Gitau20/Data-Analysis-Ministry-Of-Health

3 Projects: Clinical data analysis of infant mortality linked to gum scrubbing practices in Machakos: A data-driven mapping of mental health stakeholders, services, and system gaps, highlighting funding, capacity, and policy: An analysis of depression prevalence and key psychosocial and clinical correlates among people living with HIV # PUBLIC HEALTH DATA ANALYSIS - MINISTRY OF HEALTH KENYA Clinical and system-level data analysis to support evidence-based decision making in Kenya's health sector. Three applied projects using real hospital and survey data from Machakos Level 5 Hospital and the national mental health landscape - covering infant mortality, mental health systems mapping, and HIV-associated depression. ## Skills Demonstrated 'Python' 'pandas' 'statsmodels' 'matplotlib' 'seaborn' 'data cleaning' 'exploratory data analysis' 'health systems mapping''clinical data analysis' 'public health policy' ## Project 1: Uncovering preventable clinical risk pathways in infant mortality (2020-2025) {View Notebook - github.com **Context:** Paediatric ward clinical data, Machakos Level 5 Hospital **Tools:** Python, pandas, matplotlib, seaborn **Key Findings:** - Infant mortality **tripled** over the six-year period - Over **90% of gum scrubbing-related deaths** involved septic complications - Critical risk window identified: infants aged **7–12 months** - Seasonal mortality peak identified in **July** - Significant data quality gaps uncovered: underreported discharges, missing geographic data **Recommendations:** - Early sepsis intervention protocols - Targeted caregiver education on traditional practices - Improved clinical documentation standards - Stronger data systems for public health planning > *Data note: All patient data used in this analysis was fully anonymised in compliance with hospital data governance protocols.* ## Project 2: Mapping Kenya's Mental Health System — Structures, Gaps, and Opportunities {View Notebook - github.com **Context:** National mental health stakeholder and service survey **Tools:** Python, pandas, matplotlib, seaborn, E …