# Every hospital, one set of numbers.

> Patient, clinical and operational data from every site, unified into one trusted model with access controls built for compliance.

Canonical: https://datakeelai.com/case-studies/hospital-network-governed-lakehouse/
Updated: 2026-11-03

Industry: Healthcare · Service: Data platform
Framing: Our team's experience

How a large US hospital network brought Epic, operational and demographic data into one governed lakehouse, so leadership could finally compare sites like for like.

## Before

Patient, clinical and operational data lived in separate systems. Each team reported from its own copy, so the numbers rarely matched. Proving who could see which data, for compliance audits, was close to impossible.

## What we built

A governed lakehouse that brings Epic EHR data, operational files and demographic data into one place. Data moves through three clear layers: raw, cleaned, and ready for reporting. Everything lands in one standard data model, with access controls designed around healthcare regulations from day one.

## What we found

The same metric was being calculated in different ways by different teams. Agreeing one definition, in one model, ended the "whose number is right?" debate.

## After

- Leadership compares departments and hospital sites using the same, standard metrics.
- Cross-functional teams work from one source of truth instead of their own copies.
- Access is defined by role and can be shown to auditors.

## Stack

- Databricks
- Azure
- Delta Lake
- Power BI
