By Chesterton, R. N. and Morris, R. S. and Pfeiffer, D.U. and Tanner, C. M., New Zealand Veterinary Journal, 1989
Description
A case-control study of environmental and behavioural factors influencing foot lameness was undertaken on 62 dairy herds comprising an average of 185 milking cows in Taranaki, New Zealand. 32 case herds were identified as having had at least 10% of the cows lame during the milking season in which the herd was studied, and 30 control herds were selected on the basis that no more than 3% of cows in these herds had been lame per year for at least two years immediately prior to investigation. Each herd was visited at both a morning and an afternoon milking, and 58 risk factors were measured between the time the farmer began to assemble the cows for milking and the completion of milking. Comparison of single variables between case and control herds identified 24 which showed differences. These variables were then subjected to stepwise multivariate logistic regression, and statistically significant variables in this analysis were used to create a tentative "path diagram" of possible "causal web" relationships between the various risk factors and the outcome variable, the lameness prevalence level. Information from a review of the published literature was used to include further variables to the 24 into the initial (or null hypothesis) path model. Logistic path analysis was then used to eliminate non-significant paths from the diagram, leaving 19 arrows joining 13 variables in the final path diagram, compared with 33 joining 20 variables in the initial version. The most influential variables in explaining variation between case and control herds were the average level of maintenance of the track and the degree of patience shown by the farmer in bringing the cows in for milking. Overall, factors associated with the movement of animals to the milking shed explained 40% of the variation (deviance) with regard to the lameness prevalence level. Risk factors associated with characteristics of the milking process explain 24%, and risk factors associated with characteristics of the cows in the herd explain 9.5%. Recommendations are made on management changes which deserve further investigation as ways of reducing lameness problems in dairy herds
A case-control study of environmental and behavioural factors influencing foot lameness was undertaken on 62 dairy herds comprising an average of 185 milking cows in Taranaki, New Zealand. 32 case herds were identified as having had at least 10% of the cows lame during the milking season in which the herd was studied, and 30 control herds were selected on the basis that no more than 3% of cows in these herds had been lame per year for at least two years immediately prior to investigation. Each herd was visited at both a morning and an afternoon milking, and 58 risk factors were measured between the time the farmer began to assemble the cows for milking and the completion of milking. Comparison of single variables between case and control herds identified 24 which showed differences. These variables were then subjected to stepwise multivariate logistic regression, and statistically significant variables in this analysis were used to create a tentative "path diagram" of possible "causal web" relationships between the various risk factors and the outcome variable, the lameness prevalence level. Information from a review of the published literature was used to include further variables to the 24 into the initial (or null hypothesis) path model. Logistic path analysis was then used to eliminate non-significant paths from the diagram, leaving 19 arrows joining 13 variables in the final path diagram, compared with 33 joining 20 variables in the initial version. The most influential variables in explaining variation between case and control herds were the average level of maintenance of the track and the degree of patience shown by the farmer in bringing the cows in for milking. Overall, factors associated with the movement of animals to the milking shed explained 40% of the variation (deviance) with regard to the lameness prevalence level. Risk factors associated with characteristics of the milking process explain 24%, and risk factors associated with characteristics of the cows in the herd explain 9.5%. Recommendations are made on management changes which deserve further investigation as ways of reducing lameness problems in dairy herds
We welcome and encourage discussion of our linked research papers. Registered users can post their comments here. New users' comments are moderated, so please allow a while for them to be published.