Junkfood Science: Purple tongues and big red grins

September 25, 2007

Purple tongues and big red grins

Over a century ago, a little eleven year old boy named Frank Epperson took his favorite brightly colored soda pop, put a stick into it, and left it on his porch to freeze. Eighteen years later, he patented his ice pop, which by that time his own kids had renamed the “popsicle.” In 1902, the first Devil’s Food Chocolate Cake (also called Red Velvet Cake because of its deep red color) recipe appeared in an American cookbook called Mrs. Rorer's New Cook Book by Sarah Tyson Rorer. During the Great Depression, around the time Frank was getting his patent, Kool-Aid was born, sporting equally bright colors. And who knew that M&Ms came into being inspired by, as the story goes, soldiers in the Spanish Civil War who were eating little beads of chocolate encased in vividly-colored sugar coating to keep it from melting?

While these aren’t the earliest examples of food colorings in America’s favorite foods — foods have been dyed for centuries — it’s nice to be reminded sometimes that what’s feared today as perversions of a modern diet, aren’t all that modern at all. It’s equally helpful to remember that food additives have never been safer than those used today and far fewer are used today ever in our history. Only 7 food colors are allowed in foods and beverages in the U.S. today. Since 1938, the Food and Drug Administration (Federal Food, Drug and Cosmetic Act) has regulated food additives to prevent adulterated foods that could be harmful (prior to that, the Food and Drug Act of 1906 established a voluntary certification program of food additives administered by the U.S. Department of Agriculture). The 1958 Food Additives Amendments and the 1960 Color Additive Amendments began fresh and required food companies to provide proof that their additives were safe before permitted to use in foods.

In contrast, those making scary claims that ‘food additives are unsafe’ don’t have to provide any proof at all. You’ve heard the one about hyperactivity in kids caused by food additives...

This scare is more than 30 years old and first surfaced in 1973 when a California allergist, Dr. Ben Feingold, claimed hyperactivity was caused by artificial colors and flavors. He convinced followers that 32% to 60% of hyperactive children could improve on his special elimination diet that removed all “processed foods” and artificial colors and flavors. Saying he had no time for science or to conduct a double-blind trial, others have tried to prove his claims and, to date, no one has been able to. Instead, the most well-designed and carefully-conducted clinical studies have disproven them. None has been able to show evidence that food additives and food colors affect the behavior of children. At most, very rare numbers, about 1 or 2 in 10,000 kids, may be sensitive to certain colors resulting in itching or congestion, something that can be said about most any food ingredient. Nor is there any evidence of epidemic hyperactivity among kids in the early part of the last century when far more artificial food colorants were used.

According to the Food Color Facts brochure by the U.S. Food and Drug Administration, “well-controlled studies conducted since [the 1970s] have produced no evidence that food color additives cause hyperactivity or learning disabilities in children.” As reviews have pointed out, invariably when symptoms of hyperactivity are reported, they are subjective impressions by parents, rather than independently or objectively verified, and any changes in the diet have appeared related to changes in the family dynamics, such as paying more attention to the children.

In 1980, the National Advisory Committee on Hyperkinesis and Food Additives, a group of behavioral and medical scientists organized by the Nutrition Foundation, conducted a critical review of clinical trials that have tested Feingold’s claims and concluded:

Based on seven studies involving approximately 190 children, there have been no instances of consistent, dramatic deterioration in behavior in hyperactive children challenged, under double-blind conditions, with artificial food colorings...There are three exceptions to these generally negative conclusions, but, in all three cases, the deterioration is reported by the mother with no other objective, confirming evidence available.... Without the confirming evidence of objective tests and/or outside observers, even these exceptions cannot be considered as definite evidence that there may be an occasional, genetically determined, sensitivity to food colorings. Though one cannot prove that no such children will be found, sufficient numbers of highly selected children have been studied to feel confident that such specific sensitivity, if found, will be rare.

