Biomathematics & Statistics Scotland
The welfare of chickens farmed for meat production is important to many consumers and hence has become an important policy consideration within Government. BioSS staff have been working in collaboration with the Avian Science Research Centre at SAC on studies investigating practical ways to measure leg health in broiler chickens. In a Defra funded project, we compared a new approach to measuring chicken walking style, using a force plate, with the current standard for assessing leg health, the Bristol gait scoring method. Working in both experimental and commercial environments, we examined the relationships between each method and different leg health problems identified subsequently by post mortem of the same birds. In addition, we investigated the power of these methods to estimate prevalences in commercial flocks and showed that both methods had low power for predicting leg health problems identified post mortem when using commercially realistic sample sizes. This conclusion has led to further funding from Defra for a project to examine gait scoring and the relationship to pain, body conformation and post mortem pathology in broiler chickens. The full results were published in the Veterinary Record http://veterinaryrecord.bmj.com/content/168/3/77.full.pdf, which provides an open-access summary veterinaryrecord.bmj.com -Sandilands.pdf and which devoted an editorial veterinaryrecord.bmj.com-editorial to this work.
The relationship between test sensitivity,
when set equal to test specificity, for leg
health tests for individual birds and the
number of samples required to achieve
95% confidence interval widths of 10%
and 20% with 90% power when the
true flock-level prevalence determined
by post mortem examination is 20%.
The gait scoring (GS) method requires
measurements on more birds than the
force plate (FP) method to achieve a
specified width of confidence interval.
Further details from: Sarah Brocklehurst
Article date 2009
