Objective: In developed western populations longer legs have been shown to be a biomarker of better early childhood conditions. It was hypothesised that in transitioning populations better childhood conditions may bring forward puberty and thus decrease leg length, counteracting the overall positive effect of a favourable childhood environment on leg growth.
Design: Structural equation modelling was used to assess the interrelationship of age, education, father's job, age of menarche and leg length in a cross-sectional sample of 7273 Chinese women aged at least 50 years from the Guangzhou Biobank Cohort Study.
Results: Leg length had no significant association with education or father's occupation on bivariable testing. After including age of menarche in the model, education was associated with longer legs (0.45 cm longer per 10 years of education, 95% CI 0.20 to 0.71). Education was also associated with younger age of menarche (1.21 years younger per 10 years of education, 95% CI 1.09 to 1.34), which was in turn associated with shorter legs (0.23 cm shorter per year of menarche earlier, 95% CI 0.18 to 0.27).
Conclusions: In older Chinese women leg length is not a universal biomarker of childhood conditions, when proxied by her educational level and father's occupation. Nutritionally driven epigenetic influences operating over generations may constrain growth in very recently developed populations. Given the impact of childhood conditions on health, and the dearth of long-term records outside the industrialised world, a greater understanding of the influences on growth in the developing world is required.