The belief continued, as all good scares do. So, studies attempting to support the claim have continued, too. One study conducted in 2000 by the David Hyde Asthma and Allergy Research Centre, Isle of Wight, received a great deal of media attention and claimed to show that a cocktail of additives given to 3 year olds affected their behavior. The UK Committee on Toxicity (COT), however, reviewed the study and determined that the methodology was deficient and unable to support the conclusion. The children exhibited no difference in behavioral changes when consuming the additive cocktail versus the placebo when assessed in the clinical setting, and objective observational assessments could not detect the behavioral changes reported by the parents.

The reason why scientists and doctors continue see no sound evidence to point to food colors as causing behavioral or learning difficulties in children is that they are looking neutrally at the most objective evidence and at the design of studies. When we, parents or loved ones are emotionally close to a situation, our expectations can influence how symptoms are interpreted. Even researchers’ expectations can influence how symptoms are interpreted. It’s hard to admit, but we really can’t trust our own experiences and medical conditions is one area where they’re not reliable at all. And interpreting children’s behaviors is especially hard.

Many of the symptoms of ADHD, or attention-deficit hyperactivity disorders, are subjective and the same symptom to one person could be seen as normal, high-energy, boisterous kid behavior, while to another be viewed as abnormal. The American Academy of Family Physicians offers a helpful overview of ADHD for parents. As they note, the symptoms of ADHD hyperactivity include: fidgety, runs or climbs inappropriately, doesn’t play quietly, blurts out answers or interrupts people, can’t sit still in seat, talks too much, is always on the go and has trouble waiting his/her turn.

So many things can affect a child’s behavior and also create the perception of a link — tiredness, hunger, stress, fear, excitement, discipline, boredom, etc. — isolating the cause from confounding factors is difficult.

The only way to scientifically overcome observational bias, subjectivity and the placebo effect is through “double-blind” placebo controlled trials. These types of studies have shown scientists time and again that we can’t trust our own eyes. In these studies, a random group of participants is given the treatment and another group are given a pretend treatment (placebo) and neither the researchers nor the participants know which is which until the study ends (they are thus both “blind”). To make things even tricker, double-blind studies are not all created alike. The design, conduct and reporting of these studies can give some more credence than others. Double-blind studies are especially important when testing kids because not only can the parents be influenced by their own expectations, but kids are also very astute to parental expectations and can react to or mirror their parents’ and other adults’ fears, anxieties or positive attitudes.


The latest study in the news

If you’re a parent, no matter how many times and how much careful evidence has shown food additives to be safe, recent news reporting on a study published in the Lancet said to have found that food colorings increase hyperactivity in children may still have been frightening. It claimed to be the first ever to show a direct connection.

No one has bothered to look closely at this study, but here’s why it needn’t worry you.

First the news:

Food Additives Could Fuel Hyperactivity in Kids

Some common food colorings and preservatives appear to increase the risk of hyperactive behavior among children, British researchers report. The link between food additives and hyperactivity has long been suspected, but this is the first study to show a direct connection. The findings have already caused the British government's Food Standards Agency, which funded the study, to issue a warning to parents about food additives.

Attention-deficit hyperactivity disorder (ADHD) is an increasingly common problem, and theories abound to account for that," said Dr. David Katz, director of the Prevention Research Center at Yale University School of Medicine. “Among them is the notion that food additives induce hyperactivity....No one factor is solely responsible for rising rates of ADHD," Katz said. “Along with the hazards of a highly processed food supply, children are getting less and less physical activity as a means of dissipating their native rambunctiousness."

In the study, Jim Stevenson, a professor of psychology at the University of Southampton, and his colleagues gave drinks containing additives to 297 children. The children were in two groups: 3-year-olds and 8- and 9-year-olds. The drinks contained artificial food coloring and additives such as sodium benzoate, a preservative.... Over the six weeks of the trial, Stevenson's team found that children in both age groups who drank the drinks containing additives displayed significantly more hyperactive behavior. These children also had shorter attention spans. However, which specific additives caused specific behavioral problems is not known, the researchers said....

