Spatial and temporal determinants of particulate matter peak exposures during pregnancy and early postpartum

Background: Fine particulate matter (PM_2.5) exposure is an important environmental risk for maternal and children's health, with peak exposures especially those derived from primary combustion hypothesized to pose greater risk. Identifying PM_2.5 peaks and their contributions to personal exposure remains challenging. This study measured personal PM_2.5 exposure, characterized primary combustion peaks, and investigated their determinants during and after pregnancy and among Hispanic women in Los Angeles, CA. Methods: Continuous personal PM_2.5 exposure, Global Positioning System geolocation, and ecological momentary assessment surveys were collected from 63 women for 4 consecutive days in their 1^st trimester, 3rd trimester and 4–6 months postpartum. Based on the shape of PM_2.5 time-series, primary combustion peaks were identified, characterized (number, duration, area under the curve [AUC]), and linked with locations they occurred in. Zero-inflated generalized mixed-effect models were used to examine the spatial and temporal determinants of PM_2.5 peak exposures. Results: A total of 490 PM_2.5 peaks were identified from 618 person-days of monitoring. Spending an additional minute at parks and open spaces was related to smaller (AUC decreased 3.1 %, 95 % CI: 1.5 %–4.6 %) and shorter (duration decreased 1.7 %, 0.5 %–2.9 %) PM_2.5 peak exposure. An additional minute in vehicular trips also related to smaller and shorter peak exposure (AUC and duration decreased 2.5 %, 1.2 %–3.7 % and 1.8 %, 1.0 %–2.6 %, respectively). However, an additional minute at industrial locations was associated with greater number (3.6 %, 2.0 %–5.2 %), AUC (1.6 %, 0.1 %–3.2 %) and duration (1.0 %, 0.0 %–2.1 %) of personal PM_2.5 peak exposure. Conclusions: This study demonstrates the potential to statistically identify exposure to primary combustion PM_2.5 peaks and understand their determinants from personal monitoring data. Results suggest that visits to parks and open spaces may minimize PM_2.5 peak exposures, while visiting industrial locations may increase them in and around pregnancy.