Southampton Six — the UK warning label and what McCann 2007 found

The six colourings, the warning, and where it came from

Walk down a UK supermarket aisle and you'll occasionally see a small line of text on a pack of sweets, an ice lolly, or a brightly-coloured drink: "may have an adverse effect on activity and attention in children". That line isn't marketing copy. It is required by UK law — retained from EU Regulation 1333/2008 — on any food containing one or more of six specific artificial colourings. They are collectively known as the Southampton Six, after the 2007 trial at the University of Southampton that triggered the warning.

Here is the list, with the colour and the kinds of UK products you're most likely to find each one in:

Three of them — E102, E122 and E124 — are azo dyes, a chemistry family that has carried sensitivity questions since the 1970s for a small number of asthmatic and aspirin-sensitive adults. The other three (E104, E110, E129) sit in adjacent dye families. They produce vivid, stable colours at low cost, which is why they're still permitted at all.

What the warning label actually says

The exact required wording, sitting beside the colouring on the back-of-pack ingredient list:

That line is the law in the UK — retained from EU Regulation 1333/2008 with the labelling provisions made operational by Regulation 1129/2011. It applies to all six colourings. It does not apply to sodium benzoate (E211), even though that preservative was in both test mixes used in the McCann trial.

The warning is descriptive, not advisory: it tells you what published research has flagged, it doesn't tell you what to do about it. We treat it the same way — naming the colourings on the products you scan, quoting the FSA-required line, and leaving the choice to you.

The McCann 2007 trial — what was actually measured

McCann et al., published in The Lancet in November 2007, is the study that triggered the warning label. It was funded by the Food Standards Agency and ran in Southampton. The design:

• Around 300 children in two age groups — 137 three-year-olds and 130 eight- to nine-year-olds.
• Double-blind crossover. Each child drank three different drinks across separate weeks. Neither the child, the parents, the teachers, nor the researchers running the assessments knew which drink was which on any given day.
• Mix A: Sunset Yellow (E110), Carmoisine (E122), Tartrazine (E102), Ponceau 4R (E124) plus sodium benzoate (E211).
• Mix B: Sunset Yellow (E110), Carmoisine (E122), Quinoline Yellow (E104), Allura Red (E129) plus sodium benzoate (E211).
• Placebo: a drink that looked and tasted similar but contained none of the colourings or the preservative.

Hyperactivity was measured using a Global Hyperactivity Aggregate score combining parent ratings, teacher ratings (older group), direct observation in classrooms (older group), and a computerised attention task. The headline finding: children given Mix A or Mix B showed a statistically significant rise in hyperactive behaviour compared with placebo, in both age groups.

The honest framing — what the result does and doesn't tell you

The result is real and the trial design is solid. Two things that get lost when the study is summarised:

• It is an average effect across the group, not a per-child guarantee. Some children showed a clear behavioural shift on the active mix. Others showed nothing. Individual response varies, and the trial wasn't designed to predict which children would respond.
• The mixes contain six colourings plus a preservative. The trial cannot tell you which specific additive — or which combination — is doing the work. Sodium benzoate was in both active drinks, so its individual contribution can't be separated from the colourings.

It is also not an ADHD study. McCann et al. measured a behavioural shift. They didn't measure clinical diagnosis. The conclusion the FSA drew, and the line on the warning label, sticks closely to what the data actually says.

What the FSA and EFSA did with it

The two agencies reached different operational conclusions from the same evidence.

The UK Food Standards Agency reviewed the trial and asked manufacturers to voluntarily remove the six colourings from food sold in the UK. Many big brands reformulated through 2008 and 2009 — Smarties, Ribena and most supermarket own-label confectionery dropped them and switched to plant-based pigments (beetroot, paprika, spirulina, anthocyanins). The FSA also pushed for, and the EU adopted in 2010, the mandatory warning label that now appears on any UK product still using them.

The European Food Safety Authority reassessed the colourings between 2009 and 2014. EFSA concluded the McCann data provided "limited evidence" that the mixes affected children's behaviour at the doses tested, and on that basis did not recommend a ban. It did lower Acceptable Daily Intakes for several of the six (including E104 and E110) on the back of broader reassessments. Norway and Austria had earlier introduced national restrictions on some azo dyes; those did not become EU-wide.

