Tuesday, 9 July 2013

To sleep, perchance to boost cognitive development


Time for bed: associations with cognitive performance in 7-year-old children: a longitudinal population-based study
Yvonne Kelly, John Kelly, Amanda Sacker

JECH Online First, published on July 8, 2013 as 10.1136/jech-2012-202024

The authors have looked at the Millenium Cohort study, and linked the times at which children go to bed to their scores on the British Ability Scales reading, maths and pattern construction. I have another post in progress on this Millenium cohort, so I will defer fuller comment till then.

Girls seem to be affected by late nights, particularly those sustained when they are 3 years of age, boys far less so. The effects, such as they are, are strongly attenuated by controlling for the educational level of the parents.  The authors explain:

“Model B additionally adjusts for interview season, school year, mother’s age in years, birth order, languages spoken in the home, family income, highest parental qualification, mother’s psychological distress, discipline strategies, mother often irritated with child, mother’s parenting competence, breastfed, skips breakfast, musical activities at home, drawing/painting activities at home, helped with reading, TV hours per weekday, computer hours per weekday.”

Bluntly, if you are bright and middle class, with proper Burgher, slow-life, delayed gratification strategies, your darling brat will be in bed at a time that suits you, and not the brat.

However, there is a sliver of residual variance which makes you even more justified in putting your daughter to bed early when very young, whatever her opinion on the matter.

A 10 or 20 word vocabulary test given to parents would probably have furnished the Model B correction more efficiently, but one cannot have everything. Soon we will be getting the new birth cohort studies with full genomic analysis and repeated brain scans. That should provide some fun.