More and more testing is being performed on food products worldwide with every year that passes. Despite the advent of HACCP – and the virtual abandonment of quality control testing as a means of ensuring food safety and quality – ever lengthening supply chains and consumer concerns about safety have driven a steady increase in the volume of testing being done. But the results produced by all this activity are only as good as the methods used to obtain them. A fact brought into sharp relief this week by an announcement from the Food Standards Agency (FSA) in the UK.
Earlier this year, the Agency instructed the Bart Ingredients Company to recall a batch of ground cumin, in which routine allergen testing had detected the presence of undeclared almond. Almond protein is a recognised food allergen and is a health risk to people with sensitivity to almonds. It is among the 14 allergens that must legally be declared on food labels. But the company later challenged the findings and produced analytical results of their own that contradicted initial tests, so further analysis was commissioned from a third party lab.
That analysis eventually confirmed that almond protein was not present. What was present was a spice called mahaleb, derived from a plant related to the almond and reportedly sometimes handled in the cumin supply chain. Mahaleb is not a recognised allergen, so there were no longer any grounds for recalling the product. The Agency was left with no choice but to announce that the recall had been rescinded. This rather embarrassing climb-down seems to be the result of inadequate test methodology rather than error. The ELISA-based protein detection method used to test for almond protein apparently cannot tell the difference between almond and the very similar mahaleb. Research to validate a more discriminating method is now under way.
This story has some wider implications for the food industry. The original recall was just one of a number initiated on both sides of the Atlantic where undeclared allergens were detected in spices. Others in the UK involved almond in paprika, while in North America peanut protein was detected in ground cumin, leading to widespread recalls of foods using the cumin as an ingredient. At the time, the recalls gave rise to speculation about potential adulteration of high value food products with cheaper substitutes – food fraud in other words. But now it is worth asking just how reliable were the test results they were based on. Recalls and allegations of adulteration undermine consumer confidence and can ruin businesses completely. So they must be based on accurate and reliable evidence. Unfortunately, no existing test method is 100% reliable – just ask the FSA – and results can mislead as they did in this case. But there are other useful sources of information. Intelligence from people on the ground at different points in the supply chain is one such source. Though often undervalued, the observations of individuals with real local knowledge can give early warning of problems and help to inform testing. Any food business with access to a network of this kind can be a lot more confident that its ingredient supplies won’t contain any unwelcome surprises.