Don't stress: early life conditions, hypertension and selection into associated risk factors

Early life conditions have been linked to various domains of later life health, including cardiovascular outcomes. Using life history data from 13 European countries, I find that childhood socioeconomic status and measures of childhood health are related to hypertension, although there is cross country heterogeneity in these effects. I account for potential omitted variable bias by using aggregate measures of public health at birth, which are plausibly exogenous to the individual. I find that infant mortality at birth is positively related to hypertension, even allowing for cohort effects, and controlling for GDP at birth. Results imply that improvements in early life conditions in Europe led to an overall reduction in the hypertension rate of between 3 and 6 percentage points, for the cohort born 1931-1935, relative to the cohort born 1956-1960. An alternative strand of literature in epidemiology links contemporaneous factors, such as work place environment, to heart disease. However, theories of life cycle decision making suggest that individuals may be selected into these adverse environments and behaviours on the basis of their initial conditions. I demonstrate a strong association between early environment and these risk factors. Results imply that these should therefore be viewed as outcomes which lie on the causal pathway between initial conditions and later outcomes, in which case ignoring this selection will misattribute at least part of the effects of early life environment to current circumstance. This has important policy implications for targeting hypertension as it indicates that emphasis should also be placed on combatting disadvantage across the life course, rather than just factors which only manifest themselves in adulthood.