Environmental noise, air pollution, and momentary stress: A context- and mobility-based approach.

Air pollution and noise exposure adversely affect individuals' physical and mental health. Static environmental exposure measurement methods lack accuracy and introduce estimation bias in the relationship between environmental exposure and health outcomes. Contextual factors and temporal effects play a significant role in influencing this association due to non-stationary effects. However, few studies have considered mobility-based exposure, contextual factors and temporality together in environmental health research. This study investigates the associations between noise and PM2.5 and momentary stress level (MSL) using a mobility-based measurement method, considering both contextual and temporal effects. It collected real-time exposure data from participants using mobile sensing technologies (e.g., GPS tracking) and geographic ecological momentary assessment (GEMA). The results show that daily routines and interactions with family or friends are linked to the highest PM2.5 exposure, whereas work/study contexts and settings spent with colleagues are associated with the lowest PM2.5 and noise levels. Travel is associated with the highest noise exposure. Further, travel and work/study contexts and being with colleagues are significantly and positively related to MSL compared to daily routines and being alone separately, while being with family members is related to lower MSL. The association between noise and MSL is significant (effect size: 0.003-0.008, temporal range: 1 minute-2 hours), particularly in entertainment, work/study, daily routines, when alone, with friends, or with family. PM2.5's association with MSL becomes non-significant when contextual factors are included. This study highlights the importance of addressing temporal and contextual uncertainty in mitigating the uncertain geographic context problem.