R code and results attached to the publication:
Masselot P, Sera F, Schneider R, ..., Gasparrini A. Differential mortality risks associated to PM2.5 components: a multi-country multi-city study. Epidemiology. 2022;33(2):167-175 https://doi.org/10.1097/ede.0000000000001455
Mortality data are not available currently due to restricted data sharing agreement between the collaborators of this study. Therefore, only scripts 3 to 5 are fully reproducible.
The Data folder contains data needed to run scripts 3 to 5, i.e. results from the first-stage city-specific models and PM composition data.
The R code to reproduce the analysis is available. Scripts are meant to be executed in order:
- 0_PrepData.R Loads and and links mortality, PM2.5, composition and city-specific characteristics. Performs city selection.
- 1_DataSummary.R Creates Table 1 providing descriptive statistics.
- 2_FirstStage.R Runs the first-stage model on each selected city to produce city-level relative risks.
- 3_SecondStage_MetaRegComposition.R Pools first-stage results in a meta analysis including city-specific compositions and socio-economic indicators.
- 4_Plots.R Produces the figures of the paper.
- 5_SupplementaryResults.R Produces supplementary results and eFigures.
Figures and Tables generated for the publication (including supplemental ones) are available in the Results folder of this repository.
This work was supported by the Medical Research Council of UK (Grant ID: MR/M022625/1), the Natural Environment Research Council of UK (Grant ID: NE/R009384/1), and the European Union’s Horizon 2020 Project Exhaustion (Grant ID: 820655).