RitaAkumu/Maternal-Healthcare-Barriers

Analyzing Kenya DHS data to explore why some women deliver at home versus health facilities. Examines wealth, education, residence, and antenatal care to highlight maternal healthcare gaps. Status: In Progress # Barriers to Skilled Maternal Healthcare in Kenya (DHS Analysis) **Project Type:** Public Health / Data Analytics **Status:** In Progress ✅ ## 📌 Overview This project analyzes **Kenya Demographic and Health Survey (DHS)** data to explore why some women deliver at home instead of health facilities. The goal is to identify key barriers to **skilled birth attendance** and provide evidence-based recommendations that can support maternal health interventions. --- ## 🎯 Objectives - Understand factors influencing **facility delivery vs. home delivery** - Examine how **wealth, education, place of residence, antenatal care visits**, and other variables affect skilled maternal healthcare access - Produce insights that can guide maternal health policies and resource allocation --- ## ❓ Key Questions - Do women in urban areas access skilled delivery care more than rural women? - How does wealth index affect likelihood of facility delivery? - Does education level increase probability of skilled birth attendance? - Do antenatal care (ANC) visits increase facility delivery uptake? --- ## 📊 Dataset **Source:** DHS (Kenya) **Type:** Survey dataset (women’s health, maternal services, household factors) **Example variables:** - Wealth index - Education level - Residence (Urban/Rural) - Antenatal care visits - Age, parity, marital status - Place of delivery - Skilled birth attendance indicators ## 🛠 Tools Used - Python (Pandas, NumPy) - Matplotlib / Seaborn - Statistical analysis (Chi-square / Logistic Regression - optional) - Jupyter Notebook --- ## 🔍 Methodology 1. Data cleaning and preprocessing - missing values, recoding categories, selecting relevant variables 2. Exploratory Data Analysis (EDA) - frequency tables, group comparisons, cross-tabulation 3. Visualization - charts showing delivery patterns across groups 4. Modeling (optional) - logistic regression to estimate likelihood of facility delivery 5. Conclusions and recommendations --- ## ✅ Expected Deliverables …