The Shapiro-Wilk test is exquisitely sensitive to sample size, not normality itself. With large n, it flags trivial deviations. With small n, it misses real departures. This leads researchers to make the wrong inference about whether to use parametric tests.

Analytical question: Does income group moderate resistance rates, and does this moderation vary by pathogen?

Interaction Term Analysis in R

On 15 May 2026, DRC and Uganda declared an Ebola (Bundibugyo virus) outbreak. WHO declared a PHEIC on 16 May 2026 — the 17th DRC Ebola outbreak. As of 24 May, >1,010 cases and 130 deaths were reported across both countries. This document estimates R₀ and Rₜ using reported case data reconstructed from WHO, CDC, and ECDC situation reports

This report models weekly disease case counts in a specific district using Negative Binomial Regression, which is well-suited for over-dispersed count data. Environmental predictors include: Rainfall (mm/week) Temperature (°C — mean weekly) NDVI (Normalised Difference Vegetation Index — proxy for vegetation density and humidity)

This report presents the sub-county level distribution of Neglected Tropical Diseases (NTDs) across Kenya’s 47 counties and 290 sub-counties. Kenya bears a significant NTD burden, with diseases such as Schistosomiasis, Lymphatic Filariasis (LF), Soil-Transmitted Helminths (STH), Visceral Leishmaniasis (Kala-azar), Trachoma, and Onchocerciasis disproportionately affecting populations in coastal, western, and arid/semi-arid regions.

The MV Hondius, a Dutch expedition cruise ship, experienced an outbreak of Andes hantavirus (ANDV) beginning April 2026. The index cases — a Dutch couple — are believed to have been exposed during a birdwatching trip across Chile, Argentina, and Uruguay prior to boarding on 20 March 2026. As of 7 May 2026, WHO reports 8 identified cases (5 confirmed, 3 suspected) and 3 deaths, with the ship en route to Tenerife, Spain. Key epidemiological question: What does the incubation period distribution tell us about exposure timing and potential for further spread?

Generalized Additive Models (GAMs) extend linear models by allowing non-linear smooth functions of predictors while maintaining interpretability

This report demonstrates logistic regression using a diabetes dataset. The outcome is diabetes diagnosis (Yes/No), predicted by age, BMI, glucose level, physical activity, and family history.

This report analyses a simulated disease outbreak affecting 30 individuals across 4 communities. Network methods identify high-risk individuals, transmission pathways, and key structural features that drive epidemic spread.

Geospatial Mapping, Prediction & Interpretation Using Shapefiles

A Comprehensive Burden of Disease Study

This report presents a comprehensive geospatial analysis of flood risk in Nairobi County, Kenya. Using proximity-weighted risk modelling, multi-factor susceptibility analysis, and seasonal prediction techniques, the study identifies and quantifies flood-prone areas across the city. The findings confirm that informal settlements along river corridors — particularly Mathare, Kibera, and Mukuru — face the highest and most persistent flood risk, driven by a combination of hydrological, topographic, infrastructural, and socio-economic factors.