The Long Reach of Childhood Health

Featured in print Bulletin on Aging & Health

Policy makers everywhere are interested in the health of their populations, and therefore concerned about the link between health and economic status. Whether one looks across countries, across individuals in the same country, or across time within a given country, the story is the same - being richer is associated with better health. While this link is well-established, however, the direction of causality is not. Does low economic status lead to poor health, or does poor health lead to low economic status, or both?

In Causes and Consequences of Early Life Health (NBER Working Paper 15637), researchers Anne Case and Christina Paxson focus on a specific aspect of this question: the association between childhood health and economic and health outcomes later in life.

The authors note that relatively little is known about the impact of early life health on adult labor force outcomes because there are few data sets that follow individuals from birth through adulthood. In light of this challenge, researchers have recently begun to use height as a marker of early life health and nutrition. The authors use five longitudinal data sets from the U.S. and Britain to examine the impact of height on later life outcomes.

Previous research suggests that final adult height depends on several factors, including genes, environmental conditions, and interactions of these. Differences in average heights across countries and the increase in heights in the developed world in the 20th century are believed to be due largely to environmental factors. Environmental conditions thought to affect height include prenatal factors such as maternal smoking and low birth weight, as well as early childhood exposure to poor nutrition, infections, and psychosocial stress. Some research suggests that these factors affect health throughout life.

Turning to their analysis, the authors find that height is associated with improved economic outcomes in all five of their data sets. Each additional inch of height is associated with an increase in schooling of 0.05 to 0.16 years and an increase in the probability that men work of 0.2 to 0.6 percentage points. Height affects earnings as well - using the midpoint of their estimates, moving from the 25th to 75th percentile of height is associated with an 8 percent increase in earnings.

Height is associated with better adult health as well. Taller individuals report better overall health and fewer long-standing illnesses and are less likely to be disabled. In older adulthood, taller individuals have fewer problems with activities of daily living and are less likely to be depressed. Taller adults also perform better on cognitive tests.

Does childhood health affect adult outcomes primarily by influencing educational attainment, which then affects adult labor force and health outcomes, or does it have independent, long-run effects on these measures? To answer this, the authors re-estimate their models controlling for schooling. They find that the effects of height decline but are still significant, suggesting that childhood health has direct effects on adult outcomes.

These results underscore the importance of early childhood, but shed little light on which specific aspects of childhood are being captured by height. To explore this, the authors conduct an analysis with the National Longitudinal Survey of Youth (NLSY) and the follow-on study of children of NLSY respondents. This data set includes information on the children's educational outcomes, health during childhood, and the prenatal environment to which they were exposed. Importantly, the data set includes siblings, which allows the authors to control for genetic material and unmeasured environmental factors siblings share, so they can test whether measured differences in the environment (such as in exposure to maternal smoking) are associated with differences in outcomes.

The authors find that taller children have higher educational attainment and are more likely to be in the right grade for their age. They verify that these differences are not the result of taller children starting school earlier, nor of taller children having higher self-esteem. Rather, they are consistent with taller children having higher cognitive ability.

Finally, they explore why some children are taller than others, particularly within sibling pairs. They find that birth length and weight are positively associated with adult height. Maternal smoking and drinking affect birth length and weight, but have no independent effect on adult height. Lack of prenatal care also depresses adult height.

The authors note that understanding the long-run consequences of child health may help policy makers assess the benefits of interventions that improve child health and may be useful in making projections of trends in adult morbidity and disability. Their findings suggest "height, which reflects (in part) environmental conditions in the prenatal and early childhood periods, can be used to understand the long-run consequences of childhood health." While they caution that the environments in which children are being raised today are quite different from those experienced by the oldest individuals in their data, the fact that children born in the last two decades have poorer cognitive performance and schooling outcomes "indicate that child health is likely to remain an important determinant of health and economic status once these children reach adulthood."

The authors acknowledge funding from the Demography of Aging Center at Princeton University, funded under National Institute of Aging grant P30 AG024361.