EUVIMEDEUVIMED entdecken
Crossref

Accounting for Spatiotemporally Heterogeneous Measurement Error in Gridded PM2.5 Exposure Estimates: A Spatiotemporal Varying-Coefficient Calibration Approach

Jianxiang Tang, Feng Hong, Xiuling Zhu, Wenqian Zeng, Jingru Xu, Mi Cheng, Shaoxiong Yang, Baimayangji, Bing Guo, Xing Zhao

Environmental Pollution · 2026-11 · 129538

DOI: 10.1016/j.envpol.2026.129538 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Environmental Pollution, 2026-11. Verfasst von Jianxiang Tang und 9 weiteren Autor:innen. Beteiligte Einrichtung: West China School of Public Health and West China Fourth Hospital, Sichuan University, Chengdu, China. Die Arbeit nennt 29 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

High-resolution gridded PM2.5 products enable large-scale epidemiological studies, but their use as residential exposure surrogates is subject to measurement error that may vary across space and time. Existing error calibration approaches often rely on predefined geographic and temporal strata, which may fail to capture local variation in measurement errors, thereby biasing health-effect estimates. We developed a spatiotemporal varying-coefficient calibration model (STVCCM) that used natural spline functions of spatial coordinates and calendar month to characterize spatiotemporal variation in the calibration relationship. The STVCCM was applied to the China Multi-Ethnic Cohort, and the exposure estimates obtained with the STVCCM were compared with those from uncalibrated models, models calibrated with standard regression (RC), and models calibrated with stratified regression (SRC). Measurement error, defined as the difference between gridded PM2.5 predictions and the corresponding monitoring observations, showed substantial spatiotemporal heterogeneity: mean absolute error (MAE) ranged from 3.5 to 14.7 μg/m3 across monitoring sites and was higher in January than in July (10.8 vs 4.7 μg/m3). At both the 10 μg/m3 and 1 SD scales, STVCCM yielded the highest numerical ORs for three of five outcomes. In simulation, STVCCM had the smallest observed absolute bias and root mean square error under the specified spatiotemporally heterogeneous error mechanism. These findings suggest that STVCCM, through its flexible representation of spatiotemporal calibration relationships, shows promise for addressing heterogeneous measurement error in PM2.5 epidemiologic studies.

Schlagwörter

Stichwörter der Autor:innen

PM(2.5) exposure calibrationmeasurement errorregression calibrationthe China Multi-Ethnic Cohortvarying-coefficient model

Autor:innen und Institutionen

  • West China School of Public Health and West China Fourth Hospital, Sichuan University, Chengdu, China.
  • School of Public Health, the Key Laboratory of Environmental Pollution Monitoring and Disease Control, Ministry of Education, Guizhou Medical University, Guiyang, China.
  • Chongqing Municipal Center for Disease Control and Prevention, Chongqing, China.
  • Wuhou District Center for Disease Control and Prevention, Chengdu, China.
  • School of Public Health,Kunming Medical University, Kunming, China.
  • School of Medicine, Tibet University, Lhasa, China.

Zitieren

APA

Tang, J., Hong, F., Zhu, X., Zeng, W., Xu, J., Cheng, M., Yang, S., Baimayangji, Guo, B., & Zhao, X. (2026). Accounting for Spatiotemporally Heterogeneous Measurement Error in Gridded PM2.5 Exposure Estimates: A Spatiotemporal Varying-Coefficient Calibration Approach. Environmental Pollution, 129538. https://doi.org/10.1016/j.envpol.2026.129538

Vancouver

Tang J, Hong F, Zhu X, Zeng W, Xu J, Cheng M, et al. Accounting for Spatiotemporally Heterogeneous Measurement Error in Gridded PM2.5 Exposure Estimates: A Spatiotemporal Varying-Coefficient Calibration Approach. Environmental Pollution. 2026:129538. doi:10.1016/j.envpol.2026.129538

BibTeX
@article{Tang2026,
  author = {Tang, Jianxiang and Hong, Feng and Zhu, Xiuling and Zeng, Wenqian and Xu, Jingru and Cheng, Mi and Yang, Shaoxiong and Baimayangji and Guo, Bing and Zhao, Xing},
  title = {Accounting for Spatiotemporally Heterogeneous Measurement Error in Gridded PM2.5 Exposure Estimates: A Spatiotemporal Varying-Coefficient Calibration Approach},
  journal = {Environmental Pollution},
  year = {2026},
  pages = {129538},
  doi = {10.1016/j.envpol.2026.129538}
}
Kostenlos starten →