“However, we need to remember that there are many factors associated with hyperactive behavior in children. These are thought to include genetic factors, being born prematurely, or environment and upbringing," Dr, Andrew Wadge, chief scientist at the Food Standards Agency, said in a statement. Insights about the causes of ADHD should help parents implement preventive strategies which are urgently needed, Katz noted. “A healthful, unadulterated diet and regular physical activity seem like a good place to start," he added.

This study and the news coverage have many classic earmarks of a marketing agenda, versus sound science.


Remember Peter Wason’s admonitions?

All you have to do is find an instance — in this case, a well-done study — that doesn’t support a claim to decisively eliminate it as being true, whereas a sound claim can be replicated exhaustively. Those trying to scare us about dangers in our food, bodies or world will repeat and try to prove the same disproven scare over and over again, rather than move on. This study wasn’t using some new, more modern method of assessing hyperactivity that thirty years of studies didn’t have available. Nor was it testing some new food colorant on the market that hasn’t been tested before. And a study that was truly trying to identify a possible chemical or combination that might contribute to hyperactivity wouldn’t have created two large cocktails of chemicals, with no rhyme or reason to the combinations, making isolating any potential culprit(s) impossible. A well-designed study would have systematically tested individual chemicals or a progression of combinations, for instance.

There is another telltale sign that this is primarily food politics and being used to scare young parents about “unhealthy” foods and artificial ingredients — besides the fact that mere weeks after the first media blitz about the study, special interest groups are plastering the news again this week with complaints that the UK government hasn’t acted fast enough to ban these ingredients.

Any insinuations by American doctors or health food advocates or the American media that this study supports the dangers of food additives in the “increasing common problem” of ADHD, and the need for an “unadulterated” healthy diet, is bogus. Most of the colorants used in this study — quinoline, carmoisine and ponceau — aren’t even approved for use in foods and drinks in the United States. And another one, sunset yellow is found naturally in squashes and other yellow-colored produce. Nor is there any evidence that British children have more hyperactivity than American kids.

Whenever you find claims of “evidence” to support a fear that’s been repeated shown not to be true, it’s your alert to look closely at how they did it, and how and why you’re being manipulated. Let’s walk through a few details in this study.


Was this study truly double-blind?

Just because researchers claim that the participants and observers weren’t told which beverages were the additive cocktails and which were the placebo beverages... doesn’t mean the parents or kids couldn’t tell. As the researchers explained, the cocktail mixes A and B had different combinations of 6 different food colors and they varied the amounts of fruit juices in them to match the color and taste to the placebo drink. They said they’d conducted masked testing of the drinks on two panels of young adults (not testing individuals independently), and that any differences between the drinks couldn’t be detected. They reported that 40% thought mix A had chemicals and 65% thought the placebo had chemicals. But they didn’t do blind taste tests with any of the parents and, more importantly, they didn’t ask the kids.

And the kids, as we’ll see, were having to guzzle a lot more of these drinks than mere sips in a taste test.

The parents’ ability to detect which drink had the additives, for instance, would essentially eliminate the blindness. When parents believe that their children are getting a chemical that could contribute to hyperactivity, they will be more likely to look for or see signs in their kids. And children are more likely to react to parental expectations and anxieties.

It is also worth noting that the parents who agreed to allow their children to participate in this study were a small fraction of the original cohorts: a mere 17% of the parents of the 3 year olds and 25% of the parents of the 8/9 year olds. And after the study got started, another 20 removed their child from the study and didn’t complete it, bringing down the completion rates to 15% and 20% of the original group, respectively (for a total of 267 kids). These parents appeared to differ from the general population and, perhaps, they were more suggestible or less careful observers. It is noteworthy that only 35% of the participating mothers had attended school to age 16, with only 18% completing college, and only 20% of the parents had higher occupations.

