Livestock production losses attributable to brucellosis in northern and central Tanzania: Application of an epidemiological-economic modelling framework
Ângelo Mendes, Daniel T. Haydon, William A. de Glanville, Rebecca F. Bodenham, AbdulHamid S. Lukambagire, P. Johnson, Gabriel Shirima, Sarah Cleaveland, Emma McIntosh, Nick Hanley, Jo E. B. Halliday
Abstract
Livestock brucellosis is an endemic disease in many low-resource settings. Despite its widespread distribution, little is known about the scale of economic impacts caused by the disease. This study aimed to develop an integrated epidemiological-economic modelling framework to estimate production losses attributable to livestock brucellosis, using Tanzania as a case study. Data on livestock production and prevalence of exposure to Brucella spp. were obtained from surveys conducted in northern and central Tanzania between 2013 and 2019. A clustering algorithm was applied to classify households into pastoral and non-pastoral production systems. A Bayesian latent-class analysis model was applied to derive livestock brucellosis prevalence estimates. A herd-growth model was used to estimate production losses attributable to brucellosis. A total of 1,541 households (384 classified as pastoral and 1,157 as non-pastoral) contributed data on livestock production or prevalence of exposure to Brucella spp. The median (95% uncertainty interval, UI) individual-level brucellosis prevalence in cattle, sheep, and goats was 5.1% (3.4-6.9), 1.3% (0.1-3.0), and 2.5% (0.3-4.8) in the pastoral system, and 0.7% (0.1-1.6), 1.6% (0.2-3.8), and 2.5% (0.3-4.9) in the non-pastoral system, respectively. The median (95% UI) annual losses attributable to brucellosis in cattle, sheep, and goats, per infected animal, were 74.4 (26.2-211.7), 9.7 (3.4-23.1) and 10.6 (3.7-25.0) international dollars (int. $) in the pastoral system, and 62.3 (16.8-228.6), 6.3 (1.8-17.1) and 7.0 (2.2-17.9) int. $ in the non-pastoral system, respectively. Household-level losses were equivalent to 4.4% (2.1-8.8) and 0.6% (0.2-1.6) of the median (95% UI) livestock-derived income in the pastoral and non-pastoral systems, respectively. This study did not capture the system-wide impacts of brucellosis, including on human health. The estimated losses are only a part of the full societal economic impact of the disease. These results can be used to inform cost-benefit analyses of potential interventions and guide policy development for brucellosis control.