Randomised controlled trialindependent cross-checkabstract checked I-UPF-120 in-depth review
Food colours, sodium benzoate and children's behaviour
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.
Overview
McCann D et al.. Lancet 2007 · PMID 17825405 · DOI 10.1016/S0140-6736(07)61306-3
In a blinded trial, drinks with artificial colours plus the preservative sodium benzoate slightly increased hyperactive behaviour in children.
Does the study support the specific claim? No · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
To test additives on children's behaviour.
- How was the study done?
Double-blind placebo-controlled crossover trial.
- How many people, animals or samples were studied?
297 children.
- What was actually found?
Small increase in hyperactivity.
- How large was the observed effect?
Effect sizes 0.12-0.32.
- What do the results mean in everyday life?
Some additive mixtures may affect sensitive children.
- What are the study's limitations?
Benzoate and colours not separated.
- What does it say about Elawady's specific claim?
Limits C-UPF-11.
- Do the results apply to Germany and Europe?
Applicable.
- What can explicitly not be concluded from this study?
Cannot show benzoate alone is harmful.
Scientific in-depth review
1. Full reference
- Authors
- McCann D et al.
- Journal
- Lancet
- Year
- 2007
- Identifier
- PMID 17825405 DOI 10.1016/S0140-6736(07)61306-3
- Verification
- verified – PubMed efetch 2026-10-26; Crossref works API 2026-10-26
- Retracted
- no
- Corrections
ErratumIn: Lancet. 2007 Nov 3;370(9598):1542 (erratum notice; no retraction)
- Extent checked
- abstract only
- Conflicts of interest and funding
No COI statement in PubMed record
2. Research question
Do artificial food colours and sodium benzoate affect hyperactive behaviour in children from the general population?
3. Study design and population
- Study design
- Randomised controlled trial – Randomised, double-blinded, placebo-controlled crossover trial
- Population
3-year-old and 8/9-year-old children
- Country
UK (community sample; location not detailed in abstract)
- Period
Not stated in abstract
- Sample size
153 3-year-olds and 144 8/9-year-olds
- Inclusion and exclusion criteria
Community sample
4. Methods and endpoints
- Exposure / intervention
Drink containing sodium benzoate plus colour mix A or B
- Comparator
Placebo drink
- Endpoints
Global hyperactivity aggregate (GHA)
- Statistical methods
Per-protocol analysis; effect sizes
5. Actual results
- Results
3-year-olds: mix A vs placebo effect size 0.20 (0.01-0.39), P=0.044; not mix B; restricted analysis 0.32 (0.05-0.60). 8/9-year-olds (>=85% consumption, no missing data): mix A 0.12 (0.02-0.23), P=0.023; mix B 0.17 (0.07-0.28), P=0.001.
- Absolute risks
Not applicable (continuous outcome)
- Authors' conclusion
Artificial colours or sodium benzoate (or both) result in increased hyperactivity in children in the general population.
6. Strengths
- Double-blind placebo-controlled crossover
- General-population children
7. Weaknesses
- Cannot separate benzoate from colours
- Small effect sizes; per-protocol analysis
- Erratum published (Lancet 2007;370:1542)
8. Risk of bias
- Orienting tool
- RoB 2
- Type of assessment
- simplified assessment (abstract only)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
Per-protocol; mixtures.
9. Assessment of the statistical interpretation
Small standardised effects (0.12-0.32).
10. Relevance to the specific claim
Randomised evidence that a mixture containing sodium benzoate can affect children's behaviour; contradicts 'trivial' for children, though benzoate's own role is unclear.
11. Independent contrary findings
Effect cannot be attributed to benzoate alone; EFSA considered effects small.
12. Transferability to Germany and Europe
Applicable to EU-authorised additives; regulatory follow-up not reviewed here.
13. Overall methodological judgement
Limits a blanket statement for children; does not show harm of benzoate alone.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | low |
| Directness | direct |
| Reproducibility | partly replicated |
| Consistency with the overall evidence | mixed |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Benzoate-only effect.