Food testing labs in South Africa: what to test, how often and what to ask the lab

Many small food manufacturers send samples to a lab because a customer asked for a certificate, then file the results without reading them. Others test everything every week and still cannot say whether the process is under control. A simple food microbiology testing plan, built on your own products and risks, turns that spend into evidence you can use.

Why you test: verification, not a safety net

Testing finished product cannot make food safe. If one pack in a thousand is contaminated, a handful of samples will almost always miss it. Safety comes from the controls in your process: cooking, chilling, cleaning, hygiene and supplier approval. Microbiological testing is how you check that those controls are working.

That changes the question from “what should we test?” to “what do we need to prove?” A bakery wants to show that its cleaning removes contamination from surfaces after cooling. A ready meal producer wants to show that Listeria monocytogenes is not establishing itself in the chilled packing area. A juice maker wants to show its process and hygiene keep spoilage organisms down over the shelf life. Each of those leads to a different plan.

Choosing tests for your product: indicators and pathogens

There are two broad kinds of test, and a good plan uses both.

Indicator organisms

Indicators tell you about hygiene and process control. They are cheaper, quicker and found more often, which makes them useful for trending.

• Aerobic colony count (often called total plate count): general level of contamination and handling.
• Enterobacteriaceae or coliforms: hygiene after a heat step, and cleaning effectiveness.
• E. coli: faecal contamination, poor hand hygiene or contaminated water.
• Yeasts and moulds: spoilage, especially in baked goods, dairy and acidic products.
• Coagulase-positive staphylococci, including Staphylococcus aureus: contamination from people’s hands and handling.

Pathogens

Pathogen tests answer a specific question for a specific product. Choose them from your hazard analysis, not from a generic list. For example, Salmonella for products with poultry, egg, spices or dry ingredients eaten without further cooking; Listeria monocytogenes for chilled ready-to-eat foods; Bacillus cereus for cooked rice and starch-based products held after cooking; Clostridium perfringens for cooked meat dishes that are cooled and reheated.

Agreeing the specification

Every test needs a specification: what result is acceptable, what result triggers investigation, and what result means the product is rejected. Sources include your customer’s specification, any legal standards that apply to your product, published industry guidelines and your own history. Sit down with your lab and your main customer to agree these before you start testing, and write them into a specification sheet for each product. The lab can tell you what its methods can detect and how results will be reported. A result without an agreed specification is just a number.

Building a food microbiology testing plan: product, surfaces, water and hands

Sample type	Typical tests	What it tells you
Raw materials	Indicators, plus pathogens linked to the material	Whether suppliers meet specification
Finished product	Indicators and relevant pathogens	Whether the process and hygiene controls worked
End of shelf life	Indicators and spoilage organisms	Whether the shelf life claim holds
Food contact surfaces after cleaning	Aerobic colony count, Enterobacteriaceae	Whether cleaning is effective
Environment (drains, floors, under equipment)	Listeria species in ready-to-eat areas	Whether a harbourage site is developing
Water	E. coli, coliforms and other parameters the lab advises	Whether water is fit for use, measured against SANS 241
Hands of food handlers	E. coli, Staphylococcus aureus	Whether hand washing works, and a strong training tool

For environmental monitoring, think in zones: food contact surfaces closest to the product, then nearby non-contact surfaces, then the wider area. Testing drains and floors for Listeria species, rather than only for Listeria monocytogenes, gives you an early warning before the pathogen reaches the product.

How often to test in a small plant

There is no single correct frequency. It depends on the product risk, your volume, your history and your customer’s requirements. A sensible approach is to start more often, build a history, then reduce once results are stable.

• Test more often when a product or process is new, after a change, after a failed result, or when complaints rise.
• Rotate surface swabs so that every food contact surface is covered over a cycle, rather than swabbing the same three spots each time.
• Test water at intervals set by your risk, and whenever the supply, storage tank or treatment changes.
• Use hand swabs during training campaigns and after hygiene problems, not only on a fixed calendar.

An example from a small bakery

A bakery in Gqeberha making cream-filled products and bread agreed this starting plan with its lab and its main retail customer: finished cream products tested every two weeks for indicators and pathogens chosen in the hazard analysis; bread tested monthly for yeasts and moulds at end of shelf life; six food contact surfaces swabbed weekly on rotation after cleaning; borehole water tested quarterly; and hand swabs during each new staff intake. After six months of stable results, the customer agreed to monthly testing on cream products. When one surface swab failed, the bakery went back to weekly testing on that line until three results in a row were within specification.

Your plan should be written into your HACCP verification schedule. If you need help linking testing to the hazard analysis, our HACCP and GMP implementation consulting covers this.

Choosing a lab: SANAS accreditation and the scope

In South Africa, look for a laboratory accredited by SANAS to ISO/IEC 17025. Then go one step further: accreditation applies to specific methods on specific sample types, listed on the lab’s schedule of accreditation. A lab can be accredited for Salmonella in meat and not for Salmonella in spices. Ask for the schedule and check that each test you need, on your type of product, is on it.

Questions to ask the lab:

1. Is this test, on this kind of sample, on your accredited scope?
2. Which method do you use, and what is the detection limit?
3. How will results be reported, in what units, and how quickly?
4. Will you phone us on a presumptive positive before confirmation?
5. Do you supply sterile bags, swabs and cool boxes, and do you collect?
6. How long can samples travel, and at what temperature, before you will not accept them?
7. How long do you keep the sample after testing, in case we need a retest?

Sampling and getting samples to the lab properly

A good lab cannot fix a bad sample. Train the person who samples, and give them a written method.

• Take finished product in its sealed pack where possible. If you must open a pack, use sterile bags and utensils and wash and sanitise hands first.
• For surfaces, use swabs with the neutralising solution the lab supplies, so that sanitiser residue does not kill organisms on the swab and give a falsely clean result.
• Label each sample with product, batch, date, time, location and the sampler’s name, and complete the lab’s submission form.
• Keep chilled samples cold in a cool box with ice packs, keep frozen samples frozen, and get them to the lab within the time it has agreed with you.

From result to decision

Every result should lead to a decision, even if the decision is “within specification, no action”. Compare it with the specification, record the outcome and trend it over time. Rising aerobic counts on a surface that is still within limits are a reason to look at cleaning before it fails.

For pathogens, decide in advance what happens to product. If you are going to act on a positive result, hold the batch until the result is back; once the product has been sold, a positive result means a withdrawal or recall. An environmental Listeria positive calls for deeper cleaning, swabbing around the positive site and repeat testing until it is clear. The microbiological contamination risk assessment template helps you link each test back to the hazard it verifies. An internal food safety audit can check whether results are actually being reviewed and acted on.

What to do this week

• List what you currently test, how often and why. Cross off anything nobody can explain.
• Ask your lab for its schedule of accreditation and check your tests are on it.
• Write or update a specification for each product you test.
• Set up a simple trend sheet for surface swab results.
• Agree in writing what happens to product after a pathogen positive.

If the reports still look like a foreign language, our Introduction to Food Microbiology course for non-microbiologists explains the organisms, the tests and what the results mean for your plant.