Severe epidemics of leptospirosis yearly occurred in Thailand since 1997, causing over 1,200 deaths, especially among farmers highly exposed while working in flooded rice fields. Main vectors are wild rodents, spreading the human pathogenic bacteria (spirocheta) in the environment through their urine. Phrae province, which has recorded among the highest incidences, was chosen for a large scale spatial analysis, in collaboration with Phrae provincial Public Health Office. A health geographic database (GIS) was built, using SavGIS, a Geographic Information System (GIS) and Remote Sensing (RS) freeware developed by the Development Research French Institute (IRD), to understand the disease dynamics and assess the villagers' exposure. GIS and Remote Sensing processing have shown a great potential in analyzing vector-borne diseases but have never been applied to leptospirosis. The land use was extracted and vegetation indices computed from a TERRA ASTER satellite image and described around villages. In a second step, this information was compared with the reported disease incidence. Problems encountered due to the accuracy of biomedical and location-based data are detailed. The results are exposed and discussed in a preventive planning perspective.