BBC Breakfast – specialists in moaning and scaremongering pre-8am, and plugging new books and albums post-8am (but that’s ok, it’s still ‘public service broadcasting’, right?) – spewed this delicious news story in my face this morning. I was also delighted to hear it was the leading news story on the front page of Daily Mail.
The premise of the story is this – if your child eats junk food from an early age, they’ll be less intelligent later on in life. Following what must be standard “science news article in mainstream media” protocol, neither article links the original paper from which this story originates. I’ll save you the hassle – it’s >here< (doi:10.1136/jech.2010.111955)
A quick summary about the authors of this paper – this is from the Avon Longitudinal Study of Parents and Children (or ALSPAC) sometimes known as the ‘Children of the 90s’ – which in my mind means they’re wearing friendship bracelets whilst watching “Live & Kicking” and playing with some Micro Machines. But that’s just me.
Anyway, good on them, because I imagine these kinds of longitudinal studies can reveal a lot of interesting information. In this particular study, Dr Kate Northstone & colleagues have looked at the eating habits of children at the ages of 3, 4, 5, and 8 years old, in terms of how often they’re consuming a long list of different food and drinks. Being children of the 90s, maybe Sunny D is on the list.
These food types were then categorised into “Processed” (high fat & sugar content), “Traditional” (meat, poultry, potato, and vegetable), “Health-conscious” (salads, fish, fruit, and rice), and “Snacks” for 3 years olds (biscuits and such, though whether this included such treats as ‘Club’ or ‘Um Bongo’ is still uncertain).
Then, at 8.5 years old, the kids had an IQ test, and then the scores of the individual children are correlated with their diets for each year since they were 3 years old.
If you just compare the diets and IQs directly and don’t really adjust for any other factors, then all kinds of significant results come out – which is unsurprising, as Little Johnny who eats all his greens might have a very different life to Little Jamelia who eats all her greens, and this difference might also influence their test scores.
When the score results are adjusted for potentially important factors (such as breastfeeding, energy intake, maternal education, maternal social class, maternal age, housing tenure, something called “life events”, and their diet in the following years) , most of these significant correlations disappear, leaving only one very statistically significant finding – that children who eat more than average amount of ‘processed’ foods at 3 years of age tend to, on average, have lower IQs at the age of 8.5.
Other statistically significant findings are, to quickly summarise, that eating more snacks as a 3 year old raises your IQ at 8.5 years, and that eating more ‘health-conscious’ foods at 8 years old bumped up your IQ.
I’m stressing that these are ‘statistically’ significant findings, and I trust the maths is correct. What I’m not convinced by is that these are significant findings, in terms of real life. The average decrease in IQ that your 8.5 year old child can expect if you fed them more processed food than average is 1.67 points. Similarly, the increases for eating more snacks as a 3 year old and more healthy stuff as an 8 year are 0.9 and 1.2, respectively.
I’m no expert on intelligence tests, by any measure, but I know this much – the way IQ tests, such as the one used in this study, are scored is regularly adjusted so that the average IQ score for the population is 100, and the standard deviation is 15. That means that 62.87% of the population will have a score betwen 85 and 115. A difference of 1.67 points (to me at least, but correct me if I’m wrong), doesn’t really sound like a huge difference. That might well be the difference a child from the 90s experiences before and after common events, such as “being tango’d”, or watching “The Big Breakfast”. It might be the difference between me taking the test in the morning and taking it again in the evening. I’m not entirely sure, to be fair – correct me if I’m wrong.
But there’s another thing – this is IQ tests of an 8 year old, not of them as fully grown adults. Who’s to say naughty little Sammy who only ate pizza won’t grow up to be Prime Minister? And maybe that he’ll have a high IQ? Everyone knows that children are stupid, even the ones that’ll grow up to be PhD students. I used to believe that if you opened the plug whilst in the bath, you got sucked down the plug hole and sliced up Play-Doh style. I used to have curtains as a hair style. I owned a copy of “Mr Blobby” – and I liked it.
Now, in future years, I expect the ALSPAC study might look at their IQ scores as adults, and if they’ve got differing IQs and that’s dependent on how much processed or healthy food they did or didn’t eat (adjusting for everything else), then we’ve got something really interesting – like a massive Pog collection, or everything on ‘The X-Files’ – and not just statistically interesting.
But of course, I can’t really fault the paper or the authors; they acquired some data, they analysed it, they got some results, and they published them, and I honestly don’t know enough about these kinds of studies to really critically analyse their methods. It’s more the presentation of the results by BBC and Daily Mail (amongst others, it must be said). To their credit, the BBC article at least mention that the 8 year old’s IQ scores associated with eating more processed foods at age 3 are only “slightly lower” – though this was not the case on BBC Breakfast television this morning, I can tell you THAT.
The Daily Mail article is a little less subtle. “Toddlers fed a diet of junk food can suffer lasting damage to their brainpower, researchers warn” (No claims of how long this IQ dip lasts for are made in the paper). “The difference could be as much as five IQ points compared with children given healthier diets” (5 points? Not sure where this figure’s come from. I suppose if the minimally adjusted figures are used… and worse of them in the 95% Confidence Interval bracket is used… and they’re rounded up…). “But even if their diet improves, it could be too late as the ill-effects can persist for a lifetime” (assuming ‘ill-effects’ refer to the IQ scores, again, no claims as to how long this IQ dips lasts). It’s just a little bit sloppy. But enough Daily Mail bashing; it’s all a little bit too easy, even easier than killing a Tamagotchi.
Ah well. All this 90s nostalgia is making me hungry for some Nerds – doesn’t matter now, does it? Enjoy!