Food colours

Since 2009, EFSA has been reviewing the safety of all food additives which had been approved for use in the EU up until that point. Many additives were initially authorised for use a long time ago and in some cases new scientific data have since become available. Food colours were re-evaluated first as they were among the first food additives to be authorised for use in the EU.

EFSA’s expert Scientific Panel which deals with food additives, the ANS Panel, has started re-assessing all of the permitted food colours (45 in total), giving priority to those synthetically produced and later to those obtained from natural sources. The re-evaluation of previously authorised food colours is scheduled to be completed by 2015.

Colours used as food additives may also be authorised for use as feed additives. The evaluation of the safety of food additives and feed additives is carried out by different Scientific Panels at EFSA, with different data requirements due to the separate regulatory frameworks in the two areas. For example, the purity requirements in the food and feed areas can differ considerably. However, EFSA’s Panels co-ordinate their scientific work to ensure consistency in their risk assessment approaches and consideration of the available scientific information within these fields.

EFSA’s experts released the Statement after reviewing new data on the colour Allura Red AC (E 129) and five other chemically related colours, known collectively as ‘sulphonated mono azo dyes’. The ANS Panel has established Acceptable Daily Intakes (ADI) for these colours following previous evaluations. (The ADI is the amount of a substance that people can consume every day over the course of a lifetime without any appreciable risk to health.) In its 2013 Statement, the Panel reconfirmed that there is currently no reason to revise the ADI for Allura Red AC (E 129).

EFSA recommended, however, that new tests be carried out to address uncertainties related to the possible genotoxicity of Allura Red AC. Genotoxicity is the ability of a substance to damage DNA, the genetic materials of cells. This recommendation also applies to the other ‘sulphonated mono azo dyes’ in this group: Amaranth (E 123), Ponceau 4R (E 124), Sunset Yellow FCF (E 110), Tartrazine (E 102) and Azorubine/Carmoisine (E 122).

Except for Amaranth (E 123) which is not used in animal feed, these colours have also been or are in the process of being evaluated for use as feed additives. In 2015, EFSA concluded that Allura Red AC is not genotoxic for dogs and cats. New test data demonstrate that Allura Red AC neither damages the DNA of individual cells (comet assay) nor shows other evidence of genotoxicity (in-vivo micronucleus test). EFSA experts extrapolated safe dietary levels for dogs and cats from available toxicity studies. They calculated the highest safe dietary concentration of Allura Red AC to be 370 mg/kg complete feed for dogs and 308 mg/kg complete feed for cats.

• Statement on Allura Red AC and other sulphonated mono azo dyes authorised as food and feed additives (2013)
• Scientific opinion on the safety of Allura Red AC in feed for cats and dogs (2015)

Genotoxins are substances that can damage DNA, the genetic material of cells, and result in mutations or cancer. Genotoxicity is one of a number of factors that scientists assess when evaluating the safety of additives and other substances added to food.

Following a review of the data, EFSA’s experts considered that these azo dyes could share a pattern of effects that deserve further investigation. However, currently the overall weight of evidence suggests that they are not genotoxic.

Weight of evidence describes the consideration, where there is uncertainty, of information that either supports or opposes a particular outcome and then making a decision based on the most convincing evidence.

People make personal weight of evidence decisions all the time in their daily lives. In risk assessment, the type of evidence considered is highly technical and often difficult to understand for a non-scientist. However, there are well-established steps in the analysis and evaluation of scientific information that help scientists to weigh up all the evidence and make decisions relying on their scientific knowledge and expertise.

As part of the safety evaluation of food additives such as colours risk assessors like EFSA establish, when possible (i.e. when sufficient information is available), an Acceptable Daily Intake (ADI) for each substance. The ADI is the amount of a substance that people can consume every day over the course of a lifetime without any appreciable risk to health.

The ADI is generally derived by looking at the highest intake level at which substances do not cause harmful effects in animal experiments and applying a safety factor (typically of 100) to account for differences between humans and animals. This means that even if people exceed the ADI for a certain substance, this will not necessarily cause negative health effects.

In the case of food additives, a substance used in an additive is ingested directly via food. For feed additives for food-producing animals (livestock), consumers can be exposed indirectly. Residues of the substance may be present in such animal-derived products as meat, eggs or dairy products. Regulators set limits for feed additives called maximum residue levels (MRLs) to ensure that consumer exposure to the residues of these products is below the Acceptable Daily Intake for consumers.

The safety evaluation of feed additives also takes into consideration potential adverse effects of these substances on the target animal (i.e. the animal that consumes the feed additive directly). As well as food-producing animals, target animals could be pets and other domesticated animals (for example, horses). Specific tolerance studies are performed on the target animals to establish a safety indicator called a ‘margin of safety’ in case the additive is consumed at higher levels than recommended: the higher the margin of safety, the lower the risk for the animal.

Like other food additives, the azo dyes referred to in EFSA’s June 2013 statement can be used in a range of foodstuffs, including soft drinks, bakery products, desserts, sauces, seasonings and confectionery. European Union legislation specifies in which foods they can be used and the maximum amounts that can be added per food type.

Feed use also varies by colour. For example, Carmoisine (E 122), Allura Red AC (E 129) and another colour called Brilliant Blue (E 133) are used or proposed for use with cats and dogs only. On the other hand, the colours Patent Blue V (E 131) and Erythrosine (E127) are authorised as feed additive for a variety of non-food producing animals, including cats, dogs, ornamental fish and reptiles.

