Quantitative evaluation of the infection dynamics of bovine brucellosis in Tanzania

Brucellosis is endemic in Tanzania. A cross-sectional study was conducted at 17 cattle farms in agro-pastoral areas in Tanzania to identify risk factors associated with the within-farm prevalence of bovine brucellosis and to quantitatively assess the infection dynamics through disease modelling. Cattle blood sampling and interviews with farmers using a structured questionnaire were conducted. A total of 673 serum samples were screened using the Rose-Bengal plate test (RBPT), and sero-positivity of RBPT-positive samples was confirmed using a competitive enzyme-linked immunosorbent assay. Zero-inflated binomial regression was performed for univariable and multivariable risk factor analyses of within-farm prevalence. Several susceptible-infectious (SI) models were compared based on deviance information criteria, and age-dependent force of infection (FOI) was measured using age-specific prevalence data for the 10 infection-positive farms. Using the diagnoses of cows on the 17 farms, the basic reproduction number, R0, was also calculated. The farm-level prevalence and animal-level adjusted prevalence were 58.8 % (10/17, 95 % confidence interval: 33.5-80.6 %) and 7.0 % (28/673, 95 % credible interval: 5.7-8.4 %), respectively. The risk factor for high within-farm prevalence was introduction of cattle from other herds. A mathematical model with constant FOI showed the annual probability of infection as 1.4 % (95 % credible interval: 1.0 %-2.0 %). The R0 was 1.07. The constant FOI could have been due to the predominant mode of infection being transmission of Brucella from contaminated aborted materials during grazing. Direct purchase of infected cattle could facilitate efficient transmission between susceptible animals through abortion.