gpaasi/malaria-scd-uganda-shared-component-mapping

Code, workflow, and supporting files for spatial co-clustering analysis of malaria and sickle cell disease admissions across Ugandan health facility catchments # Spatial co-clustering of facility-recorded malaria and sickle cell disease admissions across health-facility catchments in Uganda: a Bayesian shared-component disease mapping analysis ## Summary This repository implements the analytic workflow for the paper, **“Spatial co-clustering of malaria and sickle cell disease burden across health facility catchments in Uganda: a Bayesian shared-component disease mapping analysis.”** The pipeline preprocesses modeled higher-level health facility catchments and routine DHIS2 admissions data, quantifies global and local spatial autocorrelation, fits Bayesian shared-component disease-mapping models, compares baseline, extended, and covariate-adjusted specifications, and derives hotspot, stability, and posterior-uncertainty summaries. The workflow is designed to examine whether facility-recorded malaria and sickle cell disease (SCD) admissions share residual spatial structure across modeled service geographies in Uganda. It moves from descriptive mapping to global and local spatial dependence, then to joint Bayesian modeling of shared and disease-specific spatial components, and finally to hotspot interpretation and uncertainty classification. Key outputs are cleaned analytic catchment datasets, global and local autocorrelation statistics, Bayesian model summaries, posterior risk surfaces, hotspot typologies, uncertainty maps, and manuscript-ready tables and figures. ## Folder layout . ├─ data/ │ ├─ raw/ # place source inputs here │ ├─ interim/ # intermediate analytic files │ └─ processed/ # final outputs land here ├─ R/ │ ├─ 01_q2_step1_preprocess_catchments_and_outcomes.R │ ├─ 02_q2_step2_spatial_weights_and_global_moran.R │ ├─ 03_q2_step3_bivariate_local_lisa_and_multivariate_geary.R │ ├─ 04_q2_step4_bayesian_shared_component_model.R │ ├─ 05_q2_step5_extended_shared_component_model_comparison.R │ ├─ 06_q2_step6_hotspot_typology_stability_ …