In its original evaluations of ten food colours including the group of azo dyes, the ANS Panel based its conclusions on data relating to children’s exposure to these colours from nine EU Member States and adult exposure from the UK population, which is considered to be one of the highest consumers of soft drinks in Europe. The data indicated that both average and high-level consumption levels are generally considerably higher for children than for adults.

Since then, new data have emerged on the actual levels of most of these colours found in food. Based on these data, EFSA is currently revising its assessments of human exposure to these original ten colours. These re-assessments are scheduled for completion by November 2014. Anyone interested in following the development of these assessments can check their progress on EFSA’s online Register of Questions.

• EFSA’s Register of Questions

Several factors are taken into account to calculate exposure through food or feed to potential risks. Among them are: diet, age, geographical location and consumption patterns (e.g. habitual or occasional). These factors differ greatly for humans and animals.

Take for instance, the much greater variety of foods making up the human diet. Animals typically consume the same feed on a daily basis whereas people usually vary their diet throughout the day and the week. The food people have for breakfast, lunch and dinner in a single day can already contain a greater variety of ingredients and nutrients than animals eat in a lifetime. Also the average lifetime of most farm animals and domestic pets is shorter than for humans. Therefore, scientists can predict that because exposure of animals to a potential hazard occurs more often and, in comparison with people, for a greater part of their lives, the risk for animal health from a given substance may be higher than for humans.

Under EU legislation in the food and feed areas, EFSA is required to operate under different regulatory frameworks with separate data requirements. The legislation underpinning these frameworks was developed at different times for a variety of historical, political and socio-economic reasons. As a consequence data requirements for the authorisation of substances used in food and feed may differ. 

In spite of these differences, when assessing the safety of substances such as colours and other additives used both in food and feed, EFSA’s Panels aim to ensure consistency in their risk assessment approaches including consideration of the available scientific information within both of these fields.

• Scientists at EFSA further strengthen co-ordination of food and feed additive evaluations

In 2007, the UK Food Standards Agency commissioned a study, carried out by researchers from Southampton University (McCann et al, 2007), which indicated that certain mixtures of five azo dyes (the group in Question 3 above, excluding Amaranth), the colour Quinoline Yellow (E104) and the preservative sodium benzoate (E211) may have a small effect on activity and attention in certain groups of children.

EFSA’s experts evaluated the so-called “Southampton study” in 2008 and concluded that the findings could not be used as a basis for changing the ADIs for the individual additives. One of the reasons for this was that the study looked at mixtures and not individual additives – it was therefore not possible to attribute the effects to any of the individual substances.

As part of its systematic review of food additives which are authorised for use in the EU, EFSA looked at all of the available evidence relating to each of these individual colours and assessed them as a matter of priority due to the concerns which had been raised. These scientific opinions were adopted in November 2009.

• EFSA updates safety advice on six food colours

In 2009, after reviewing all of the available evidence, the ANS Panel reduced the ADI for three of the colours in question, namely Quinoline Yellow (E104), Sunset Yellow FCF (E110) and Ponceau 4R (E124). As a result, the Panel concluded that exposure to these colours through the diet could exceed the new ADIs for both adults and children. In 2014, EFSA established a Acceptable Daily Intake of 4 mg/kg bw/day for Sunset Yellow FCF. Experts also concluded that estimates of consumer exposure are well below this ADI for all age groups.

The Panel did not change the existing ADIs for the three other colours it assessed, Tartrazine (E102), Azorubine/Carmoisine (E122) and Allura Red AC (E129). The Panel also concluded that only children who consume relatively large amounts of food and drinks containing Azorubine/Carmoisine or Allura Red AC could exceed the ADIs for these colours.

Risk assessors sometimes set a temporary ADI when data are insufficient to show that use of the substance is safe over a lifetime but sufficient to show that it can be used safely over the relatively short period of time required to generate and evaluate further safety data. A higher-than-normal safety factor is used when establishing a temporary ADI and an expiry date is established by which time data to resolve the safety issue are needed.

In EFSA’s 2008 opinion and reconfirmed in subsequent assessments, EFSA’s experts concluded that the available scientific evidence, including the “Southampton study”, did not substantiate a causal link between these individual colours and possible behavioural effects.

In the EU food safety system, all additives are currently assessed and authorised individually. To protect consumers, the safety of each additive must be demonstrated on an individual basis before its authorisation for use in foods can be considered by EU risk managers (European Commission, European Parliament and EU Member States).

It is not always possible to assess the safety of mixtures of chemicals when using standard risk assessment approaches. The number of possible combinations of additives and other substances naturally present in the diet is practically infinite, taking into account differences in food composition, consumer food choices and dietary patterns.

The scientific community, including EFSA, is attempting to develop methods for evaluating possible risks from exposure to multiple chemicals in food. For example, the Authority has made important headway in the risk assessment of multiple pesticides and contaminants for humans, and multiple pesticides for bees. In July 2013, EFSA published its first major report on combined exposure to multiple chemicals and will continue to contribute to scientific advances in this area.

• News story: Assessing ‘chemical mixtures’ – current approaches and future priorities

EFSA will continue to provide independent scientific advice on colours used in food and feed. For Allura Red AC (E 129) and the other five sulphonated mono azo dyes, EFSA’s scientific experts have recommended new studies to investigate the potential genotoxicity of these substances using the most recent and validated experimental protocol. Based on the results, EFSA’s experts will, if necessary, reconsider the existing ADIs for these six substances.

Any subsequent follow-up action is the responsibility of EU risk managers who authorise the use of additives in food and feed.