Bite Sized Science: Food for thought?
By Keir Liddle
The British Dietetic Association (BDA) are the professional association that represents and registers dietitians:
Registered Dietitians (RDs) are the only qualified health professionals that assess, diagnose and treat diet and nutrition problems at an individual and wider public health level. Uniquely, dietitians use the most up to date public health and scientific research on food, health and disease, which they translate into practical guidance to enable people to make appropriate lifestyle and food choices.
In other words, registered dietitians represent those people who are properly qualified to give advice on dietary concerns, and not the Gillian McKeiths of this world.
They have recently published results from a longitudinal study that the media have picked up on, reporting its findings as evidence that healthy eating improves IQ, and that a link between a diet high in processed foods and a slightly lower IQ exists. The study mentioned, The Avon Longitudinal Study of Parents and Children, has a website here, and the headlines relate to a statistical association (a correlation if you will) between the consumption of junk food at young age and subsequent IQ test results.
IQ is likely to be affected by many factors, and it seems likely that socioeconomic factors are a huge confound in the study, and that this will have a mediating effect on both junk food consumption and IQ scoring. Living in an environment with a low socioeconomic level is a predictor of high BMI, and the link between low socioeconomic level and the consumption of cheap junk foods is also well established. The relationship between IQ and low socioeconomic status is more controversial, and there may be many confounds associated with this. However, generally speaking those with low socioeconomic status score lower on IQ tests these results can be explained using environmental factors: significant effects from environmental factors on with 80% of the variance in IQ being explained in poor children but with the opposite being found with children from wealthy backgrounds.
However the paper does state that the following factors corrected for:
“breastfeeding duration, energy intake, maternal education, maternal social class, maternal age, housing tenure, life events, HOME score and all other dietary pattern scores”
However corrected versus uncorrected values aren’t shown or even which factor is most meaningful – also the paper doesn’t look at father’s education and social class which may not always match the mother’s exactly. It also does not mention whether a family is single or two parent – so it’s difficult to impossible to know how well thase have actually been normalised. A lack of detail on confounding factors casts some dount on the robustness of their epidemiology. However this is a signpost study and the researchers have not overstated their case so as the saying goes: more research is needed.