One source of truth for your entire organization

Example view - dashboards are customized for your institution

What is an EHR-independent data warehouse?

An EHR-independent data warehouse is a central data layer for your institution where treatment data, quality data and operational data come together, separate from the EHR they originate from. The data is extracted via connectors from the source system, transformed into a uniform model and stored in an environment that you own. The advantage: you no longer depend on what the EHR vendor does or does not let out of the system. If you switch EHR in the future, the data warehouse and the history remain intact.

Which topics does the data warehouse cover?

The core consists of six fact tables that cover the most requested reporting needs: care activities, admissions and treatments, diagnoses and DBCs, financials, quality indicators and staff deployment. Around them sit dimension tables for patients, care professionals, locations, time and care products. The model is a star schema that fits how controllers, quality officers and researchers actually ask questions. We add additional fact tables when a specific question requires it.

How does RSAD work in the data warehouse?

RSAD is the standardized delivery of care activity and DBC data to parties such as the NZa. The data warehouse generates this delivery from the integrated source data, with the same definitions and validations laid down in the RSAD specification. The advantage is that the delivery is based on the same data you use internally for steering and accountability: no more difference between what you submit externally and what you report internally. The query and validation runs automatically, so manual assembly is no longer necessary.

What are the six fact tables?

Care activities (the individual procedures at patient and date level), Admissions and treatments (clinical admissions, outpatient visits, day treatments), Diagnoses and DBCs (registered diagnoses linked to care products), Financials (the financial side of care production), Quality indicators (aggregated outcome data for steering and accountability) and Staff deployment (the deployment of care professionals related to production). Each table is modeled so that cross-sections by patient, time, location and care product are possible quickly.

Where does the data warehouse run?

In your own Google Cloud environment (BigQuery) or Azure environment (Microsoft Fabric or Synapse). You own the project, the billing and the access rights. We handle the configuration according to NEN 7510 and GDPR, with encryption in transport and at rest, audit logging and an access model that aligns with your existing identity management. The data does not leave your own tenant. If you want to move to another cloud later, you can: the data model is portable and the transformations are written in dbt, which runs on both platforms.

How do we stay owners of the data?

The data, the transformations, the connectors and the documentation all belong to your institution. We work in your cloud tenant, in a repository you manage, with code you can read and change. At delivery your IT department receives training and documentation so they can run it themselves. We remain available for operations, further development and the annual updates to the RSAD specification, but that is a service model, not a vendor lock-in.

How long does an implementation take?

A first working version with one or two fact tables and the related dimensions is typically in place within three to four months. The full model with all six fact tables, RSAD delivery and dashboards takes six to nine months, depending on the complexity of the source systems and the available time of your own people for validation. We work in sprints with intermediate deliveries, so you can already report after three months and do not have to wait for the final delivery.