Hay K E, Morton J M, Clements A C A, Mahony T J, Barnes T S
The University of Queensland, Queensland Alliance for Agriculture and Food Innovation, St Lucia 4072, Queensland, Australia.
The University of Queensland, School of Veterinary Science, Gatton 4343, Queensland, Australia; Jemora Pty Ltd., P.O. Box 2277, Geelong 3220, Victoria, Australia.
Prev Vet Med. 2016 Jun 1;128:23-32. doi: 10.1016/j.prevetmed.2016.03.017. Epub 2016 Mar 30.
Bovine respiratory disease (BRD) is the major cause of clinical disease and death in feedlot cattle. A prospective longitudinal study was conducted in a population of Australian feedlot cattle to assess associations between factors related to feedlot management and risk of BRD. In total, 35,131 animals in 170 pens (cohorts) inducted into 14 feedlots were included in statistical analyses. Causal diagrams were used to inform model building to allow separate estimation of total and direct effects. Multilevel mixed effects logistic regression models were fitted within the Bayesian framework. The placement of pen water troughs such that they could be accessed by animals in adjoining pens was associated with markedly increased risk of BRD (OR 4.3, 95% credible interval: 1.4-10.3). Adding animals to pens over multiple days was associated with increased risk of BRD across all animals in those pens compared to placing all animals in the pen on a single day (total effect: OR 1.9, 95% credible interval: 1.2-2.8). The much attenuated direct effect indicated that this was primarily mediated via factors on indirect pathways so it may be possible to ameliorate the adverse effects of adding animals to pens over multiple days by altering exposure to these intervening factors (e.g. mixing history). In pens in which animals were added to the pen over multiple days, animals added ≥7 days (OR: 0.7, credible interval: 0.5-0.9) or 1-6 days (OR: 0.8, credible interval: 0.7-1.0) before the last animal was added were at modestly reduced risk of BRD compared to the animals that were added to the pen on the latest day. Further research is required to disentangle effects of cohort formation patterns at animal-level and higher levels on animal-level risk of BRD. Vaccination against Bovine herpesvirus 1 at feedlot entry was investigated but results were inconclusive and further research is required to evaluate vaccine efficacy. We conclude that there are practical interventions available to feedlot managers to reduce the risk of cattle developing BRD at the feedlot. We recommend placement of water troughs in feedlot pens so that they cannot be accessed by animals in adjoining pens. Further research is required to identify practical and cost-effective management strategies that allow longer adaption times for cattle identified prior to induction as being at higher risk of developing BRD.
牛呼吸道疾病（BRD）是育肥牛临床疾病和死亡的主要原因。在澳大利亚的一群育肥牛中进行了一项前瞻性纵向研究，以评估与育肥牛场管理相关的因素与BRD风险之间的关联。共有14个育肥牛场的170个围栏（群组）中的35131头动物纳入了统计分析。因果图用于指导模型构建，以便分别估计总效应和直接效应。在贝叶斯框架内拟合了多层次混合效应逻辑回归模型。围栏水槽的放置位置使得相邻围栏中的动物可以使用，这与BRD风险显著增加相关（比值比4.3，95%可信区间：1.4 - 10.3）。与将所有动物在同一天放入围栏相比，在多天内将动物添加到围栏中与这些围栏中所有动物的BRD风险增加相关（总效应：比值比1.9，95%可信区间：1.2 - 2.8）。大幅减弱的直接效应表明，这主要是通过间接途径的因素介导的，因此有可能通过改变对这些中间因素的暴露（例如混群历史）来减轻在多天内将动物添加到围栏中的不利影响。在那些在多天内将动物添加到围栏中的围栏中，与在最后一头动物添加当天放入围栏的动物相比，在最后一头动物添加前≥7天（比值比：0.7，可信区间：0.5 - 0.9）或1 - 6天（比值比：0.8，可信区间：0.7 - 1.0）添加的动物患BRD的风险适度降低。需要进一步研究以厘清动物层面及更高层面的群组形成模式对动物层面BRD风险的影响。对育肥牛场入场时接种牛疱疹病毒1疫苗进行了研究，但结果尚无定论，需要进一步研究以评估疫苗效力。我们得出结论，育肥牛场管理者可以采取实际干预措施来降低育肥牛场中牛患BRD的风险。我们建议在育肥牛场围栏中放置水槽，使相邻围栏中的动物无法使用。需要进一步研究以确定切实可行且具有成本效益的管理策略，以便为在入场前被确定为患BRD风险较高的牛提供更长的适应时间。
