State immunization information systems run a coding workload that depends on CVX, MVX, NDC, and the CDC vaccine-administered code sets, all reconciled against a state-specific code map. A FHIR terminology server feeding immunization registry mapping has to host these vocabularies, translate cleanly between them, and accept the regular CDC update cadence without operator pain. The six servers below have shipped against real state IIS workloads in 2026. For the FHIR implementation reference, the broader reference walks through the surrounding pieces.
The Six Servers Running in Immunization Mapping Production
- Smile Digital Health Tx. Commercial. Several state IIS deployments use the bundled terminology server because CVX, MVX, and NDC loaders ship with the CDR. The update cadence is automated against CDC releases.
- HAPI Terminology. Open-source. Public-health agencies adopting HAPI for the registry's FHIR API use the bundled terminology layer for CVX-to-MVX-to-NDC translations.
- Ontoserver. CSIRO server. Has gained adoption in state IIS programs because of its strong `$translate` performance on the CVX-to-RxNorm linkage that drives EHR-side immunization records.
- InterSystems IRIS Terminology. Enterprise stack. State IIS programs running IRIS for the broader registry data layer find the terminology piece integrates cleanly with the rest.
- Tx-Server. The reference server. Used as a comparison baseline; some state IIS programs run it in parallel with their production server for behavior-question arbitration.
- Aidbox Terminology. Hosted FHIR platform. CVX, MVX, and NDC load together with translation linkages preconfigured, which suits smaller state IIS programs that need a low-operator-effort setup.
The six represent the realistic FHIR-native options for state immunization registries in 2026.
Three Capability Tests Worth Running
A state IIS team evaluating a terminology server should run three capability tests during pilot. CVX-to-NDC translation: the registry has to round-trip a CVX vaccine code to one or more NDC manufacturer-specific codes, since EHR submissions arrive in either form. CDC update absorption: the server has to accept the CDC's CVX and MVX updates on the published cadence without breaking existing mappings. ValueSet expansion for adolescent-vs-adult vaccine schedules: the registry has to differentiate which vaccines are scheduled for which age groups, and the value sets supporting that have to expand cleanly.
A two-month pilot against real EHR-submitted immunization records surfaces most operational issues. The oncology cornerstone covers the broader selection framework, and the tribal health coding walkthrough shows the same logic applied to a different public-health-aligned setting.
Why Public-Health Fit Matters
State IIS programs operate under federal and state reporting deadlines, and a terminology server that lags on update absorption creates real submission friction. The six servers above all clear the bar of supporting CDC's update cadence without operator drama in 2026 deployments. That practical reliability is the main reason they made the list ahead of other technically capable but public-health-untested servers.
State IIS programs that pilot two servers against the same EHR-submitted immunization records will surface the operational realities quickly. The CDC update cadence, the cross-vocabulary translation, and the age-group value-set handling together separate the servers ready for state IIS workloads from those that handle the demos and stumble on the data.
Beyond the six servers above, the CDC's own immunization reference services have started exposing FHIR endpoints, which lets state IIS programs federate against the CDC layer for the most current code data. That federation pattern is worth designing for, regardless of which server hosts the state IIS itself.
State IIS leadership teams that talk to other states often discover their peers run one of these six servers. The practical reason is reliability under the CDC update cadence; the rest is engineering preference, and any of these will support a defensible production deployment in 2026.