The result is the regulatory picture you have today: all six colourings remain permitted across the UK and EU within their ADIs, and the warning label is the compromise.

Where the Southampton Six still appear

UK reformulation was substantial but not total. The colourings still turn up in:

• Some children's confectionery — a portion of pic-n-mix, novelty sweets, and budget-brand boiled sweets. Pink and red mixes are the most common offenders (E122, E124).
• Certain ice lollies and frozen desserts, particularly multi-coloured layered varieties.
• Cake decorations and dessert mixes — coloured icing pens, sprinkles, jelly powders. Bakery decorations are a common source of E124.
• Some imported drinks and snacks, where the original formulation was for a market without the same labelling rule.
• A small number of medicines and supplements, including some children's syrups, where colour is used to differentiate strengths or flavours. The same warning label requirement applies under medicines regulation.

If you don't want them, the warning line and the E-number list make them straightforward to spot. Manufacturers can't hide the warning — it has to sit next to the ingredient.

What it doesn't mean

A few things worth being clear on, because the Southampton Six get summarised badly online:

• It is not a ban. All six colourings are legal in UK and EU food within their ADIs.
• It is not a definitive ADHD link. The trial measured a behavioural shift in a group of children given a mix; it did not measure clinical diagnosis or causation in any individual child.
• It is not a guarantee that every child will react. The effect was an average across the group; individual children varied.
• It is not a verdict on the colourings in isolation. The trial used mixtures, so per-additive attribution is not possible from this data alone.

And — this is the line we stay behind — if you have a specific concern about your child's behaviour, attention or development, that is a conversation for your GP or health visitor, not for a website. Diet is one factor among many.

How NutraSafe surfaces them on a UK barcode

When you scan a UK barcode in our app, we read the ingredient list and check for the six E-numbers. If any is present, we name it explicitly, show the FSA-required warning text, and add it to the additive breakdown alongside any other E-numbers in the product. You don't need to remember the codes — the scan does the spotting.

The barcode scan, the per-product grade, and the public E-number lookup at /e-numbers/ are on the free tier (along with up to 25 food log entries a day). The full processed-food and NRV insights view, AI Coach, AI meal scan, allergen warning detail, fasting features and suspected-triggers analysis sit on NutraSafe Pro.

Frequently asked questions

What are the Southampton Six?

Six artificial food colourings flagged by the UK Food Standards Agency: E102 Tartrazine, E104 Quinoline Yellow, E110 Sunset Yellow FCF, E122 Carmoisine, E124 Ponceau 4R and E129 Allura Red AC. Any UK product containing one or more must show the warning text on the pack.

What does the UK warning label actually say?

"[Name or E-number of the colour(s)]: may have an adverse effect on activity and attention in children." It must appear on the back of the pack alongside the named colouring. The requirement was introduced UK-wide in 2010.

What did the 2007 McCann Southampton study find?

A double-blind, placebo-controlled crossover trial in around 300 children aged 3 and 8–9 in Southampton. Children given a mix of these colourings plus sodium benzoate showed a statistically significant rise in hyperactive behaviour compared with placebo. The effect was an average across the group; individual response varied.

Are the Southampton Six banned in the UK or EU?

No. EFSA reassessed the colourings, lowered some Acceptable Daily Intakes, but did not ban any of them. The mandatory warning label, introduced in 2010, is the regulatory compromise. All six remain permitted within ADI.

Where will I see the Southampton Six on UK labels?

Read the back-of-pack ingredients. Look for E102, E104, E110, E122, E124 or E129, or their chemical names — Tartrazine, Quinoline Yellow, Sunset Yellow, Carmoisine, Ponceau 4R, Allura Red. Many UK brands reformulated post-2008, but they still appear in some children's sweets, certain ice lollies, fruit-flavoured sodas, cake decorations and some imported products.

Related reading

• Are E-numbers bad for you? UK guide
• What are E-numbers? UK guide
• Food additives and children's behaviour
• NutraSafe E-Numbers lookup (330+)
• Food additive scanner UK

Last reviewed: 7 May 2026.