Exploratory Space–Time Analyses of Reported Lyme Borreliosis Cases in France, 2016–2019

Abstract

In recent decades, the incidence of Lyme borreliosis (LB) in Europe seems to have increased, underpinning a growing public health concern. LB surveillance systems across the continent are heterogeneous, and the spatial and temporal patterns of LB reports have been little documented. In this study, we explored the spatio-temporal patterns of LB cases reported in France from 2016 to 2019, to describe high-risk clusters and generate hypotheses on their occurrence. The space–time K-function and the Kulldorf’s scan statistic were implemented separately for each year to evaluate space–time interaction between reported cases and searching clusters. The results show that the main spatial clusters, of radius size up to 97 km, were reported in central and northeastern France each year. In 2017–2019, spatial clusters were also identified in more southern areas (near the Alps and the Mediterranean coast). Spatio-temporal clustering occurred between May and August, over one-month to three-month windows in 2016–2017 and in 2018–2019. A strong spatio-temporal interaction was identified in 2018 within 16 km and seven days, suggesting a potential local and intense pathogen transmission process. Ongoing improved surveillance and accounting for animal hosts, vectors, meteorological factors and human behaviors are keys to further elucidate LB spatio-temporal patterns.

Case presentation and management of Lyme disease patients: a 9-year retrospective analysis in France

“…Several regions of France, Auvergne-Rhône-Alpes, Grand-Est, Bretagne, Normandie and Ile-de-France, were identified as more at risk than others. This is in agreement with a recent study, which showed that main spatial clusters of LB cases were reported in central and northeastern France every year between 2016 and 2019 ( 23 ). While we cannot exclude the possibility of bias due to the uneven distribution of doctors willing to participate to this survey, tick collection studies carried out in the various regions mentioned showed a high rate of infection of ticks with Borrelia burgdorferi sensu lato (Elisabeth Ferquel personal communication), which is in agreement with our observations.…”

Spatial and seasonal determinants of Lyme borreliosis incidence in France, 2016 to 2021

“…Therefore, we assumed the COVID-19 lockdowns did not affect the overall spatial and seasonal patterns of human encounters with ticks throughout 2020. In addition, in 2020, 87.6% of LB cases reported their tick-bite in their department of residence, consistent with previous years (84.4-88.8% in 2016-19) [33].…”

“…Therefore, we assumed the COVID-19 lockdowns did not affect the overall spatial and seasonal patterns of human encounters with ticks throughout 2020. In addition, in 2020, 87.6% of LB cases reported their tick-bite in their department of residence, consistent with previous years (84.4–88.8% in 2016–19) [ 33 ]. For prediction and model validation, the quarterly proportions of human tick bite frequency were processed for 2020 and 2021 separately at the departmental level, displaying similar spatial patterns compared with previous years.…”

Temporal dynamics of antibody level against Lyme disease bacteria in roe deer: Tale of a sentinel?

“…The seasonality of tick abundance often shows two peaks in temperate regions, one in spring and another in autumn (Hauser et al, 2018 ; Schulz et al, 2014 ). The highest risk of exposure for humans corresponds to the spring peak of tick density (Fu et al, 2021 ), when roe deer samples for serology may be difficult to get. Detecting the density of active ticks positive for Bbsl in winter via roe deer serology could nevertheless be useful if those are related to the acarological risk, that is the density of infected ticks in spring.…”

“…However, trends at large geographic scales remain difficult to assess. Surveillance systems for LD are heterogeneous across countries, with human LD being classified as a notifiable disease in some countries, while in others, like in France, the data collection is based on both a network of volunteer doctors and hospitalization data (Fu et al, 2021 ; Smith et al, 2006 ; Van Den Wijngaard et al, 2017 ; Vandenesch et al, 2014 ). LD risk can be also assessed by tick sampling using cloth dragging to track apparent tick densities followed by polymerase chain reaction (PCR) to search for Bbsl DNA in ticks, but this approach has limits.…”