Rock-bottom Quality at The Lancet

Plenty of red faces at The Lancet in a few years’ time when somebody will decide to carefully read what they have allowed onto such an esteemed publication:

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– – published online September 6, 2007 DOI:10.1016/S0140-6736(07)61306-3

To the eye of the busiest readers, such a paper could not be more explicit about the dangers of artificial food colouring and preservatives :

Interpretation: Artificial colours 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

It’s just too bad that the results published in the very same article warrant such a conclusion not at all.

(1) The study included a tiny sample of 300 children, hardly something meaningful for the “general population”

(2) Unbelievably, importance is given to result of very little statistical significance.

Statistical significance is indicated, as usual, as “p”: in the article, “p=0.044” means the probability of the result being by chance is  1/22. “p=0.02” corresponds to  1/50 and “p=0.023” means 1/44.

In other words, out of 22 results with “p=0.044“, one of them will be statistically bound to be due to chance: and thus, meaningless. In fact, it is best practice for statistical significance to be granted only for “p=0.01” (1/100) or less.

(3) The only result with an acceptable p is “mix B” with “p=0.001” (1/1000). However, that corresponds to an increase in hyperactivity of just 0.17, that is around 8.5% of the threshold (2.0) defined by the authors for Hyperactivity Disorder/ADHD

Such a low value, and the fact that “mix A” has shown no statistically-significant results, can only be interpreted by saying that the impact of artificial colouring and preservatives on ADHD is irrelevant

Note also that if I am not mistaken not even one of the children in the study ever showed any indication of Hyperactivity Disorder. And I will not even be drawn in the discussion of if and how ADHD could truly be measured as claimed in the article.


Oh boy! Could any of that have stopped the UK’s Food Standards Agency (sponsors of the study) from abusing the results to call for a lower use of artificial colouring and preservatives in food? ‘Course not.

Let’s give the FSA their due, though: having classified honey as junk food, incredibly claiming “science” to be on the side of such an abysmally stupid choice, they have to defend their reputation and therefore can only keep misusing “science” to provide foundations to their prejudices, for the foreseeable future.


Should we try to avoid using artificial colouring and preservatives in food, especially for children? Yes. But should we base our choice on inconclusive evidence masquerading it as “scientific”? No. Never.

Because: is it ethical to add meaningless worries to parents already 100% busy with their children and ADHD? No. And it will never be.


And by the way: shame to the science editors that don’t properly read the original articles they decide to write about. Critical eyes should not be confined to movie reviews.