Identification of geographic factors associated with early spread of foot-and-mouth disease

Objective - To explore whether early analysis of spatial data may result in identification of variables associated with epidemic spread of foot and mouth disease. Sample Population - 37 farms with infected cattle (ie, case farms) reported within the first 6 days of the 2001 Uruguayan foot-and-mouth disease epidemic. Procedure - A georeferenced database was created and retrospective analysis was performed on case farm location in relation to farm density, cattle density, farm type (ie, beef vs dairy cattle production), road density, case farm distance to the nearest road, farm size, farm ownership, and day of infection. Mean or median results of 1 to 3 day versus 4 to 6 day spatial data were compared. Spatial-temporal associations were investigated by correlation analysis. Results - Comparison of mean or median values between the first 3 days and days 4 to 6 of the epidemic and results of correlation analysis indicated a significant increase in road density, cattle density, and dairy cattle production and a significant decrease in farm size and case farm distance to the nearest road that developed over time. A route that linked most case farms by the shortest possible distance and also considered significantly associated variables was created. It included 86.1% of all case farms reported by 60 days into the epidemic. Conclusions and Clinical Relevance - Epidemic direction can be assessed on the basis of road density and other spatial variables as early as 6 days into an epidemic. Epidemic control areas may be more effectively identified if local and regional georeferenced data are considered.