tiwa-codes/health-dqa-pipeline

A reproducible Data Quality Assessment (DQA) pipeline for routine health facility reports in Nigeria. This project implements comprehensive data quality checks, scoring, and interactive visualization for monthly facility data including immunization, ANC, deliveries, and OPD services. # Health Data Quality Assessment Pipeline A reproducible Data Quality Assessment (DQA) pipeline for routine health facility reports in Nigeria. This project implements comprehensive data quality checks, scoring, and interactive visualization for monthly facility data including immunization, ANC, deliveries, and OPD services. ## 📋 Project Overview ### Objective Create a complete, automated pipeline to assess and report on the quality of routine health facility data, helping identify issues with completeness, timeliness, consistency, outliers, and data anomalies. ### Key Features - **Comprehensive DQA Checks**: Completeness, duplicates, timeliness, outliers (MAD method), spikes, consistency - **Quality Scoring**: 0-100 scores per facility-month with configurable weights - **Multi-Level Summaries**: Facility, state, and national level aggregations - **Interactive Dashboard**: Streamlit app with filters, charts, and data downloads - **Synthetic Data Generation**: Realistic Nigerian health data with intentional quality issues - **Reproducible**: Seeded random generation, configuration-driven, offline execution ### Data & Indicators **Facility Types**: Primary Health Centers (PHC), General Hospitals **Nigerian States**: 37 states/territories (36 states + FCT Abuja) **Health Indicators**: - `opd_visits`: Outpatient department visits - `anc1_first_visit`: First antenatal care visit - `anc4_visits`: Four or more ANC visits - `facility_deliveries`: Deliveries at facility - `bcg_doses`: BCG vaccine doses - `penta1_doses`, `penta3_doses`: Pentavalent vaccine doses - `measles1_doses`: Measles vaccine doses - `fully_immunized_under1`: Fully immunized children under 1 year - `vaccine_stockout_days`: Days with vaccine stockouts - `eligible_under1_population`: Eligible population denominator ## 🚀 Quick Start ### 1. Setup Environment ```bash # Clone repository git clone github.com cd health-dqa-pipeline # Create virtual environme …