View Report Details

To Make Recommendations On The Best Practical Procedures To Sample And Test Poultry Flocks For Salmonella

Project Code: B15003

30/04/2005

Background and Objectives

A survey undertaken by the Food Standards Agency (FSA) in 2004 showed that 5.7% of raw chicken meat offered for retail sale to British consumers was contaminated with Salmonella. Although this was a significant improvement over previous surveys, it still means that more than 1 in 20 chickens purchased in the UK have the potential to cause an outbreak of Salmonella food poisoning. The FSA have an obligation to protect consumers from foodborne illness and are therefore keen to further reduce Salmonella contamination of chickens. One way that this can be achieved is to test flocks on farm to determine whether they are infected with Salmonella. When Salmonella-positive flocks are identified, these can be slaughtered separately, at the end of a days processing, so that the risk of crosscontaminating Salmonella-free carcasses is minimised. The purpose of this study was to find an effective way to sample poultry flocks on-farm so that accurate determinations of their Salmonella status can be made. The results would then be used to provide appropriate guidance to the poultry industry.

Approaches

Review of previously-gathered information and published literature for on-farm Salmonella sampling and theoretical appraisal of sampling methods.

In 2003, the FSA funded a study (reference code ZB00023) to gather information on the Salmonella testing and scheduling practices used by the UK poultry industry.

The first part of this study was to update these findings and include additional information on the practices followed by smaller-throughput producers and slaughterhouse operators. The findings from this part of the study showed that, although a large percentage of British birds are tested for Salmonella before slaughter, there was no standard method in use either for the collection of the test samples or their detection in the laboratory. It was found that the sampling and laboratory examination methods used for on-farm Salmonella detections were chosen largely for historical reasons and specific assessments were rarely made regarding the sensitivity of these methods. Since there was no evidence that scientific evaluation of any sampling and analyses method had ever been undertaken, the relative effectiveness of the methods currently in use were unknown. To find out how the sampling and testing practices in use in the UK related to other EU countries and the USA; guidelines for Salmonella testing and scheduling produced in these other countries were obtained and concise comparative summaries were written. Some of the methods used in other countries (e.g. milksoaked gauze drag swabs) were discounted for routine use in the UK on the grounds that they were awkward and impractical for routine use.

A literature review was also undertaken which identified and summarised over 40 key scientific publications. Although a number of publications described experiments that compared how the method of sampling related to Salmonella incidence, it was difficult to decide equivocally which sampling method was most sensitive for the detection of Salmonella in flocks. The difficulties stemmed mainly from differences in detection methods and the lack of a single standardised study which encompassed all routinely-encountered sampling methods.

Practical evaluation of sampling and testing procedures

Chicken houses on four Salmonella-positive commercial farms were visited and samples were collected from the floor litter by boot swabs, sock swabs, and direct sampling. Boot swab sampling involved walking (at least 100 steps) through broiler houses wearing a pair of industrial overshoe protectors moistened with liquid to help pick up litter. Sock swabs were fabricated from cut lengths of Tubegauze sports injury bandage and collected in a similar manner to the boot swabs. Direct sampling involved collecting at least 20 small handfuls of material from throughout the broiler house. Dust was collected as an individual sample, and in combination with boot swab samples. Dust samples were collected by taking pinches of dust from fans, ledges and with the exception feed dispensers and heaters, other areas of the house where dust tended to accumulate. Samples were examined for the presence of Salmonella and the most effective sampling method determined. When the total number of positive detections were counted, direct litter sampling and diluent-moistened boot swabs were equally as effective as sampling methods for the detection of Salmonella. Further statistical analyses of the results revealed that litter sampling and boot swabs were significantly most effective for sampling on two of the farms visited. These methods however were not significantly different than the other sampling methods at the remaining two farms.

What it means and why it’s important

The method that is used to sample broiler houses on-farm for the presence Salmonella in chickens can influence the laboratory analysis results. This study has shown that boot swabs and direct litter sampling were significantly the most effective for sampling Salmonella at broiler houses on two farms. The cost of taking samples by boot swabs or direct litter sampling and the microbiological testing is approximately £9.20 and £8.60 respectively. The costing includes sampling consumables, and all required postage using a standard Royal Mail service.

Since there was no significant difference in either performance or cost between boot swabs or direct litter; either is recommended for the sampling of broiler houses. Both methods allow the cost-effective sampling of broiler houses and subsequent detection of Salmonella-positive flocks.

Improvements to the detection of Salmonella-positive flocks and consequent slaughterhouse scheduling will help prevent cross contamination of chicken carcases during slaughter and processing and is likely to help lower the risk of Salmonella from chickens causing foodborne illness to British consumers.

Main File

Some of the files on this site may be in a format that your computer can't read. However, you can download Readers and Viewers for the following document types below: