Grow fast, die young? The causes and consequences of adult height and prolonged growth in nineteenth century Maastricht

Background: Both adult body height and the developmental growth trajectory have been found to be important predictors of later-life mortality. However, evidence for these relationships largely comes from contemporary populations, where most people live until old-age. It is an open question how height and growth impact later-life mortality in a population where death before old-age is more common. We therefore study the causes and mortality consequences of height and growth in a high-mortality, nineteenth-century Dutch population.

Methods: We exploit a unique dataset from three sources: conscription records with late-adolescent height, standing militia registers with adult height, and individual cause-of-death and age-at-death information. Our study is set in the Dutch city of Maastricht. To determine the causes of height and growth (either early-life environmental conditions or shared family inheritance), we use Pearson's correlation tests and multilevel linear models. To determine height and growth's consequences, we use survival analyses, for all-cause mortality and cause-specific mortality.

Results: Regarding the causes of height and growth, we find that prolonged growth and adolescent height are more strongly associated with external environmental factors than shared family inheritance. Adult height is more strongly related to shared family inheritance. Regarding the consequences of height and growth, we find that being taller in adulthood and growing faster are significantly associated with an increased hazard of death for the all-cause mortality model.

Conclusions: While we find the 'usual suspects' for the causes of height and growth, our findings for the conse-quences are surprising: the tallest individuals who grow the fastest have the highest hazard of death. Our results may be explained by a selection effect: the tall, fast growers may be the least-selected in early-life, and are therefore more vulnerable than their peers in adulthood.