molly-cliff/Meningitis-belt-location

Spatial distribution and geographic risk of meningitis epidemics across Africa # Updating the meningitis belt: associations between environmental factors and epidemic meningitis risk across Africa This repository contains R and Stata code used to create a logistic regression model and risk map for meningitis outbreaks across Africa. A 2003 study by Molesworth et al investigated the spatial distribution of meningitis epidemics in Africa occurring between 1980 and 1999 to determine the geographical risk of meningitis across Africa. This analysis was carried out to present an update on the associations between environmental variables and meningitis epidemic risk for meningitis across Africa and to understand if the geography of the meningitis belt and risk factors associated with meningitis have changed. We examined epidemics from 2003- 2022, assessing the impact of specific humidity, windspeed, dust, rainfall, and land coverage as well as including a population weighting analysis. We conducted a sensitivity analysis without the Democratic Republic of Congo (DRC), due to potential overreporting. # Repository Contents 1. Enviromental Cluster Analysis: Processes monthly temporal data into single surface comprising of climatic variable profiles. This is largely carried out using Principle component analysis and the K means algorithm to divide each seasonal variable raster into distinct clusters based on data similarity. This folder also includes code to extract average population density for each ADMN2 district within an African continental GADM shapefile. 2. Data Cleaning: Stata code to assign meningitis epidemics to their nearest ADMN2 district within the GADM shapefile (Note this is data cleaning for the full analysis as opposed to subsetted vaccination data). 3. Intersection Analysis: Maps epidemics reported to older adminstrative boundaries to exisiting (2024) ADMN2 districts using a WHO Polio shapefile due to the changing nature of district boundaries. ADMN2 level district intersections between the WHO Polio and GADM shapefiles were …