Health Across the Life Course

Abstract

This dissertation consists of three chapters examining health across the life course. A common theme is that health reflects cumulative exposures to risk or protective factors in one’s own life and in previous generations.

Two chapters examine inequalities in birth outcomes by race/ethnicity and immigrant generation using administrative birth records linked between mothers and daughters. Births to foreign-born women generally have better outcomes than births to US-born women of the same race/ethnicity, but it is unclear whether this immigrant advantage persists in subsequent generations.

The first chapter focuses on birthweights of descendants of foreign-born black women, using Florida birth records from 1971-2015. Following an initial advantage relative to US-born black women, birthweights of descendants of black immigrant women converge toward those of descendants of US-born black women within one generation. In contrast, descendants of both US-born and foreign-born white and Hispanic women maintain relatively high birthweights in both generations. These results suggest an environmental exposure specific to black women.

The second chapter focuses on other immigrant populations using California birth records from 1978-2015. Initial foreign-born advantages in low birthweight and preterm birth are present among most racial/ethnic groups with the exception of Asian women, though there is substantial heterogeneity within this broad ethnic category. In the next generation, advantages diminish for most racial/ethnic groups while an immigrant disadvantage in low birthweight emerges for descendants of Asian women, driven by descendants of Asian Indian, Cambodian, and Filipino women.

The final chapter turns to the end of the life course and assesses how changes in physical functioning predict subsequent mortality. Performance-based measures of physical functioning have been proposed as objective alternatives to self-reports that could better detect declines, but this advantage has not been verified. Using the Social Environment and Biomarkers of Aging Study of older Taiwanese adults, I find that, conditional on initial performance, steeper declines are predictive of increased mortality risk. However, conditional on current performance, previous changes do not improve predictions of mortality, suggesting that the benefit of repeated tests lies primarily in updating information about functioning, not in detecting subtle changes that potentially reveal underlying health deterioration.