192 - Association Between 2023 Maui Wildfire Exposure and Perinatal Outcomes: A Comparative Interrupted Time Series Analysis

Publication Number: 2185.192

Miho Sodeno, Department of Global Health and Research, Juntendo University, Los Angeles, CA, United States; Rie Sakai-Bizmark, Lundquist Institute at Harbor-UCLA Medical Center, Torrance, CA, United States

Background: Wildfires and exposure to smoke have been linked to adverse perinatal outcomes such as preterm birth and low birth weight (LBW). The August 2023 Maui (Lahaina) wildfires caused unprecedented damage in recent U.S. history; however, prior studies have focused on excess mortality or respiratory and mental health effects in adults, and specific impacts on perinatal outcomes remain unknown.

Objective: To evaluate the impact of exposure to the 2023 Maui wildfires on the risk of perinatal outcomes using a Difference-in-Differences (DID) design.

Design/Methods: County-level monthly birth data for Hawaii (Jan 2018-Dec 2024) were obtained from the Centers for Disease Control and Prevention's Wide-ranging Online Data for Epidemiologic Research (CDC WONDER) database. Maui County was defined as the intervention group, and other counties (Honolulu, Hawaii) as controls. Exposure categories were defined by the estimated pregnancy stage during the Maui wildfire: T3 (late pregnancy; births in Sep-Nov 2023), T2 (mid-pregnancy; births in Dec 2023-Feb 2024), and T1 (early pregnancy; births in Mar-Jun 2024). Births in August 2023 were excluded. Dependent variables were counts of preterm birth, LBW, and abnormal conditions at birth (per CDC WONDER definitions, e.g., assisted ventilation or NICU admission). A linear regression model weighted by total births was applied, with robust standard errors clustered at the county level. Models adjusted for county-level maternal education, WIC participation, race, ethnicity, and fixed effects of month, year, and county. The parallel trends assumption was assessed using a joint test of pre-fire interaction terms.

Results: We identified 105,123 births across three counties in Hawaii (2018-2024). No significant pre-intervention trend differences were observed between Maui County and the comparison counties. DID analysis showed that during the T3 period after the wildfire, mothers in Maui County had higher rates of preterm births (+22.18 per 1,000; p < 0.01) and LBW (+19.95; p < 0.01) compared with changes in the control group before the fire. In contrast, during the T1 period, preterm and LBW rates declined (−17.52 and −21.76 per 1,000; p < 0.01 and p = 0.01), while abnormal conditions at birth increased (+26.42; p < 0.01).

Conclusion(s): Findings suggest that maternal exposure to the wildfire in various stages of pregnancy may have affected the variable outcomes of neonatal health and their care. As maternal exposure during pregnancy can be influenced by multiple pathways and future studies are warranted.

Preterm birth per 1,000 births by period and county
Rates in preterm birth rates (per 1,000 live births) among mothers in Maui County compared with control counties (Honolulu and Hawaii), 2018-2024. Rates increased in Maui during the T3 period (1-3 months after the wildfire), whereas rates decreased during the T1 period (7-10 months after the fire).

Low birth weight (LBW) per 1,000 births by period and county
Rates in low birth weight (LBW) infants ( <2,500 g) per 1,000 live births among mothers in Maui County compared with control counties (Honolulu and Hawaii), 2018-2024. The rate of LBW increased in Maui County during the T3 period but declined during the T2 and T1 periods relative to controls.

Difference-in-differences estimates of perinatal outcomes before and after the Maui wildfire
Difference-in-differences (DID) regression results for preterm birth, low birth weight, and abnormal conditions at birth among births in Maui County compared with control counties, 2018-2024. Models were adjusted for county-level maternal education, WIC participation, race, ethnicity, and fixed effects of month, year, and county.