Printer-friendly format

Food additives and hyperactive behaviour in
3-year-old and 8/9-year-old children in the community:
A randomised, double-blinded, placebo-controlled trial

Dr. Jim Stevenson and his team of researchers headed by Donna McCann at Southampton University have just completed and published their much-anticipated research on normal children in England given drinks with a mix of coloring plus a preservative. On our research page, it is called the McCann 2007 study, and the full text from The Lancet is also available for those who want to read it.

The study was published online in The Lancet on September 6, 2007.


Artificial colors or a sodium benzoate preservative (or both) in the diet result in increased hyperactivity in 3-year-old and 8/9-year-old children in the general population.

We are happy that the study is getting so much press attention -- it is certainly time to stop ignoring good research showing the link between additives and behavior. However, if you are like me, you also want to know what the study itself says, so here it is, in a (somewhat long) nutshell:

153 three-year-old and 144 eight/nine-year-old children were included in the study.

Artificial coloring and the preservative to be tested were removed from their diet during the time of the study.

The challenge drink contained a preservative and artificial colors in a fruit juice mix. The placebo drink was just a fruit juice mix that looked and tasted the same.

The researchers delivered the juice mixes to the family each week, and neither the children, the parents, or the researchers knew which was which at that time, making it a double-blind-placebo-control study.

Salicylates were not considered and were obviously included in the juice mix since two children could not be included in the study because of an allergy to blackberry juice (one of the salicylates eliminated by the Feingold Program).

The main outcome measure was called a global hyperactivity aggregate (GHA), and it was based on recorded behaviors and ratings of the children by teachers and parents, plus a computer test of attention for the older group.

There were two kinds of challenge drinks: MIX A and MIX B

MIX A contained the following food dyes:

  • Sunset Yellow -- FD&C Yellow #6
  • Carmoisine -- a red dye not used in US
  • Tartrazine -- FD&C Yellow #5
  • Ponceau 4R -- FD&C Red #4 not used in US
MIX A given to the 3-year-olds contained 20 mg of the mixed artificial colors.
MIX A given to the 8/9 year-olds contained 24.98 mg of mixed artificial colors.

MIX B contained the following food dyes:

  • Sunset Yellow -- FD&C Yellow #6
  • Carmoisine -- a red food dye not used in US
  • Quinoline Yellow -- D&C Yellow #10
  • Allura Red -- FD&C Red #40
MIX B given to the 3-year-olds contained 30 mg of mixed colors.
MIX B given to the 8/9 year-olds contained 62.4 mg of mixed colors.

Both mixes contained 45 mg of sodium benzoate as the preservative to be tested.

Note: 20, 30, even 62 mg of food dye may be the amount of dye in a few ounces of sweets (depending on how brightly colored the candy, whether it is colored all the way through or only on the surface, etc.), but remember that children also consume food dye in their cereal, their juice drinks, their mac'n cheese, etc. and are also exposed to it in their toothpaste, their shampoo (through their scalp), and their hand lotion (through their skin). In fact, in 1977 the National Academy of Sciences did a huge study on 12,000 people and determined that most people in the United States ingest up to an average of 327 mg of food dyes per day. The amount children in the UK consume is likely to be close to that.

As far as we know, the reason that they did not use BHA, BHT, or TBHQ, is that these preservatives have already been removed from most food for children in the UK. Possibly, therefore, the children consume much more sodium benzoate than American children. Although this preservative is not eliminated on the Feingold Program, products containing it are marked in the Foodlist so that people beginning the diet may avoid it, and so that people who are sensitive to it may continue to avoid it.

Most children who completed the study drank more than 2/3 of the drinks.

64% of the younger children reported 2 or fewer infractions per week.
66% of the older children reported 2 or fewer infractions per week
Very few children had more than 4 infractions per week.

In recruiting the children, the researchers were careful to find children from all socioeconomic backgrounds, and to make sure that they were a random sample including some hyperactive children and some non-hyperactive children, as is found in the general population. The researchers also used additional controls to avoid placebo effects.

Parents completed a daily diary of juice consumption and compliance, and any bottles of juice not consumed were returned and contents measured to validate parent reports.

Parents rated the children's behavior each week for the seven weeks of the study, and the children were also observed by psychology graduate students given special training. The children were observed 3 times a week for 8 minutes each.

The Conners continuous performance test was used for the older children. It is a computerized test of attention.


For the younger children, both MIX A and MIX B both had a significant effect, but the calculation of MIX B versus placebo was not significant for them because their response to it varied very widely.

For the older children, both Mix A and Mix B showed a significantly adverse effect when analysis was restricted to those children consuming at least 85% of the drinks with no missing data.

The researchers concluded that artificial colors or a sodium benzoate preservative (or both) in the diet result in increased hyperactivity in 3-year-old and 8/9-year-old children in the general population.

Before starting this study, the researchers also took DNA samples of the children, intending to find out whether certain genes make the children more likely to respond to the additives. The results of the genetic analysis will be published in the future.

What's wrong with a little
petrol in your food?

Synthetic food dyes are
made from petroleum.

  Fast Facts on Coloring:
  • Yes - they are made from petroleum
  • They contain lead, mercury & arsenic
  • They may give you cancer
  • They may damage your sperm
  • They may damage your liver
  • They may lower your immunity
  • They may give you an asthma attack
  • They may trigger hives
  • They may make you hyperactive
  • They may damage your nerves

See our research pages and the "Bluebook" research pages for proof. Some of the above research has only been done on animals since such research on humans is not feasible.


  Fast Facts on Benzoates:
  • It forms free radicals which damage cells
  • It reacts with citrus to form benzene
  • It may trigger hives
  • It may trigger an asthma attack
  • It may give you cancer
  • It may give you cirrhosis of the liver
  • It may give you Parkinson's disease
  More on sodium benzoate
Note: The choice is not between dangerous preservatives or NO preservatives. There ARE safer preservatives available.