Skip to content
SugarMyth
Busting the sugar myth...
The evidence is settled

Sugar does not make your kids hyperactive.

Most people think it does. Researchers have been testing that belief since the 1970s and keep coming up empty. Below: where the idea started, and the 1994 experiment that explains why you see it at every birthday party anyway.

Trials pooled
23
Measures tested
39
Effect found
None

Origin

The belief was never based on a study about sugar.

In 1973, allergist Benjamin Feingold popularized an elimination diet aimed at artificial dyes and flavorings, which he believed caused hyperactivity in children. Sugar was not part of his hypothesis.

It got swept in anyway. Because parents already suspected sugar, it was filed under the general heading of “food additives.” Once it was on that list, the association hardened into common knowledge. The belief spread first and got tested second.

The record

Then researchers went looking for it.

Over five decades, in better and better controlled designs, nobody has found it.

  1. 1973Myth forms

    The Feingold diet reaches the mainstream

    An additive-elimination diet becomes widely adopted. Sugar is folded in by association, not by evidence.

  2. 1982First verdict

    The National Institutes of Health finds no proven link

    An NIH consensus statement concludes that no link between sugar and hyperactivity has been scientifically demonstrated.

  3. 1994Controlled diets

    39 measures, no significant differences

    Children ate three separate three-week diets: high sucrose, aspartame, and a saccharin placebo. Across 39 behavioral and cognitive variables, none of the three diets produced a significant difference.

    New England Journal of Medicine [2]
  4. 1994The twist

    The effect turns up in the parents

    35 boys aged 5 to 7, all given an aspartame placebo. Mothers told their son had received sugar rated him significantly more hyperactive than mothers who were told the truth.

    Journal of Abnormal Child Psychology [3]
  5. 1995Meta-analysis

    23 trials pooled. No effect on behavior or cognition.

    A meta-analysis of 23 within-subject trials drawn from 16 reports finds that sugar does not affect children’s behavior or cognitive performance. The authors note a small effect, or an effect in some subset of children, cannot be ruled out.

    JAMA [1]
  6. 2017Modern check

    287 children, no relationship to behavior or sleep

    Total sugar consumption in Australian children aged 8 to 12 was not associated with sleep or behavioral domains, and did not help predict sleep or behavior problems.

    International Journal of Food Sciences and Nutrition [4]

The numbers

How small is the effect they measured?

The 1995 meta-analysis sorted its outcomes into 14 measurement constructs and computed a mean effect size for each. Effect size is expressed in standard deviations, so it can be compared against the conventional thresholds for a small, medium and large effect.

Mean effect sizes across all 14 constructs

Wolraich, Wilson & White, JAMA 1995 [1]

Observed range of mean effect sizes: −0.14 (direct observations) to +0.30 (academic tests) below “small” 0.2 SMALL 0.5 MEDIUM 0.8 LARGE −0.14 +0.30 direct observations academic tests −0.4 −0.2 0 0.2 0.5 0.8 EFFECT SIZE (STANDARD DEVIATIONS)
All 14 construct means fall between −0.14 and +0.30, and the 95% confidence interval for every one of the 14 includes zero. The largest value in either direction is smaller than the conventional threshold for a small effect. The bar shows the observed range; the 14 individual values are not published separately in the abstract, so they are not drawn.

The 1994 controlled-diet trial: 39 outcomes measured

Wolraich et al., New England Journal of Medicine 1994 [2]

measured, no significant difference: 39 significant difference: 0
Three diets, three weeks each, run double-blind. Zero of the 39 behavioral and cognitive variables differed significantly between the high-sucrose diet and either placebo.

The experiment that explains it

So why does everyone see it anyway?

In 1994, researchers recruited 35 boys aged 5 to 7 whose mothers described them as sugar-sensitive. All of them got an aspartame placebo. Not one received sugar. The only variable was what their mothers were told.

Randomized, double-blind n = 35 boys, ages 5 to 7
Mothers told: “sugar”
Child actually received
Aspartame placebo
Rating of hyperactivity
Significantly higher
Mother’s behavior
More controlling, more criticism
Mothers told: “placebo”
Child actually received
Aspartame placebo
Rating of hyperactivity
Baseline
Mother’s behavior
Baseline
Both groups of children got exactly the same thing. The only measured difference was in the mothers. [3]

The sugar rush is real. It is happening in the person watching.

The evidence base

Four studies, four designs, one result.

Study Design Participants Outcomes Effect found
NEJM 1994 [2] Three 3-week diets, double-blind 25 aged 3–5
23 aged 6–10		 39 variables None
J Abnorm Child Psychol 1994 [3] Expectancy trial, all children on placebo 35 boys aged 5–7 Mother ratings, observed interaction In the mothers only
JAMA 1995 [1] Meta-analysis of 23 trials from 16 reports 23 trials 14 constructs None
Int J Food Sci Nutr 2017 [4] Cross-sectional, Australian cohort 287 aged 8–12 Sleep and behavior domains None

What is going on instead

Sugar and excitement arrive at the same parties.

Birthdays. Halloween. Christmas morning. The last day of school.

Children are wound up at these events, and these events serve cake. The two show up together, so one gets blamed for the other. Take away the presents, the other kids and the broken bedtime, hand a child sugar in a quiet room, and the behavior does not follow. The occasion was doing the work.

None of this makes sugar good for children. There are real reasons to limit it: teeth, weight, and what it crowds out of a diet. Hyperactivity is not one of them.

References

The primary literature.

  1. Wolraich ML, Wilson DB, White JW. The Effect of Sugar on Behavior or Cognition in Children: A Meta-analysis. JAMA 1995;274(20):1617–1621.
    jamanetwork.com
  2. Wolraich ML, Lindgren SD, Stumbo PJ, Stegink LD, Appelbaum MI, Kiritsy MC. Effects of diets high in sucrose or aspartame on the behavior and cognitive performance of children. New England Journal of Medicine 1994;330(5):301–307. PMID 8277950.
    PMID 8277950
  3. Hoover DW, Milich R. Effects of sugar ingestion expectancies on mother-child interactions. Journal of Abnormal Child Psychology 1994. doi:10.1007/BF02168088. PMID 7963081.
    PMID 7963081
  4. Total dietary sugar consumption does not influence sleep or behaviour in Australian children. International Journal of Food Sciences and Nutrition 69(4):503–512. Published online 18 October 2017. doi:10.1080/09637486.2017.1386628.
    tandfonline.com

Further reading

Plain-language explainers.

  1. Sugar: Does it Really Cause Hyperactivity? Academy of Nutrition and Dietetics
  2. Does sugar really make kids hyper? Australian Academy of Science