Childhood intelligence and being a vegetarianBMJ 2007; 334 doi: https://doi.org/10.1136/bmj.39107.671412.80 (Published 01 February 2007) Cite this as: BMJ 2007;334:216
- Marcus Richards, MRC programme leader and UCL reader in cognitive epidemiology ()
Evidence increasingly suggests that intelligence is associated with health and survival,1 2 3 although the reasons for this are not fully understood. To varying degrees, intelligence could mediate the long term impact of early adverse circumstances (such as overcrowding), influence the acquisition of factors that protect health, and reflect underlying biological mechanisms that regulate health. A cohort study in this week's BMJ by Gale and colleagues4 assesses whether intelligence can influence the acquisition of protective factors. In a large representative population study of more than 8000 British men and women, intelligence in childhood was associated with a vegetarian diet in mid-adulthood, and this was independent of educational attainment and social class.4
So what do the results say about the relation between intelligence and personal values, and whether intelligence influences lifestyle choices that protect health? An analysis of five prospective studies found that vegetarians had a mortality rate 76% lower than that of non-vegetarians, after adjusting for age, sex, and smoking.5 A randomised controlled trial found that higher intakes of vegetables, legumes, fruit, and bread (as well as more fish and chicken instead of red meat) were associated with reduced total mortality, death from heart disease, and incident cancer in men who had survived myocardial infarction.6
Like all good research, the study by Gale and colleagues raises provocative but testable questions. Firstly, given that childhood cognition is itself modified by nutrients, including iodine, iron, zinc, vitamin B-12, folate, and omega 3 fatty acids,7 do dietary patterns established in childhood influence food choice in adults? If so, might this long term influence in part explain the association between intelligence and vegetarianism? Little is known about how diet in childhood relates to that in adulthood, although preliminary evidence from the 1946 British birth cohort suggests that people from families who ate large amounts of fruit in childhood continued to do so in midlife, whereas those from families who ate little fruit in childhood also had low fruit consumption in midlife.8
Secondly, observational studies have reported positive associations between the intake of nutrients such as B complex vitamins, omega 3 fatty acids, and unsaturated unhydrogenated fats in adults and cognitive function in later life,9 10 11 although a randomised controlled trial found no effect of long term vitamin E supplementation in healthy women aged 65 years or more.12 Might the associations with cognitive ageing reported in observational studies be explained to some extent by people choosing their dietary habits according to prior cognitive ability?
So to return to where we began, does diet mediate associations between childhood intelligence and adult health, as Gale and colleagues ask? Significant mediation of this kind was found in relation to obesity in the British 1958 birth cohort, whereby a high healthy diet score—based on consumption of fried food and fresh fruit at age 33—along with educational attainment significantly accounted for the association between childhood intelligence and weight gain between 16 and 42 years.13 If diet does mediate (or at least partly mediate) the effect of intelligence on other health outcomes—such as cardiovascular disease, colorectal cancer, diabetes, and Alzheimer's disease—then public health interventions to encourage eating a consistently healthy diet may be beneficial. Such an approach should begin with parents encouraging children to eat healthily, with the aim that they continue to make healthy food choices as adults, while simultaneously teaching adults themselves about healthy food choices and how these choices may have developed during childhood.
While Gale and colleagues found that intelligence is associated with food choice independently of educational attainment, it is also important to know if education is associated with food choice independently of intelligence and, if so, in what way. If schools make an important contribution to the development of healthy food choices, then it may be best if parents and teachers provide consistent advice on these choices.
Competing interests: None declared.
Provenance and peer review: Commissioned, not externally peer reviewed.