Skip to content

0 of 15 lessons complete · 0%

Module 2 · Hazard analysis in depth

Control measures

Lesson 6 of 15 · about 4 minutes

A control measure is any action or activity that can be used to prevent or eliminate a hazard, or reduce it to an acceptable level. At step 6 the team considers which control measures, if any, can be applied to each hazard.

Control measures work in different ways. Some eliminate or reduce a hazard, such as cooking. Some stop growth, such as chilling, freezing, acidity, low water activity and preservatives. Some prevent contamination, such as separation, cleaning and filtering. Some find and remove a hazard, such as sieves, magnets and metal detectors. Some rely on information, such as correct allergen labels and clear cooking instructions.

Codex makes two useful points. More than one control measure may be needed for a specific hazard. To control Listeria monocytogenes, a heat treatment may be needed to kill it in the food, and cleaning and disinfection may be needed to prevent transfer from the environment. And more than one hazard may be controlled by one control measure. A heat treatment can control both Salmonella and E. coli O157.

A control measure must be capable of doing the job. Ask four questions. Does it work on this specific hazard? Is it reliable every time? Can we show that it is working? Does it still work in the worst case, such as the largest batch or the coldest part of the oven?

Know the limits of each control. A metal detector does not reliably find glass or plastic. A sieve stops only what is bigger than its mesh. Cooking removes neither allergens nor every toxin. Washing can remove soil, but do not assume it removes all pathogens from ready-to-eat produce.

Not every control belongs at a CCP. Codex says that for some significant hazards, control may be achieved by GHPs, some of which target a specific hazard. Examples are cleaning equipment to control contamination of ready-to-eat foods with Listeria, or preventing food allergens from being transferred from one food to another. In other cases, control measures need to be applied within the process, at CCPs.

Annex I of the Codex text compares the two. GHPs are general conditions and activities, usually not specific to one hazard, although some do target a hazard. Their validation is generally not carried out by the business itself. Data from authorities, published science or manufacturers is adequate, as long as the business can show that it follows the instructions. Control at a CCP is specific to a process step and a significant hazard. It has critical limits, validation, monitoring that ensures the limit is met, and pre-determined corrective actions for the product.

If no control measure exists at any step for a significant hazard, the product or process should be modified. In our example, the team first wanted to use a supplier's ready-made mash. The specification did not show how allergens were controlled, so there was no way of knowing what the mash might contain. The team changed the supplier.

Finally, check that a control does not create a new hazard. A sanitiser can leave a residue if it is misused. A test piece for a metal detector can end up in the product if it is not controlled. Add these to the hazard analysis.

Key points

  • One hazard may need several controls; one control may cover several hazards
  • Test whether a control works on this hazard, every time, in the worst case
  • Some significant hazards are controlled by GHPs, others at CCPs
  • If no control exists, change the product or process
Check your understanding

A jar filling line has a metal detector. Which hazard does it leave uncontrolled?