Illegal dyes study published
Tuesday 12 December 2006
The Agency has published details of FSA-funded work undertaken by local authorities on imported foods checking for a range of illegal dyes. This has shown that the vast majority of foods tested comply with the law.
Food was analysed for Sudan I to IV, Para Red, Rhodamine B, Orange II, Red G, Butter Yellow and Metanil Yellow. These dyes are allowed to be used in industry, for example for colouring solvents, oils and waxes, but should not be added to food. However, in some cases inadvertent contamination of some food products has been identified.
A total of 149 local authorities across the UK carried out tests on food between December 2005 and March 2006. Sampling was carried out randomly on spices, sauces and oils at a variety of premises including ports, warehouses and retailers.
Out of 893 samples, six were found to contain illegal food dyes. Sudan I was found in one sample of tandoori masala mix from an unknown country of origin. Sudan IV was found in three samples of palm oil from Ghana, and Orange II was found in a ground red chilli product and pepper soup mix, both from Nigeria.
Where illegal dyes were identified, the FSA worked with local authorities to ensure the contaminated products were traced and that appropriate action was taken to withdraw any such products from the market. The affected materials were then disposed of in accordance with approved protocols.
Bixin – a colour permitted under EC legislation for specified foods, not including spices – was detected in 18 of the samples analysed. Bixin is derived from annatto seeds, which are sometimes added (ground) for their flavouring properties as an ingredient in food. We are currently considering the possible distinction of these two uses of bixin and annatto, as a colour and as a flavouring. The issue will be discussed with relevant stakeholders and with the European Commission.
The science behind the story
Sudan I has been shown to cause cancer in laboratory animals and may contribute to the development of cancer in people. The other dyes in the survey have a similar structure and are thought to pose a similar risk. However, at the levels found there is no immediate risk of illness.