Background The maintenance of physical and cognitive function for the maximal period of time is a key component of healthy ageing, associated with continued independent living, better quality of life and reduced morbidity and mortality risk. There is growing interest in investigating life course influences on healthy ageing to identify potential intervention targets, beyond those already identified by chronic disease epidemiology.

Using data from the MRC National Survey of Health and Development (NSHD), we assess:

1. the relative importance of socioeconomic position (SEP) in childhood and adulthood across a range of physical and cognitive functional ageing outcomes at age 60–64; 2. the contribution of key social, behavioural and developmental intermediary factors to the relationship between child SEP and functional ageing

Methods Ageing outcomes (lung function, grip strength, chair rise time, standing balance, timed up and go (TUG), verbal memory, processing speed and simple reaction time) were regressed on childhood and adult SEP ridit scores (calculated as the proportion of the population higher than the midpoint for each category) and sex. The ridit score coefficient is the slope index of inequality (SII), interpreted as the absolute difference in outcome between the hypothetical top and bottom of the SEP gradient. These were converted to the relative index of inequality to provide an estimate of the relative SEP difference across outcomes and over time. Regression models were then adjusted to examine the influence of potential mediators (education, smoking, BMI, height and childhood cognition) on the SII.

Results Substantial childhood and adult socioeconomic gradients were observed in all physical and cognitive outcomes. The hypothetical top of the childhood SEP distribution performed between 9 and 18 per cent better, relative to the hypothetical bottom (p ≤ 0.05 for association between childhood SEP with all outcomes). These associations persisted on adjustment for adult SEP, with the exception of standing balance. The hypothetical top of the adult SEP distribution performed between 6 and 26 per cent better, relative to the hypothetical bottom (p ≤ 0.05 for association between adult SEP with all outcomes). With the exception of processing speed (in women only), chair rise time, and TUG, associations between adult SEP and outcomes persisted on adjustment for childhood SEP. Adjusting for potential mediators attenuated specific associations.

Conclusion Child and adult SEP were independently associated with physical and cognitive ageing at age 60–64. The different social, behavioural and developmental pathways partly mediating these associations may guide appropriate intervention strategies.