Heat Waves, Early Birth, and Vulnerable Populations: Epidemiological and Spatial Analyses in the United States

Loading...
Thumbnail Image

Authors

Fitch, Amy

Issue Date

2025

Type

Dissertation

Language

en_US

Keywords

Early-term birth , Heat wave , Preterm birth , Vulnerable populations

Research Projects

Organizational Units

Journal Issue

Alternative Title

Abstract

Heat waves are a serious and growing threat to human health and society. While there is no single definition of a heat wave, elevated morbidity and mortality during extreme heat is consistently observed in addition to other societal impacts, such as strain on the energy infrastructure, drought, and wildfire risk. This dissertation examines extreme heat events through a public health and equity lens. In Aim 1 we estimate acute effects of heat waves on preterm and early-term birth in eight US states from 1990-2017, adjusting for seasonal patterns of conception. In Aim 2 we evaluate if the acute effects of heat waves on early birth vary by individual-level factors (maternal age or education) or area-level factors (impervious land cover or social deprivation). In Aim 3 we evaluate spatial patterns in overlapping social vulnerability and heat in nine western US cities during the series of heat waves that occurred in June-July 2021. For Aims 1 and 2, we obtained temperature data from the novel High-resolution Urban Meteorology for Impacts Dataset. Daily mean temperatures were extracted at the zip code tabulation area (ZCTA) level for up to 28 years and linked to birth records from eight US states: California, Florida, Georgia, Kansas, Nevada, New Jersey, North Carolina, and Oregon. Defining heat waves in multiple ways, we performed time-stratified case-crossover analyses to estimate acute effects of heat waves on preterm (<37 weeks) and early-term (37-38 weeks) birth. We calculated state-specific and pooled odds ratios (Aim 1). We adjusted for seasonal patterns of conception and error in last menstrual period reporting by including the probability of early birth among ongoing pregnancies at risk. In Aim 2 we conducted stratified analyses for both preterm and early-term birth to determine if the association between heat wave and early birth was modified by individual level and area-level factors: maternal age, maternal education, ZCTA-level land cover and ZCTA-level social deprivation. The main analysis included 2,966,661 early-term and 945,869 preterm births occurring from May - September across the eight states from as early as 1990 to 2017. Results showed modestly elevated odds of early-term birth for heat waves occurring in the four days preceding birth. The adjusted pooled odds ratios across heat wave metrics ranged from 1.008 to 1.022 for preterm birth and from 1.007 to 1.020 for early-term birth. For the association of acute heat wave exposure and early-term birth, we found evidence of effect modification across all stratification variables (Aim 2). Stronger effects were evident among younger and less educated mothers as well as those living in areas of higher social deprivation and higher impervious land cover. For preterm birth, the highest heat-related risk was less consistently linked to socioeconomic disadvantage, as associations were also evident among those living in low deprivation areas, older mothers, and more educated mothers. In the third aim, we investigated a series of heat waves that occurred in June-July 2021 in nine western US cities (Boise, Las Vegas, Los Angeles, Oakland, Portland, Sacramento, Salt Lake City, San Francisco, and Seattle) to determine to what extent vulnerable populations experienced a disproportionate burden of extreme heat. We defined vulnerability at the census tract level using the Agency for Toxic Substances and Disease Registry’s social vulnerability index (SVI) and linked to temperatures modeled at a 1km2 resolution to capture urban heat island gradients. We examined correlations between SVI and minimum and maximum temperatures averaged across heat wave days. We found that vulnerability was positively correlated with heat wave temperatures in most cities, the exceptions being Oakland and San Francisco. There was significant spatial clustering, or hotspots, in the same cities, of census tracts that experienced both high vulnerability and heat wave temperatures. Such areas may be appropriate targets for heat wave mitigation efforts, such as cooling centers or residential greening efforts. Using both epidemiological and spatial analysis methods, this dissertation contributes to our knowledge of the effects of heat wave exposure on early birth risk and identifies populations most affected by heat waves.

Description

Citation

Publisher

License

Journal

Volume

Issue

PubMed ID

DOI

ISSN

EISSN