All of this is significant because the children’s hyperactivity was, in part, determined by their parents. Each week, the parents rated their child’s behaviors — “switching activities, interrupting or talking too much, wriggling, fiddling with objects or own body, restless, always on the go, concentration” — which were totaled into a score. Two types of outside observations were also gathered. Children were observed in the classroom for 24 minutes every week (three observation sessions of 8 minutes each) by psychology graduates and a total weekly mean score was derived; and the kids were rated on abbreviated ADHD scales by teachers on questionnaires each week. But since 3 year olds aren’t in school, preschool attendants merely scored the younger children’s behaviors when they changed activities. Finally, just one objective measure was taken, but only on the 8/9-year-old children each week: the Conners continuous performance test II, a test that uses visual stimuli of 14-minute duration to measure attention and control.

But in reporting their results, the researchers combined the objective measures with outside observations and the more subjective impressions of parents, making credible interpretations impossible.


Going to the source

Even so, the results reported in the actual study are considerably different than the spin being reported in the media. For most of the observations, in fact, the confidence intervals are so close to zero that the slightest reduction would have shifted them all into statistical insignificance. Here’s what the researchers actually reported:

In model 2, in which the effects of other factors were controlled, the effect of mix A for the entire sample was not significant (p=0·123) but mix B did have a significantly adverse effect compared with placebo (p=0·012). When the analyses are restricted to those children who consumed at least 85% juice, the adverse effect of mix A on behaviour remained non-significant (p=0·066) but was significant for mix B (p=0·003).... For mix A, the effect for 3-year-old children was greater than for 8/9-year-old children. The effects for mix B were not significant for 3-year-old children because there was greater variability in the response to the active challenges than placebo in this age group. Thus, we recorded substantial individual differences in the response of children to the additives.

Among 8/9 year olds, they reported a 17% increase in ADD behavior for those who finished their juice with Mix B, but not otherwise. So, they found nothing tenable with mix A and only a tiny purported increase, using problematic data, among the older kids with mix B. Whether this represents true clinical hyperactivity, or older kids becoming more impatient and resistant to the study is unknown, but the later is not unreasonable to imagine.

What hasn’t been reported in the news is that these kids were being given 21 ounces a day of a drink (that they might not have even liked), which is considerably more than children their age drink daily of sweetened carbonated drinks (7 oz.) or fruit drinks (3 oz.), according to NHANES and the American Academy of Pediatrics. In fact, only 11% of kids their age drink more than 12 oz. a day of sweetened drinks. Before even beginning the additive cocktail part of the study, the children also had to drink wash-out placebo drinks for six weeks, before being randomized to placebo or additive cocktail mix A or B during weeks 2, 4 and 6 of the study. By then, the kids might easily have been able to tell a difference between the drinks. They also might not have liked them or been sick of them and began acting up more in rebellion at having to drink the cocktails. Or maybe, they just needed to pee. :-)

What were the adverse effects of filling their little tummies with such extreme and unrealistic volumes of sweet drinks every day — over twice what even the small numbers of the highest-consuming kids would ever normally drink — and displacing calcium-rich milk and a variety of foods for so long? Was the potential harm to the overall health to these children considered or part of the informed consents the parents signed? By not monitoring electrolytes and minerals, such as calcium, for example, it’s impossible to know how much of their fidgetiness and other symptoms, even if they’d been statistically significant, might have been due to nutritional issues.

The study concluded by saying:

We have found an adverse effect of food additives on the hyperactive behavior of 3 year old and 8/9 year old children.. The implications of these results for the regulation of food additive use could be substantial.

Not mentioned was that that 'adverse effect' wasn’t statistically significant, but the political angle was. This study has not provided any credible, objective evidence to negate over 30 years of evidence finding that food colorings used in foods and drinks are safe and do not cause hyperactivity in children.


For more information on the causes of ADHD and what parents can do to help their children, the American Academy of Family Physicians has a free brochure, ADHD: What Parents Should Know. Other things that do not cause ADHD, they note, include:

· Bad parenting

·Too much sugar

·Too little sugar

·Aspartame (artificial sweeteners)

·Food allergies or other allergies

·Lack of vitamins

·Fluorescent lights

·Too much TV

·Video games

© 2007 Sandy Szwarc

Bookmark and Share