Cycles of Disadvantage: Eviction and Children’s Health in the United States

Abstract

The United States is in the midst of a housing affordability crisis, driven by decades of stagnant wages, little construction of low-cost housing, few tenant or foreclosure protections, anemic investment in public housing, and rising demand in large cities. Evictions have risen apace, forcing families out of their homes and neighborhoods, straining their finances and mental health, and disrupting their daily routines. Of the cohort of infants born in large US cities between 1998 and 2000, 1 in 7 were evicted by the time they turned 15, including 1 in 4 born into deep poverty. Yet despite how many families are affected, remarkably little research has examined what eviction is doing to the public’s health. This dissertation examines longitudinal connections between eviction and children’s health, from birth through middle childhood; I am among the first to do so. Throughout, I analyze data from the Fragile Families and Child Wellbeing Study, a national urban birth cohort. In Chapter 1, I provide evidence on whether eviction increases children’s risk of lead poisoning, a health condition tied to the aging and low-quality housing evicted children are often forced to move into. Using a proportional hazards marginal structural model, I estimate that each eviction a child experiences from birth through age 5 increases their prospective hazard of lead poisoning by 2.45 times (p=0.064; 95% confidence interval=0.95, 6.32). This risk compounds as evictions accumulate. In Chapter 2, I test whether families with children born premature, with low birthweight, or with other medical complications are more likely to be evicted in the first five years of children’s lives. I fit proportional hazards models using piecewise logistic regression, controlling for an array of confounders and applying inverse probability of selection weights. I find that being born low birthweight/preterm is associated with a 1.74-fold increase in children’s hazard of eviction (95%CI=1.02,2.95), while lengthy neonatal hospital stays were independently associated with a relative hazard of 2.50 (CI=1.15,5.44), compared to uncomplicated births. Given recent findings from other researchers that unstable housing during pregnancy is associated with adverse birth outcomes, my results suggest eviction and health may be cyclical and co-constitutive. These results are also important for eviction and health research, indicating that health can drive eviction just as eviction can drive health. In Chapter 3, I ask whether evictions occurring at specific developmental periods (infancy, early childhood, middle childhood) are associated with children’s cognitive skills at age 9. My four cognitive outcomes included executive functioning/memory (WISC-IV Digit Span), mathematical reasoning (Woodcock-Johnson Applied Problems test), symbolic learning and written language skills (Woodcock-Johnson Passage Comprehension test), and receptive vocabulary and verbal skills (Peabody Picture Vocabulary Test). I find that children who were evicted in middle childhood scored 0.2 to 0.4 standard deviations (SDs) lower on all four cognitive skills tests than children who had not been evicted. I also find strong, marginally significant associations between eviction in infancy and cognitive scores eight years later, of roughly 0.25 SDs. No associations were found with early childhood eviction. Together, these chapters suggest eviction and poor health may be cyclical, ratcheting children down the ladder of well-being and opportunity and imperiling their prospects of a healthy adulthood. At minimum, they identify eviction as a powerful predictor of which children are in need of extra monitoring and support from the health care, social safety net, and public education systems, and of which medically vulnerable children are at risk of eviction.