Find out how to take samples from processed ABPs and find laboratories to test them, and what to do if one of your samples fails a test.
Applies to England, Scotland and Wales
If you process animal by-products (ABPs), such as animal carcasses or kitchen waste, you may need to test samples of your products to check they are safe to use. You need to submit samples for testing if you operate:
- an ABP processing facility (often known as a rendering plant)
- a composting or anaerobic digestate facility
- a petfood factory
- a fertiliser factory (if you use ABPs in your fertiliser)
- a biodiesel plant (if you use ABPs to make your biodiesel)
- a blood processor
- a milk processor
For more information, see the guidance for the animal by-products industry.
What bacteria you need to test for and when to take samples
The testing requirements for different types of processing facility are summarised in the following table:
|Salmonella||Enterobacteriaceae||Clostridium perfingens||Escherichia coli (E.coli) or Enterococcaceae|
|Biogas, compost or manure processing facility||Yes||Yes|
|Facilities applying for approval for a new (method 7) processing method||Yes||Yes||Yes|
|Any other ABP processing facility (rendering plants, pet food and fertiliser factories, blood processors, biodiesel plants, milk processors)||Yes||Yes|
The Animal and Plant Health Agency (APHA) may ask you to carry out additional tests as one of your operating conditions.
For all bacteria except Salmonella, you should take samples immediately after you finish processing your ABP material.
Salmonella tests are used to check that your products are not re-contaminated in storage after processing. APHA will discuss when you should take samples for Salmonella testing, when they approve your plant.
How to take samples
First sterilise and clean any equipment that you will need to use, such as scoops or dippers. You can choose any sterilisation technique or follow this method:
- Wash with detergent.
- Rinse with water.
- Boil in a fish kettle for 30 minutes.
Making sure samples are representative
To make sure samples are representative:
1) Mix the material thoroughly before you collect samples.
2) Use this table to find out how many sub-samples you need to collect and what size each sub-sample should be.
|Volume of material to be sampled||Number of sub-samples||Minimum sub-sample volume (for liquid material)||Minimum sub-sample weight (for solid material)|
|Up to 1,000 cubic metres||12||0.33 litres||330 grams|
|1,000 to 1,999 cubic metres||16||0.25 litres||250 grams|
|2,000 to 3,600 cubic metres||22||0.18 litres||180 grams|
|Over 3,600 cubic metres||30||0.13 litres||130 grams|
3) Collect the correct number of sub-samples from different parts of the mixture and mix them together thoroughly in a clean polythene bag or container.
4) Take a single final sample of 0.5 litres (for liquid material) or 500 grams (for solid material) from the bag or container.
5) Place the final sample in a clean container suitable for transport to the laboratory.
Representative sampling isn’t possible for some solid products such as dog chews, which can not be easily mixed together. In this case select a sample at random from among the products instead. The sample should be at least 500g.
Taking and storing a backup B sample
The sample that you send to the laboratory for testing is known as the A sample.
When you take the A sample, you must take a second B sample in exactly the same way.
Seal the B sample in a container and place it in a fridge that is not used for storing food.
You will need to send the B sample to the laboratory if the laboratory can’t test your A sample for any reason (eg if it is lost or damaged).
If your A sample fails the test (the laboratory results show it contains too high a level of the bacteria it was tested for) you can not send your B sample in place of the A sample. You must tell APHA you failed the test.
You can destroy your B sample if your A sample passes the test (laboratory results show it contained an acceptable level of bacteria).
How to find a laboratory to carry out tests
You must send your samples to a laboratory accredited by the United Kingdom Accreditation (UKAS).
You can find a list of all accredited laboratories by searching for ‘Defra animal by-products’ on the UKAS website.
Laboratories are accredited on a test-by-test basis so you may have to send samples to more than one laboratory if you need to test for more than one type of bacteria.
Laboratories do not need to be separately approved by Defra.
Selecting a test method (laboratories)
In order to comply with UKAS requirements, your selected laboratory must use methods that are published in national or international standards where available. The following methods must be used for:
- Salmonella: BS EN ISO 6579:2017
- Enterobacteriaceae: BS EN ISO 21528-2:2017
- Clostridium perfringens: BS EN ISO 7937:2004
- E. coli: BS EN ISO 16649-1 or 16649-2:2001
Reporting salmonella positive results (laboratories)
If you isolate salmonella in any ABP sample, you must send a sub-culture for serotyping to:
Test results must be reported to the ABP operator, who submitted the sample as quickly as possible. In addition, you must email APHA the positive result and give the business name of the ABP approved operator who submitted the sample.
How to interpret laboratory results
The laboratory will divide your sample into 5 sub-samples.
The laboratory will give a result for each sub-sample using the notation: bacteria name, x per y grams
Where bacteria name is the name of the bacteria tested for (for example, Salmonella), and x is the number of colonies of that bacteria counted in y grams of the sub-sample.
Instead of giving a number, the laboratory may also report the result for a sub-sample as:
This means the laboratory wasn’t able to detect any bacterial colonies in that sub-sample.
How much bacteria your samples can contain
Your sample will fail if any sub-sample contains any Salmonella colonies
A sample of raw pet food will fail if:
- any of the sub-samples contains more than 5,000 Enterobacteriaceae colonies per gram
- 3 or more sub-samples contain more than 500 Enterobacteriaceae colonies per gram
A sample of any other ABP product (including processed pet food) will fail if:
- any of the sub-samples contains more than 300 Enterobacteriaceae colonies per gram
- 3 or more sub-samples contain more than 10 Enterobacteriaceae colonies per gram
Your sample will fail if any sub-sample contains any Clostridium perfingens colonies
Escherichia coli (E.coli)
Your sample will fail if:
- any of the sub-samples contains more than 5,000 E.coli colonies per gram
- 2 or more sub-samples contain more than 1,000 E.coli colonies per gram
No UK laboratory is currently accredited to test for Enterococcaceae, so most compost and biogas facilities test for E.coli instead. But you can send your samples to a laboratory in another EU state that is accredited to test for Enterococcaceae.
Your sample will fail if:
- any of the sub-samples contains more than 5,000 Enterococcaceae colonies per gram
- 2 or more sub-samples contain more than 1,000 Enterococcaceae colonies per gram
What to do if your sample fails a test
If one of your samples fails a test you must tell APHA immediately.
You will need to investigate what caused the failed test and explain to APHA how you plan to fix the problem. APHA will need to agree that your solution addresses the problem.
You will have to record any failed test, and the actions taken as a result.
How long to keep test records
You must keep test results for at least 2 years in written or electronic form, and present them to APHA officers if you are asked.