Clustering of fatal fibrinous pneumonia (shipping fever) in feedlot calves within transport truck and feedlot pen groups

Abstract

Feedlot owners often state that shipping fever mortality does not affect calves in a random fashion across the feedlot; instead, mortality can be abnormally high in certain truckloads of calves or in certain pens. However, these apparent “clusters” of disease might be no more than coincidental concentrations of fatal shipping fever cases selected from a truly random distribution of cases throughout the feedlot. The purpose of this study was to distinguish between these two possibilities by critically examining the pattern of fatal shipping fever affecting calves placed in a large beef feedlot. Management and mortality data from 36 339 spring-born calves entering a large commercial beef feedlot in SW Alberta, Canada from 1985 to 1988 were used for the analysis in this study. Once at the feedlot, calves were placed in pens of approximately 300 animals. Truck manifests (which include freight or cargo documentation) and feedlot processing records were used to determine the truck and auction market origin of all incoming calves. Because the prevalence of shipping fever mortality varied dramatically among years, an analysis was performed on each of the 4 years separately. To determine whether clustering occurred within the transport truck, tests of homogeneity of binomial samples were run on the truckloads of calves making up each individual pen. To determine whether clustering occurred within a pen, a test for homogeneity of binomial samples was run within each year using the proportion of mortality due to fibrinous pneumonia in each pen; the intracluster correlation coefficient was used to correct for the nested effect of truck within pen. When the incidence of fatal shipping fever was high (greater than 2%), the disease clustered within truckload groups of calves and also, one year, within pens. Further work is necessary to determine whether contagious or non-contagious factors are responsible for the clustering that was documented.