This repository stores the updated R code to reproduce a simpler and simulated version of the linkage process presented in the article:
Vanoli J, Mistry MN, De La Cruz Libardi A, Masselot P, Schneider R, Ng CFS, Madaniyazi L, Gasparrini A. Reconstructing individual-level exposures in cohort analyses of environmental risks: an example with the UK Biobank. Journal of Exposure Science & Environmental Epidemiology. 2024;34(6):1012-1017. DOI: 10.1038/s41370-023-00635-w. PMID: 38191925. [freely available here]
This work was supported by the Medical Research Council-UK (Grant IDs: MR/Y003330/1 and MR/R013349/1), the European Union’s Horizon 2020 Project Exhaustion (Grant ID: 820655), Nagasaki University “Doctoral Program for World-leading Innovative and Smart Education” for Global Health (WISE), KENKYU SHIDO KEIHI (“the Research Grant”) and KYOIKU KENKYU SHIEN KEIHI (“the Stipend”).
- data: this folder contains the daily exposure predictions of PM2.5 for a coastal area of South England over a 1x1 km grid. The data are split into yearly datasets for 2017, 2018, and 2019. The datasets were produced using a method described in a previously published work (freely available here). However, the same linkage process can be performed with any other environmental exposure dataset with similar features, simply by adapting the R code (particularly 02_exposure_series.R).
01_resid_health_data.R: the script defines sample datasets of individuals, including information regarding residential histories located in the specific UK region (for graphical purposes).02_exposure_series.R: the script produces the linkage between residential histories and exposure series following steps 1) and 2) of the article.03_plots.R: the script produces the following graphs: a map of the locations for all the subjects, a map for the residential locations corresponding with subject A, a plot of the full PM2.5 series at all the residential locations for subject A, and a plot of the final exposure series for subject A (step 2 of the article).