Managed FHIR servers on the cloud dominate the server narrative.
Azure, AWS and GCP are heavily invested in FHIR servers, and many smaller companies exist to serve the same market.
But is going Cloud the right solution for your product and business?
It’s easy to get started, easy to scale up, and easy to manage. None of which makes it the right decision.
You need to be aware of the limitations and factor them in before deciding.
- No database access
You cannot query the underlying database directly. Managed FHIR servers are black boxes. All data queries flow through the API. - No control over release cycles
You don’t control when a new release is pushed and you don’t know if it’s about to deliver a bug that cripples your app. - Support can take days or weeks
Unlike your own developers who can drop everything in an emergency and fix or roll back an update, with cloud providers you’re one customer of many. - Server locations are limited
The last FHIR server I spun up in Europe had to be hosted in Sweden. All other EU locations were at capacity. Storing patient data outside the patient’s country can be a regulatory problem. - Cloud can be a hard sell
Not all hospitals and healthcare providers are open to having their patient’s data stored in the cloud. Your sales team could find doors slamming in their faces as soon as they mention cloud. - Scaling costs
Getting started is usually cheap and easy. But moving from one server to a 1,000 — not so cheap and not so easy.
The push to Cloud Native is everywhere in recent years. You’d be forgiven for thinking it was an answered question when it comes to healthcare data.
It’s not. The jury is very much still out.
You need to do proper due diligence. Ignore the hype and prevailing narrative. Choose the FHIR server approach that works best for your customers and your business.
---