Air pollution and preterm birth: A time‐stratified case‐crossover study in the San Joaquin Valley of California

Abstract

Background Air pollution is linked to preterm birth (PTB), but existing studies are primarily focused on chronic exposures, conducted in areas with moderate pollution, and/or subject to confounding. Objectives We investigated short‐term associations between two pollutants [particulate matter <2.5 microns (PM2.5) and ozone] and PTB, and estimated excess PTB cases potentially attributed to these pollutants. Methods This time‐stratified case‐crossover study includes 196,970 singleton pregnancies affected by PTB and early term birth from the San Joaquin Valley (SJV), California, USA (2007–2015). Daily ozone and PM2.5 concentrations were estimated by the SJV Air Pollution Control District and geospatially linked to maternal zip code. We used conditional logistic regression models to estimate the odds ratio (OR) and 95% confidence intervals (CI) for the associations between an interquartile range (IQR) increase in pollutants and very preterm (VPTB, 20–34 weeks), moderate preterm (MPTB, 34–36 weeks) and early term births (ETB, 37–38 weeks). We adjusted all models for co‐pollutants and meteorological factors. Results During warm seasons (May‐October), an IQR increase in ozone was associated with 9–11% increased odds of VPTB from lag 0 (ORlag0 1.09, 95% CI 1.04,1.16) to lag 7 (ORlag7 1.11, 95% CI 1.04,1.16). Findings were consistent for MPTB and ETB. Ozone was potentially responsible for an excess of 3–6 VPTBs, 7–9 PTBs and 24–42 ETBs per 1,000 singleton deliveries. During cold seasons (November‐April), increased PM2.5 exposure was associated with 5–6% increased odds of VPTB beginning at lag 3 (ORlag3 1.06, 95% CI 1.02,1.11). PM2.5 was associated with an excess of 1–3 VPTBs, 0–3 MPTBs and 6–18 ETBs per 1,000 singleton deliveries. Conclusions PM2.5 and ozone are associated with increased risk of VPTB, MPTB and ETB within one week of exposure and are potential contributors to the increasing PTB trend. More research is needed to further understand the role of air pollution on PTB risk.

The Temporal Relationship Between the Coronavirus Disease 2019 (COVID-19) Pandemic and Preterm Birth

“…[8][9][10][11] Conversely, the pandemic, particularly during the most extensive lockdowns, has resulted in less pollution and other adverse environmental factors that are considered to be potentially protective of preterm birth. [12][13][14] Nevertheless, the epidemiologic studies that have evaluated the relationship between the COVID-19 pandemic and preterm birth have had inconsistent results. Some have shown no change, some have shown a decreased risk, and some have shown an increased risk of preterm birth after the onset of the pandemic.…”

Selecting exposure data and identifying relevant exposure windows for birth outcome studies

“…In contrast, Ha and colleagues investigated the associations between short-term exposure to PM 2.5 and ozone (O 3 ) and preterm birth in the San Joaquin Valley, California, using a time-stratified case-crossover design. 4 A strength of using this study design is that adjusting for time-invariant factors, such as race/ethnicity, was not necessary in the statistical analysis. They found that PM 2.5 exposures at lag 5 and at lags 5-7 were associated with very preterm birth (gestational age <34 weeks) and early term birth (gestational age between 37 and 39 weeks), respectively, but associations were not found for preterm birth at any lags.…”

“…Nevertheless, they found some associations between preterm birth and the summary measurement of PM 2.5 and diesel PM, from community mapping tools, indicating that places with higher levels | 91 SPECIAL ISSUE of PM 2.5 have greater risk for preterm birth. In contrast, Ha et al 4 focused on short-term pollutant exposures with lags up to seven days. The heterogeneous results by lag days indicated that exposure to PM 2.5 or O 3 is time-sensitive, but it is unclear why some specific lag days had positive associations with very preterm birth, whereas earlier lags (e.g., lag 1) did not.…”

“…They did not find any adverse relationships for preterm birth with the traffic measures but found associations with PM 2.5 and diesel PM, using summarised data obtained from CalEnviroScreen. In contrast, Ha and colleagues investigated the associations between short‐term exposure to PM 2.5 and ozone (O 3 ) and preterm birth in the San Joaquin Valley, California, using a time‐stratified case‐crossover design 4 . A strength of using this study design is that adjusting for time‐invariant factors, such as race/ethnicity, was not necessary in the statistical analysis.…”

Impacts of climate change on reproductive, perinatal and paediatric health

“…Most research on air pollution and preterm birth has focused on long‐term exposures (during the entire pregnancy or trimester‐specific exposure). Ha and colleagues 11 used a case‐crossover design to measure the extent to which acute air pollution exposure triggers preterm birth. This study design allows for identification of the health effects of acute exposures and inherently controls for time‐invariant confounding, as each participant with the outcome is compared with themselves at different time points.…”