223 - The Spectrum of Housing Instability and Adverse Perinatal Outcomes

Publication Number: 223.6683

Daria Murosko, The Children's Hospital of Philadelphia, Philadelphia, PA, United States; Alexander S. Hill, Children's Hospital of Philadelphia, Philadelphia, PA, United States; Michael J. Luke, Children's Hospital of Philadelphia, Philadelphia, PA, United States; Scott A. Lorch, The Children's Hospital of Philadelphia, Philadelphia, PA, United States

Background: Families experience housing instability on a spectrum, ranging from early signs such as challenges affording housing costs or forced mobility due to unaffordability, to homelessness. Low-income families are at greatest risk. While homelessness during pregnancy is linked to poor perinatal outcomes, the effects of early signs of housing instability are poorly defined.

Objective: To identify the prevalence of the housing instability spectrum among pregnant individuals and its association with adverse perinatal outcomes.

Design/Methods: We used survey data from the Pregnancy Risk Assessment and Monitoring System (PRAMS) and birth certificates from 2016–2022 to analyze a weighted cohort of 9117567 pregnant individuals across 24 states and New York City. We constructed the housing instability spectrum based on stressors reported in the year prior to birth, ranging from 1) unaffordability only, 2) mobility associated with unaffordability (defined as presence of both mobility and unaffordability), and 3) homelessness. Outcomes of interest were preterm birth (PTB, < 37 weeks gestation), low birth weight (LBW, < 2500g), and parental report of infant death. Multivariate logistic regression determined the association of housing instability and each outcome, adjusted for maternal and infant factors. We repeated these models stratified by income level.

Results: Prevalence of housing instability varied by state (Figure 1). There were significant differences in infant and maternal demographics across the housing instability spectrum (Table 1). Mothers who reported homelessness were more likely to have public insurance, to identify with Black or Other race, to smoke, and had lower educational attainment. In unadjusted models, odds of PTB and LBW were increased across the entire spectrum of housing instability. In adjusted models, unaffordability was only associated with PTB (OR 1.08, 95% CI 1.00-1.16) and homelessness was associated with PTB (OR 1.14, 95% 1.01-1.29) and LBW (OR 1.17, 95% CI 1.05-1.30) (Table 2). Among lower earners, odds of LBW and death were increased in the group endorsing homelessness greater odds of LBW and infant death. Among higher earners, the group endorsing unaffordability challenges had greater odds of PTB and LBW; odds of death were lower in the group that had mobility associated with unaffordability.

Conclusion(s): The link between housing instability and perinatal outcomes varies across the spectrum. While homelessness is most consistently associated with adverse outcomes, proactive interventions aimed at addressing unaffordability may be protective, particularly for those with higher incomes.

Table 1: Maternal Demographics by Spectrum of Housing Instability of 9,117,567 Dyads in PRAMS 2016-2022


Table 2: Odds of adverse birth outcomes across the housing instability spectrum, compared to no instability and stratified by maternal income


Figure 1. Annual prevalence of housing instability spectrum indicators by states, 2016 – 2022.
Housing instability is presented by unaffordability challenges only (Panel A), unaffordability associated with mobility (Panel B) and homelessness (Panel C), with darker colors indicating greater rates. New York City data are not shown.