. However, the characteristics of their geographical distribution and how territorial specificities are associated and intertwined with ED use are limited. Investigating healthcare use and territorial factors would help targeting local health policies. We aim at describing the geographical distribution of ED's FUs within the Paris region.

Methods: We performed a retrospective analysis of all ED visits in the Paris region in 2015. Data were collected from the universal health insurance's claims database. Frequent Users (FUs) were defined as having visited ≥3 times any ED of the region over the period. We assessed the FUs rate in each geographical unit (GU) and assessed correlations between FUs rate and socio-demographics and economic characteristics of GUs. We also performed a multidimensional analysis and a principal component analysis to identify a typology of territories to describe and target the FUs phenomenon.

Results: FUs accounted for 278,687 (11.7%) of the 2,382,802 patients who visited the ED, living in 232 GUs. In the region, median FUs rate in each GU was 11.0% [interquartile range: 9.5-12.5]. High FUs rate was correlated to the territorial markers of social deprivation. Three different categories of GU were identified with different profiles of healthcare providers densities.

Conclusion: FUs rate varies between territories and is correlated to territorial markers of social deprivation. Targeted public policies should focus on disadvantaged territories.

Keywords: Access to care; Emergency department; Frequent users; Health geography; Healthcare use.

MeSH terms

  • Emergency Service, Hospital*
  • Health Policy
  • Humans
  • Public Policy
  • Retrospective Studies
  • Vulnerable Populations*