Session Information
Date: Wednesday, September 25, 2019
Session Title: Epidemiology
Session Time: 1:15pm-2:45pm
Location: Les Muses, Level 3
Objective: The objective of the study was to determine whether PD prevalence exhibit any spatial dependence in order to get hypothesis on PD risk or protective factors.
Background: The cause of PD is unknown. Studying its spatial distribution could provide indication on its pathogenesis.
Method: An extensive case-finding approach was used to find PD cases in the Canton of Geneva. Controls were derived from a sample of geolocated individuals sourced from an ongoing cross-sectional population-based study used to control for the variable population density across the canton, and from a comprehensive population census dataset used to adjust for demographic and socio-economic confounders. All individuals were geolocalized. Age, gender, ethnicity and socioeconomical status were chosen as confounding variables. A logistic mixed-effect model with a random intercept was fit to the populational dataset in order to control for confounding variables. Unit-level statistics were transformed into individual-level data. Getis-Ord Gi* statistics were computed to identify hotspots and coldspots of disease prevalence. Hot- and cold-spots were compared to maps illustrating the spatial distributions of pesticide-associated landcovers, atmospheric pollution and water supply. The results were compared with those obtained with the more commonly used aggregation method.
Results: A total of 1,115 patients were compared with 12,614 geolocated controls and were adjusted for confounders with 474,211 individuals from the census dataset. Individual-level method revealed 5 PD hotspots and 8 PD coldspots. Coldspots were preferentially located in the peripheries of the Canton apart from one centrally located. Hotspots were concentrated into three major clusters near the lake. After adjustment for confounders, clusters shrank drastically but persisted, implying that some external, spatially-dependent factor may impact disease risk. Hotspots did not match with high pesticide exposure, high atmospheric pollution levels and source of water supply. PD spatial distribution obtained with the individual-method was surprisingly different than the one obtained with the aggregation method.
Conclusion: Our study points out a spatial dependence of PD with clusters of high and low prevalence in the canton of Geneva. The clusters did not conform to the dichotomous rural or urban dominance that is sometimes reported in the literature.
To cite this abstract in AMA style:
V. Fleury, R. Himsl, S. Joost, N. Nicastro, I. Guessous, P. Burkhard. Spatial distribution of Parkinson’s disease in the Canton of Geneva, Switzerland [abstract]. Mov Disord. 2019; 34 (suppl 2). https://www.mdsabstracts.org/abstract/spatial-distribution-of-parkinsons-disease-in-the-canton-of-geneva-switzerland/. Accessed October 31, 2024.« Back to 2019 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/spatial-distribution-of-parkinsons-disease-in-the-canton-of-geneva-switzerland/