Use Case
Multi-Server Comparison
You govern dozens of connected suppliers. Each delivers different FHIR quality. How do you compare?
For: MII coordinating centers (37 university hospitals), national program architects, payer FHIR leads
The problem
No standardized way to compare FHIR quality across suppliers.
Germany's Medical Informatics Initiative (MII) connects 37 university hospitals through a common FHIR-based Kerndatensatz. Each site implements the same profiles — but no standardized quality signal exists to compare conformance across sites. Without cross-site measurement, data integration centers cannot assess federation readiness, and BMBF-funded research projects cannot guarantee comparable data quality.
US payers face the same challenge: CMS-0057-F and CMS-9115-F require conformant Da Vinci and CARIN APIs across dozens of data-supplier vendors — but there is no objective measurement to enforce those SLAs. When a vendor claims conformance, you have no evidence to confirm or dispute it.
At national scale — gematik (ePA, ISiK), NHS Digital, and the EHDS, where implementing acts require cross-border interoperability by March 2027 — the problem is the same: connected systems without comparable quality metrics.
Without cross-server comparison, quality is anecdotal. Supplier scorecards are impossible. Federation readiness is a guess.
How Records solves it
Side-by-side conformance comparison across all connected servers.
Records registers each supplier or connected system as a distinct server. The same validation configuration — profiles, terminologies, thresholds — runs against every server. Because the reproducibility contract pins validator version, IG packages, and terminology snapshot, the results are directly comparable. This is the measurement basis for supplier scorecards and SLA enforcement.
For the MII: each university hospital is validated against Kerndatensatz profiles across all 37 sites. The Germany Pack — ISiK + MII + KBV + HL7 DE profiles in one configuration — ensures all three parallel German FHIR ecosystems are covered in a single comparison run. For EHDS: cross-border conformance comparison against HL7 Europe IGs on the same measurement scale.
For US payers: each data-supplier vendor is validated against Da Vinci and CARIN profiles on schedule. Quarterly scorecards show which vendors meet CMS-0057-F and CMS-9115-F conformance SLAs — and which don't. Records provides the objective measurement to enforce them.
Cross-server comparison
Same validation config against every server. Results are comparable because measurement parameters are identical.
Per-supplier registration
Each connected system is a distinct server in Records. Independent baselines, independent history, comparable signals.
Trend tracking
Per-server conformance tracked over time. Quarterly comparison reveals whether suppliers are improving or degrading.
Supplier evidence reports
Per-supplier evidence PDFs for SLA enforcement, acceptance decisions, and program governance.
How to integrate
Register each supplier as a server. Run the same validation against all of them.
Validate each supplier
# Register and validate each supplier
records validate \
--server=<supplier-1-fhir> \
--ig=<program-ig> \
--environment=supplier-1
records validate \
--server=<supplier-2-fhir> \
--ig=<program-ig> \
--environment=supplier-2
# Compare results in the Records dashboardWhat gets pinned
| Pin | Purpose |
|---|---|
ig_packages | Same profile set across all suppliers |
terminology_snapshot | Same code systems for all comparisons |
validator_version | Same engine for all suppliers |
environment_label | Distinguishes suppliers in comparison view |
For MII sites: deploy on-premise via Docker. Each university hospital's FHIR façade is registered as a distinct server. Use the Germany Pack to validate ISiK, MII Kerndatensatz, and KBV profiles in one run. For national programs and EHDS: on-premise with HL7 Europe IGs. For payers: SaaS deployment with each data-supplier vendor registered as a separate environment.
What you get
Conformance comparison
Side-by-side per resource type across any number of servers.
Alignment gap identification
See where suppliers diverge from program requirements.
SLA-evidence reports
Per-supplier reports with vendor-neutral, third-party-verifiable measurement.
Per-server drill-down
From server comparison to specific resource types and field-level issues.
Federation readiness
Compare conformance across regions, suppliers, or partner systems with a single measurement framework.
Trend analysis
Per-server conformance tracked over quarters to measure supplier improvement or degradation.
Why vendor self-attestation is not measurement
Claims are not evidence. Measurement requires a standardized, independent ruler.
| Vendor self-attestation | Records comparison | |
|---|---|---|
| Measurement basis | Vendor-defined | Standardized across all suppliers |
| Comparability | Not comparable | Directly comparable (same contract) |
| Evidence | Verbal or vendor-provided | Independent, third-party-verifiable |
| Trend tracking | ne | Per-supplier history over time |
| SLA enforcement | Based on claims | Based on measured conformance |
Frequently asked questions
There is no hard limit. Each server is registered independently with its own baseline and history. National programs with hundreds of connected systems register each as a distinct server.
No. Records is vendor-neutral. It has been tested against HAPI, Firely/Vonk, Blaze, Spark, Medplum, and SMART Health IT. The comparison is at the data level, not the implementation level.
Yes. Records is available as a Docker container. On-premise deployments support air-gapped environments with no outbound internet requirement. Data never leaves the program's network.
Records supports FHIR R4 and R5. If suppliers use different FHIR versions, they are validated against version-appropriate profiles. Cross-version comparison is possible at the conformance signal level.
See this on your own data.
We'll compare validation across your FHIR servers — live.