Study fails to highlight effects of additives in isolated context

Results of the recent University of Southampton food colouring and hyperactivity study are "uninterpretable" and will not help European food safety bosses decide which, if any, of the additives used in the study to ban, according to a leading toxicologist.

Ronald Walker, emeritus professor of food science at the University of Surrey, said the study, which used two drinks containing food colours and the preservative sodium benzoate, was poorly designed.

He added: "I have no idea why they did it like that. The findings are uninterpretable because it's impossible to define which components in the drinks were responsible for the reactions."

The study has prompted calls for the European Food Safety Authority (EFSA) to ban all of the additives in question in the course of its review into E-numbers. But it raised more questions than it answered, said Walker. "I am not disputing the results, but I am surprised that the Food Standards Agency (FSA) [which funded the study] approved its design, given the limited value of the results. Even the placebo drink contained fruit juice, which is a common source of benzoic acid."

In order to make decisions on the safety of individual E-numbers, EFSA had to consider them in isolation as well as in combination, said a spokeswoman, adding, "a further detailed assessment is required." Issues to be studied in depth included "the robustness of the study design and methodology"

Professor Jim Stevenson, who led the study, accepted that more research was needed: "The effects we observed could be an interaction between the colours, just down to one of the colours, or a reaction between the colours and the benzoate. We simply don't know. But additives are consumed in combination and we think it's the mix that has the toxic effect."

While the FSA's food allergy branch head Sue Hattersley defended the study, saying, "children don't consume additives in isolation", FSA board member and former Unilever director Chris Pomfret said: "It would be nice if we could isolate the effects of the different colourants, but clearly we can't."

Researchers found some children given drinks containing combinations of sunset yellow (E110), quinoline yellow (E104), carmoisine (E122), allura red (E129), ponceau 4R (E124) and tartrazine (E102) and sodium benzoate (E211), became "significantly more hyperactive" than those in control groups.

Sarah.britton@william-reed.co.uk