WELFARE OF BROILER CHICKENS
 
In the UK alone, over 800 million birds a year are killed for their meat every year and the welfare of these birds is of increasing public concern. Particular criticisms have been raised about the stocking density (degree of crowding) the birds experience. We recently conducted a large scale study on commercial broiler farms in the UK and Denmark with the aim of finding out whether and how the welfare of broilers was related to stocking density. Somewhat surprisingly, stocking density itself turned out to be less important to the physical health and mortality of the chickens than the poorer air and litter quality that can result from increased stocking density if no steps are taken to control them. Some commercial producers seemed to be much better at controlling the environment of their birds than others. This does not mean that stocking density has no effect on bird welfare. Rather, it means that it would be possible to reduce stocking density without necessarily improving the welfare of the birds at all. Genuine improvements in broiler welfare will therefore come through improving all aspects of the birds' environment.
 
Dawkins, M.S., Donnelly, C.A. & Jones, T.A. (2004) Chicken welfare is influenced more by housing conditions than by stocking density. Nature 427: 342-344.http://users.ox.ac.uk/~abrg/papers/dawkins/Nature.pdf
 
 
Jones, T.A., Donnelly, C.A. & Dawkins, M.S. (2005) Environmental and management factors affecting the welfare of chickens on commercial farms stocked at five densities. Poultry Science 85: 1156-1165. http://users.ox.ac.uk/~abrg/papers/dawkins/ps.pdf
 
 
 
FREQUENTLY ASKED QUESTIONS
 
·      Does your study show that stocking density doesn't matter to broiler chicken welfare?
o      No. We did find that stocking density had no significant effect on some of the most obvious measures of broiler welfare, such as mortality and leg defects. But we also found that other measures of welfare, such as walking ability, were affected by stocking density. At higher stocking densities, growth rate was reduced, birds jostled each other more and walked less. However these effects of stocking density were overshadowed by much larger differences between producer companies. Some companies coped with higher stocking densities much better than others because they were able to control the environment within the houses and ensure good air and litter quality even at higher stocking densities. Our results do not show that stocking density has no effect on chicken welfare. They do show that other (environmental) factors have even more effect.

·      Aren't you encouraging producers to stock at ever higher densities?
o      Certainly not. We did find some adverse effects of stocking density, particularly at the higher two stocking densities 42 and 46kgm2. We do not know what happens above this. On the other hand, we caution against the oversimplistic view that bird welfare must necessarily be better at low densities. It would be possible for a farmer to stock at low density and still have poor bird welfare. Other factors, particularly air and litter quality need to be ensured.

·      Did you rely on data collected by the poultry producers themselves?
o      Almost all the data used we collected ourselves. We had our own data loggers tracking temperature and humidity; we took our own measurements of bird weight, leg & foot health, gait and behaviour; we took our own air and litter and faecal samples; we had our own video records and our own measurements of light levels. We used producers' records of daily mortality but audited these against our own checks of numbers of chicks placed, total mortality reported, numbers of birds arriving at the processing plant and downgrades. We followed each flock to the processing plant and took a random selection of birds off the line for post mortem examination.

·      Isn't your work compromised by having worked with the poultry industry?
o      We could not have done the work without the cooperation of the poultry producers but it was made clear to all companies from the outset that the results would be published, whatever they showed. We also made it clear that we went into the study with no pre-conceived ideas either way as to what the results would be. We paid particular attention to experimental design and made sure that the five different stocking densities were randomly assigned to different houses within a company and that the data were analysed 'blind' ­ i.e. by coding all houses so that the person analyzing the data had no idea of which houses or farms or stocking densities were involved until the analysis was complete. The value of working with producers is that the results are directly relevant to commercial broiler production and cannot be dismissed as 'irrelevant'. There is no way in which the work was compromised by being relevant to real farming.

·      Were you paid by the poultry industry?
o      No. The entire study was funded by Defra. No money was contributed by any of the companies. MD's salary was paid by Somerville College, Oxford.

·      How is this study different from previous studies?
o      Many previous studies have been done on a small scale, by measuring only a limited number of variables or keeping several stocking densities in one house, making it difficult to study enviornmental effects. Our study was done by experimentally manipulating whole houses of birds, measuring many different variables and on such a large scale that it was like a peek into the future: what would happen to bird welfare if stocking densities were reduced on commercial farms? The answer should be a caution to anyone who thought that the quick and easy solution to improving broiler welfare is to reduce stocking densities. Reducing stocking density from very high densities such as 46 or 42 kg/m2 would probably improve welfare but further reductions will have relatively little effect unless other aspects of the environment such as air and litter quality and measures of bird welfare, such as moratlity, leg and foot health are also taken into consideration.

·      What steps need to be taken to improve broiler welfare?
o      What matters is improving bird welfare in practice and the only way this can be achieved is by setting welfare outcomes and making sure they are met. Simply specifying how much space a bird should have or what sort of ventilation should be installed in the houses is not enough. The welfare might not after all be improved, even though well meaning people might hope it would be. It is essential to specify 'good welfare' in terms of outcomes such as lowered mortality and improved leg and foot health. These outcomes should be easily measurable, objective, and cheap to apply on a large scale. Producers could be set welfare goals to achieve (such as particular levels of mortality or hock burn) and the ideal would be that these could be assessed automatically at slaughter. What is now needed is to find the combination of outcome measures that most reliably reflects the conditions experienced by birds during their lifetime.
o      Improving the welfare of broiler chickens by improving their environment may have its limits. Heavy selection for high growth rate and breast meat yield has altered the genetics of broiler chickens to such an extent that it may be impossible to achieve high welfare for them in any environment. One possible future course of action may be to concentrate on breeding a bird that is genetically less prone to welfare problems.