Banning food colours. Bad science ~ Bad legislation

Philippus Aureolus Theophrastus Bombastus von Hohenheim was born in 1493 and, perhaps not surprisingly, changed his name to Paracelsus. Celsus was a Roman physician who gave the element zinc its name and this pompous Swiss German medic decided to re-name himself  ‘Equal to Celsus’. He is famous for his dictat: ”Sola dosis facet venum”, which translates from the Latin into “The dose alone makes the poison”. In other words, everything is toxic at the right dose and under the right circumstances. Pure Alpine air is highly toxic if a certain dose is injected intravenously. The exact dose remains unknown since no such experiment has ever been done but you get my drift, I hope.

In 2007, a group of researchers from Southampton University published a paper in the prestigious medical journal, The Lancet, showing that a cocktail of food additives (6 colours and 1 preservative) caused hyperactivity and reduced cognitive function in children.  The study was well conducted. It used a placebo (identical in look and taste to the active cocktail) and it used a cross over design, meaning that each child received the active cocktail for a week and at another occasion the placebo for a week. They used two age groups, 3-year olds and a group of 8-9 year olds. The study was analysed by two important committees of experts, the UK Committee on Toxicology and the European Food Safety Authority (EFSA). Both gave their approval.

The consequence of this publication was very serious. The EU introduced a law mandating that a food containing any one of the 6 colours had to be labeled to state that the food contains a colour, which “may have an adverse effect on activity and attention in children”.  The findings were championed by the UK Food Standards Agency (FSA), which pushed for the ban. However, EFSA held the view that the effects of the food colourings on children's behaviour were small and the significance for children's development and education uncertain. The FSA view prevailed within the EU regulatory process and the ban came into effect.

Without doubt, this entire affair calls into question the entire approach that the EU and its agencies to the application of science to risk assessment. Lets just think about linking evidence to policy. No study has ever been completed in which any ONE of the 6 colours, on their own, had any effect on childhood attention deficiency.  Now one could respond to that and say that there may only have been one active colour so lets just label them all. But what about the 7^th additive used, the preservative, benzoic acid. That was excluded from the ban because it reduces the risk of food poisoning whereas colours are just colours. Now suppose the active ingredient was benzoic acid? There is a second flaw to the study, which is relative to the issue of dose as raised by Paracelsus. According to the authors in the paper, the doses were chosen because they represented normal exposure of children to food additives in UK children. Thus they used as their reference intake, the equivalent of three 56g bags of sweets for the younger children and double that for the older ones. Now 4 x 56 is 224 grams per day and at 4 calories per gram, that translates into almost 900 calories. This would mean that UK 8-9 year olds were deriving about 50% of their calories from sweets. That would quite simply lead to wholesale nutritional deficiencies since sweets are just pure sugar with some fat – no minerals or vitamins. Now, at the time this paper was written, the official UK data showed that in this age group, the contribution of all sugar confectionary was just 7%. How could they have got this dose so wrong? Why didn’t the FSA and EFSA not challenge the dose aspect of the paper?

In Ireland, we conduct food intake surveys for different age groups and what makes our approach unique is that we always collect the packaging of all foods consumed. These are photographed and the ingredients list entered on to a database. So we set out to ask if among teenagers and children, there was even one eating occasion among the 118,000 such occasions during the 7-day survey period, when all 7 additives were consumed. The answer was zero. In other words, no child or teenager in Ireland ever, even on just one occasion, consumed all 7 additives at the one time. We then went on to see if the levels of colours used in the Southampton study were ever achieved in our samples. Even at the top 1% of consumers of any colour, the intakes came nowhere near the levels used in the Lancet study.

The ban has come into force and most food producers have complied well ahead of the build up period. The public wasn’t screaming for this and the vast majority doesn’t even know it happened. But it satisfied the worried well who lobby the European Parliament and it was grist to the mill to their anti science approach to risk analysis. But some day the development of bad legislation based on bad science will come back to haunt them.