Developing a Predictive Model for Waterborne Disease Outbreak Risk in Urban Slums in Lagos Metropolis Using Environmental and Social Determinants

Waterborne disease outbreaks in urban slum communities are driven by the complex interplay of environmental infrastructure deficits, social vulnerability, microbial ecology, and behavioral factors, yet predictive models integrating these determinants for Nigerian urban contexts are non-existent. This dissertation develops and validates a predictive model for waterborne disease outbreak risk in urban slum communities across Lagos Metropolis, South West Nigeria. A mixed-methods design combined prospective environmental surveillance, social vulnerability assessment, and epidemiological outbreak data analysis. Environmental predictors, including fecal coliform contamination levels in shared water sources, latrine-to-household density ratios, flooding frequency, and rainfall intensity data, were collected over twenty-four months across forty slum communities. Social determinants including poverty index, housing density, community network strength, and health literacy scores were quantified through community surveys administered to 1,800 households. Outbreak records from the Lagos State Health Service Commission for 2019 to 2024 were used as the outcome dataset. A hierarchical logistic regression model incorporating environmental and social predictor domains was developed and validated using an independent community dataset. The model demonstrated a sensitivity of seventy-eight percent and specificity of seventy-three percent in predicting outbreak-prone communities above a defined risk threshold. The dissertation makes a methodological and applied contribution by providing the first validated predictive outbreak risk tool for Nigerian urban slum contexts. Recommendations include integration of the model into the Lagos State Integrated Disease Surveillance and Response system and trigger-based rapid response protocols. Keywords: waterborne disease, outbreak prediction, urban slums, Lagos, predictive model.