Determine the most important predictors of diarrhoea in children under five in South and Southeast Asia (Pakistan, India, Nepal, Bangladesh, Myanmar, Cambodia, Philippines, Timor-Leste) by exploring the spatiotemporal association between diarrhoeal incidence and various behavioural, socio-demographic, and environmental factors.
Dr Syeda Hira Fatima
Global Ecology | Partuyarta Ngadluku Wardli Kuu, Flinders University, Adelaide, Australia
September 2024
e-mail
and
Prof Corey J. A. Bradshaw
Global Ecology | Partuyarta Ngadluku Wardli Kuu, Flinders University, Adelaide, Australia
September 2024
e-mail
Project collaborators: Dr Melinda Judge, Prof Peter Le Souëf, Dr Lewis Weeda, Naomi Henry
Fatima, SH, MA Judge, PN Le Souëf, CJA Bradshaw. Impact of climate change on diarrhoea risk in low- and middle-income countries. In review.
DHSDiarrProcessing_1.R: R code to load and merge DHS survey and GPS data and select the appropriate list of variables.DHSDiarrProcessing_2.R: R code to preprocess, recode, and create new variables where necessary.DHSDiarrProcessing_3.R: R code for imputation of variables with missing data.DHSDiarrProcessing_4.R: R code for processing of raster data.DHSDiarrProcessing_5.R: R code for processing of variables at the cluster level and standardization.DHSDiarrAnalysis.R: R code to reproduce the resampled boosted regression tree analysis for determining the relationships between probability of diarrhoea, and socio-economic, maternal, child, climate data (full dataset).DHSDiarrAnalysisIndaOnly.R: R code to reproduce the resampled boosted regression tree analysis for determining the relationships between probability of diarrhoea, and socio-economic, maternal, child, climate data (India only).
- DHSclusterLevelDiarrData.csv.zst: Demographic and Health Surveys data summarised by cluster with central parameter (mean, proportion, etc.) and variance per cluster. Overlaid (cluster-level) climate data derived from WorldClim bioclimatic variables (mean annual temperature, temperature annual range, total annual precipitation, precipitation seasonality, and precipitation of the driest quarter). Unzip .csv data file prior to analysis. The file is a high-compression .zst of the .csv base file; use the following command in Terminal to decompress:
zstd -d 'DHSclusterLevelDiarrData.csv.zst'
cowplot,dismo,dplyr,foreign,gbm,GGally,ggplot2,ggpubr,gridExtra,haven,leaflet,mice,raster,reshape2,sf,sp,spatstat.random,tidyr,truncnorm,usdm
