2011 transcript from the FDAs Food Advisory Committee on Food Dye and behaviour

When I find an excerpt from someone trying to change food dye regulations, long before I knew of their harms, I'm especially grateful. They tried. They didn't succeed, but I get to imagine a world where I had a 3, 4 and 5 year old whose mental and physical health wasn't disrupted, eventually turning the whole families life upside down. A little girl who didn't have that awful aggression for 4  months before we connected the dots. 

Excerpt from: Food Advisory Committee meeting transcript, March 30, 2011|

Dr. Michael Jacobson, founder of CSPI

DR. JACOBSON: Good morning. I'd first like to thank the FDA for holding this meeting on food dyes and children's behavior. 

For too long, the agency failed to examine the research, but simply stated flatly that there is, "no evidence that food color additives cause hyperactivity or learning disabilities in children." While this committee is charged with reviewing the scientific evidence on dyes, I'd like to start by emphasizing the legal standard for judging the safety of dyes. 

In implementing the 1960 Color Additives Amendment, the FDA apparently had special concerns about the safety of colorings. The FDA's standard for color additives states that, "safe means that there is convincing evidence that establishes, with reasonable certainty, that no harm will result from the intended use of the color additive." The term "convincing evidence" is not in the definition of safety for non-color additives, preservatives, thickening agents, and the like. Anything short of convincing evidence should disqualify a color additive from being used. 

And as you hear presentations today and tomorrow, I hope you'll keep in mind that benchmark of convincing evidence of no harm. 

The FDA has prepared an exhaustive review of the several dozen studies on the topic. However, the more I read the report, the more I thought it was a terrible indictment of the peer-review system for publishing scientific studies. It would appear that almost every study is so flawed, that it's a wonder that any of them were funded, any of them were published. But I do agree with some of the reports' conclusions. The literature is confusing and inconsistent, with some studies finding an effect of dyes and others not. 

 Also, I agree that dyes are not the underlying cause of behavioral reactions, but rather trigger some pre-existing predisposition in some children. And, third, dyes are not unique in disturbing children's behaviors. Other foods or ingredients have similar effects, but that certainly shouldn't let dyes off the hook, especially because dyes, unlike some of those other triggering ingredients like wheat, eggs, milk, are totally unnecessary additions to the food supply, and indeed are used largely to fool consumers into thinking a food contains fruit, egg, or other natural ingredient that it doesn't. 

While the body of research is inconsistent, a number of studies identify children who clearly were affected by dyes, and I'll just mention a few studies. In the one study sponsored by the Food and Drug Administration itself, Dr. Bernard Weiss, who will talk just after me, and colleagues, tested 22 children who were kept on a diet free of dyes and certain other additives in foods, and then they challenged the children with dyes on certain days. 

 One child reacted dramatically, according to her parents, and a second reacted more moderately. In one of the few studies that used high doses of dyes, 150 milligrams, researchers in Toronto challenged 40 kids, half of whom were considered hyperactive. After eating a diet free of dyes and certain other ingredients, the children were challenged on one day with a mixture of dyes and on another day with a placebo. Compared to the placebo, the dyes decreased the attention span of the hyperactive children, but not the other children. Seventeen of the 20 hyperactive kids suffered impaired performance in a learning test, though other tests didn't detect an effect. 

Finally, a British study tested the effect of dyes on 19 children. Their parents suspected that they were affected by foods and then put the children on restricted diets. In this study, the children were kept on their restricted diets, but then challenged with 125 milligrams of a mixture of four dyes. Seventeen of 19 sets of parents rated their children's behavior as worse, and in several cases, sharply worse, when their children consumed the dyes. 

There's a large body of research, or moderate body of research, on kids with hyperactivity or other behavior issues. In the last few years, there have been some studies that helped clarify that body of research and also extend it.

 First, in 2004, David Schab and a colleague published that meta-analysis of 15 controlled studies in which children with hyperactivity were administered mixtures of most dyes used in the United States. And a meta-analysis, of course, is never definitive because studies never use identical protocols, and the design of some could be flawed. But the Schab-Trinh study concluded that, on the whole, the studies found a clear effect of dyes on behavior, especially in the eyes of parents. The authors characterized the magnitude of the effect as being about one-third to one-half that of children ceasing the use of stimulant drugs; in other words, a significant impact. And that average obscures greater and lesser individual responses. 

You'll also hear a presentation today from Dr. James Stevenson, who with his colleagues conducted the two largest studies ever done on dyes and behavior. And unlike the previous studies, these two studies looked at ordinary British kids, kids not suspected of being sensitive to food dyes or other food substances. The two placebo-controlled studies administered mixtures of four dyes, plus, for some reason that I don't know, the preservative sodium benzoate, and they were given to young children. Between the two studies, six dyes were used, including the three dyes that account for 90 percent of all dyes used in the United States, Red 40, Yellow 5, and Yellow 6. Even though the children were not being suspected of being sensitive to dyes or other food ingredients, a battery of tests detected a small but significant effect of one mixture of dyes. There are two mixtures of dyes. So they detected an effect of one mixture of dyes on the behavior of three-yearolds, while both mixtures of dyes affected the eightand nine-year-old kids. 

Now, as I'm sure you've deduced, the FDA staff report downplays the findings of most of those studies, partly on the grounds that blinding wasn't proven, or that only parents and not clinicians or teachers associated the dyes with problems, or that the findings in a given study weren't internally consistent. 

I recognize that some studies didn't find any effects of dyes. And whether that was due to the luck of the draw in terms of including sensitive kids, inadequate dosages, inadequate testing methods, or other factors is unknowable. Still, some of the studies demonstrate clearly that normally-consumed amounts of dyes can impair the behavior of some children with hyperactivity or other behavioral issues, while other studies are equivocal, as the FDA report acknowledges. 

