In most epidemiological studies, geographical information is available. This can be used either as a cartographic representation to reinforce study results, or for spatial analysis. Spatial analysis involves assessing whether the health situation under study has a particular spatial structuring, in particular by searching for spatial clusters, or by interpolation. If so, factors associated with such spatial structuring, themselves spatialized, can be sought out, requiring, in addition to technical aspects, an interpretation that distinguishes between concomitance and explanation.
This training course is aimed at anyone wishing to carry out geo-epidemiological studies to assess the spatial heterogeneity of health situations and the associated factors.
Training will cover the basic principles and techniques of cartographic representation, open-access data research, spatial analysis methods and the interpretation of territorial profiles.
Program
- Spatial data typology;
- Graphic semiology in cartography (R).
- Open-access geographic data sources (land use, socio-economic, etc.) ;
- Data harvesting;
- Cartographic data processing.
- Principles of spatial autocorrelation;
- Spatial and temporal cluster analysis (Satscan, LISA).
- Spatial interpolation (IDW, Kriging) ;
- Spatial regression (BYM, GAM).
- Construction (unsupervised) of descriptive profiles of geographical areas / exploratory data analysis (PCA, ACM, CAH, KMeans) ;
- Interpretation of maps and profiles.
Module organization
5 days of 6 hours each (9am-1pm and 2pm-5pm).
Coordinators* & Speakers
Jean Gaudart*, Michael Genin, Martin Grau, Alain Sandoz, Sébatien Gadal, Emmanuel Bonnet, Flore-Apolline Roy, Luka Canton, Mady Cissoko, Paul Taconet, Laurent Lehot, Pierre Schalkwijk, Ibrahima Syll, Jebraiel Ben Raies, Perrine De Crouy-Chanel, Sarah Goria, Tiemoko Galboni.
Course type
Face-to-face and live online.
Module price
Individual: €500
Institutional: €1,500