Geographical variation of seropositivity to Ehrlichia risticii (equine monocytic ehrlichiosis) of horses in New York state.

A total of 2,579 serum samples from horses in New York state during 1985-1986 were examined for seropositivity to Ehrlichia risticii using the indirect fluorescent antibody technique. Cluster analysis statistical technique was used to group counties according to their estimated EME-disease rate (seropositive proportion of sampled horses). Counties were clustered into 4 groups of different EME-disease rates, representing high (86% seropositive), medium (66% seropositive), medium-low (47% seropositive) and low (6% seropositive) risk regions. The logistic regression statistical technique was used to examine whether the geographic variation of E. risticii seropositivity is due to other EME-risk factors. Risk factors known to be associated with EME (age, breed, sex, month of sample submission) have not contributed to the apparent geographical variation of the disease in New York state. This lack of an association between equine host factors or temporal factors with the geographical variation of EME suggests that conditions necessary for establishment of EME endemic foci involves other factors. The county equine population density and presence of a major equine racetrack were not associated with the geographical variation of the disease in New York state. The high risk regions for EME were located predominately along river valleys, but an explanation of why EME activity varies geographically remains speculative.