The FDA report states that, "The effects of dyes appear to be due to a unique intolerance to these substances and not to any inherent neurotoxin property." However, that unique intolerance is not terribly unusual, with possibly millions of children destined to be adversely affected over the years. And remember, we have 300 million people who have moved through the childhood years, and we're going to be getting 300 million more kids, moving through those years where this may be an effect. I should mention it's totally unknown whether dyes have effects on adults. There are really no studies on that. 

The FDA staff report says that this latter scenario of unique hypersensitivity can best be addressed by continuing to understand the biomolecular factors that may predispose an organism, meaning a child in this case, to this type of unique disruptive behavioral response, to otherwise nonneurotoxic chemical substances. I maintain that from a health perspective, the supposedly unique hypersensitivity would be best addressed by getting dyes out of the food supply. 

The staff report portrays dyes as being similar to allergens. The problem is that some people are sensitive, rather than dyes having some impact. In the case of traditional food allergens,  because only a small minority of people are sensitive, the FDA holds that declaring those ingredients on labels offers sufficient protection. Susceptible people then could figure out if they have an allergy and avoid the allergens. And often, that works. Learning that one is allergic to foods that cause obvious physical symptoms like hives, diarrhea, or vomiting is a lot easier than identifying substances such as dyes that disturb behavior.

 After all, how many parents even think that dyes might affect a child's behavior? And that's why Ben Feingold's announcement 35 years ago raised such public interest, that nobody had ever thought that common chemical additives in food could disrupt the behavior of children. Moreover, no one has suggested that traditional foods be outlawed, like milk or peanuts or wheat. Those have been in the food supply, they're basic foodstuffs, and they're not going to be outlawed. Nobody's pushed for that. 

But synthetic food dyes have no nutritional value, no other health benefits. They're not preservatives. They're used primarily to trick people into thinking that a food contains fruit or other valuable ingredients. Dyes would not be missed in the food supply, except by the dye manufacturers. Europe is managing quite well with minimal use of dyes. While FDA's staff clearly worked hard to identify shortcomings in the studies on dyes, as well they should, they did not highlight why some of the studies may have underestimated the effects of dyes. 

First, most of the studies used doses much smaller than what many children consume. A dose of 15 to 30 milligrams is less than what children may get in an individual meal or snack. In fact, in 1976, an FDA nutritionist estimated that 10 percent of children between 1 and 5 years old consumed more than 121 milligrams of dye per day. Moreover, according to FDA's certification data, dye usage is now about twice what it was in the 1970s. It's possible that more children would have reacted or reacted more forcefully if higher doses had been used in these various studies. 

Most toxicological studies use exaggerated doses to increase the likelihood of seeing an effect in humans or animals because they know that they're not dealing with the most sensitive strain of rat or the most sensitive individuals in a small clinical study. They use, what, 20, 30, 40 people in some of the larger ones. Even the highest doses, 150 and 250 milligrams per day, were no higher than what some children consume on some days. While the staff report acknowledges a study by Rowe & Rowe, they acknowledged that study was a good dose response study, one of the only dose response studies, which found a linear dose response relationship. 

The FDA staff report argues that dyes' impact on behavior does not increase linearly with dose, but plateaus somewhere between, perhaps, 10 and 100 milligrams, somewhere where most of the studies -- doses that most of the studies used. Also, several factors may account for why it can be difficult to detect the effect of a dye challenge. 

For one thing, kids' behavior is normally erratic. These are kids. So with that kind of a noise level, it's much harder to detect an effect. Also, children may control their behavior in a clinical testing environment or a classroom better than they do at home. Aren't we all on better behavior here than we are at home and at a doctor's office? Moreover, while the FDA staff report concludes that foods or ingredients other than dyes, such as wheat or milk, affect some children's behavior, most studies make no effort whatsoever to control the consumption of these other substances, foods that can affect children's behavior. So is it a combination of dyes plus a piece of bread or a glass of milk? 

Most studies didn't consider that at all. I started my talk by discussing the legal framework for judging the safety of dyes. I think that a fair reading of the evidence is that there is not convincing evidence of no harm. On the contrary, I believe that there is convincing evidence of harm to at least some children, and then one could debate how many kids -- what is the magnitude of the harm to some of the kids, either judging by these controlled studies or in some cases the anecdotal evidence, which I think deserves some attention. 

If you conclude that dyes significantly impair some children's behavior, or that dyes have not been adequately demonstrated to be safe, the logical next step would be to advise the FDA to bar them from the food supply. However, I suspect that banning dyes would be a challenging and probably futile legal process because, except for Yellow 5, the dyes have been tested only in mixtures, not individually. So there's no way of knowing from all these studies if there is an effect, is it from Yellow 5, or is it Red 3, is it Red 40, or a combination of Yellow 5 and Green 3? No evidence on that. 

Hence, as a weaker measure, the FDA could simultaneously require a health notice on labels, as the European Union has done, and urge companies to stop using synthetic dyes, as the British government has done. A warning notice might state something like, "Warning or Notice, the artificial colorings in 76 this food cause hyperactivity and behavior problems in some children." A warning label would not be nearly as protective as a ban because each parent would still bear the burden of recognizing that their child is affected by dyes, and then try to prevent the child from consuming dyed foods. 

And, moreover, restaurant menus likely would not bear warnings. If you buy food from a vending machine, you don't read the label before you put in your money. However, warning labels certainly would help educate consumers and spur companies to reformulate their products without dyes. Thank you.

https://www.talkingaboutthescience.com/studies/FDA-FACTranscript-03-30-2011.pdf