Modelling Human Brucellosis Based on Infection Rate and Vaccination Coverage of Sheep and Goats.

In this study, the vaccination coverage, serological sampling and infection rate of sheep and goats were evaluated as predictors for the modeling of human brucellosis in Greece. The human brucellosis disease frequency per local regional unit (RU) varied significantly (RR90) among consecutive years. The notification rate was higher (p < 0.001) in the RUs with implementation of vaccination in sheep and goats (vaccination zone-VZ) with a median of 1.4 (IQR 0.0-3.1) compared with the RUs of the eradication zone (EZ) with a median of 0.0 (IQR 0.0-0.0). In VZ, the increased frequency of human cases was associated with delayed vaccine administration (estimate: 0.14 (0.04; 0.29), p = 0.03) and higher vaccination coverage of the animals (estimate: -0.349 (-0.72; -0.07), p < 0.01). However, the flock sampling rate was highly heterogenous among RUs (IQR 10.56-52.93), and inconsistent within RUs throughout the period of the study 2013-2017 (p = 0.001), limiting the reliable estimation of the infection rate in livestock and the design of an integrated One Health model for human disease.