Weakening of the cognition and height association from 1957 to 2018: findings from four British birth cohort studies

Abstract

BackgroundOn average taller individuals have been repeatedly found to have higher scores on cognitive assessments, yet it is unclear whether the magnitude of this association has systematically changed across time. Recent studies have found that this association can be explained by genetic factors, yet this does not preclude the influence of environmental or social factors that affect the genome. We tested whether the association between cognition and height has weakened across time.MethodsWe used four British birth cohorts (born 1946c, 1958c, 1970c, and 2001) with comparable data available at 10/11 and 14/17 years (N = 41,418). Height was measured at each age, and cognition via verbal reasoning (10/11 years) and vocabulary/comprehension scales (14/16 years) and via mathematical tests at both ages. We constructed age-adjusted height and cognition measures and converted cognition measures to ridit scores to aid interpretation. We then used linear and quantile regression to investigate whether cross-sectional associations between cognition and height differed in each cohort, sequentially adjusting for sex, childhood socioeconomic position, and maternal and paternal height.ResultsTaller participants had higher mean cognitive assessment scores in childhood and adolescence, yet the associations were weaker in later (1970c and 2001c) cohorts – after adjustment for sex the mean difference in height comparing the highest with lowest verbal cognition scores at 10/11 years was 0.57 SD (95% CI = 0.44, 0.7) in the 1946c, 0.59 SD (0.52, 0.65) in the 1958c, 0.47 SD (0.41, 0.53) in the 1970c, 0.30 SD (0.23, 0.37) in the 2001c. This pattern of change in association was observed across all specifications (ages 10/11 and 14/16 years, and for each cognition measure used), and was robust to adjustment for social class and parental height, and modelling of plausible missing-not-at-random scenarios. Quantile regression suggested that these average differences were driven by differences in the lower centiles of height. This pattern was most evident in older cohorts – for example, in 1958c, the difference in height was 0.73 SD (0.64, 0.82) at the 10th centile, yet 0.46 SD (0.34, 0.57) at the 90th centile.ConclusionAssociations between height and cognitive assessment scores in childhood-adolescence weakened by at least half from 1957 to 2018. These results support the notion that environmental and social change can markedly weaken associations between cognition and other traits.