Population-level effects of risk factors for bovine respiratory disease in Australian feedlot cattle

Abstract

Results obtained from a nationwide longitudinal study were extended to estimate the population-level effects of selected risk factors on the incidence of bovine respiratory disease (BRD) during the first 50 days at risk in medium-sized to large Australian feedlots. Population attributable fractions (PAF) and population attributable risks (PAR) were used to rank selected risk factors in order of importance from the perspective of the Australian feedlot industry within two mutually exclusive categories: ‘intervention’ risk factors had practical strategies that feedlot managers could implement to avoid exposure of cattle to adverse levels of the risk factor and a precise estimate of the population-level effect while ‘others’ did not. An alternative method was also used to quantify the expected effects of simultaneously preventing exposure to multiple management-related factors whilst not changing exposure to factors that were more difficult to modify. The most important ‘intervention’ risk factors were shared pen water (PAF: 0.70, 95% credible interval: 0.45–0.83), breed (PAF: 0.67, 95% credible interval: 0.54–0.77), the animal's prior lifetime history of mixing with cattle from other herds (PAF: 0.53, 95% credible interval: 0.30–0.69), timing of the animal's move to the vicinity of the feedlot (PAF: 0.45, 95% credible interval: 0.17–0.68), the presence of Bovine viral diarrhoea virus 1 (BVDV-1) in the animal's cohort (PAF: 0.30, 95% credible interval: 0.04–0.50), the number of study animals in the animal's group 13 days before induction (PAF: 0.30, 95% credible interval: 0.10–0.44) and induction weight (PAF: 0.16, 95% credible interval: 0.09–0.23). Other important risk factors identified and prioritised for further research were feedlot region, season of induction and cohort formation patterns. An estimated 82% of BRD incidence was attributable to management-related risk factors, whereby the lowest risk category of a composite management-related variable comprised animals in the lowest risk category of at least four of the five component variables (shared pen water, mixing, move timing, BVDV-1 in the cohort and the number of animals in the animal's group-13). This indicated that widespread adoption of appropriate interventions including ensuring pen water is not shared between pens, optimising animal mixing before induction, timing of the animal's move to the vicinity of the feedlot, and group size prior to placing animals in feedlot pens, and avoiding BVDV-1 in cohorts could markedly reduce the incidence of BRD in medium-sized to large Australian feedlots