MNH Analytics Framework
This sub-module covers Objective 3 of the maternal mortality work: turning data that has already been ingested into working analytics and decision-ready dashboards. The subject is maternal mortality analytics — from routine records in DHIS2, case-level surveillance in MPCDSR, and periodic DHS survey estimates, to metrics, charts, and dashboards built in Apache Superset.
The sub-module deliberately starts with theory and ends hands-on, because a dashboard is only as trustworthy as the concepts and data quality behind it.
Structure
The material is organised in two parts:
| Part | Session | What it covers |
|---|---|---|
| 01 — Foundations | Analytics Foundations | What analytics really is, the four types of analytics, the national data sources, data quality assessment, and the analytical questions the dashboards answer |
| 02 — Building | Lab: Building MMR Analytics in Superset | The four-step build recipe — dataset, metric, chart, dashboard — applied end to end to the maternal mortality ratio, and the insights the finished dashboard surfaces |
Part 01 gets the concepts straight before touching the tooling. Part 02 then applies them: every artefact in the lab follows the same repeatable recipe, so the workflow transfers directly to other health programmes beyond maternal health.
What to have ready
Before starting the lab, make sure you have access to an Apache Superset instance that is connected to the data warehouse holding the ingested maternal health data. All dataset registration, metric definition, and dashboard assembly happens inside Superset — nothing is built locally.
Start with Analytics Foundations, then move on to the lab.