Refugee and migrant health clinics see patient-matching challenges that stretch the assumptions baked into most MPI products. Names may be transliterated differently across encounters, birthdates may be approximate, prior records may exist in another country, and the patient may pass through multiple resettlement programs before stabilizing in a primary-care home. The five patient-matching tools below have shown up in real refugee and migrant health deployments in 2026. For related FHIR tooling reviews, the broader reference covers the related comparisons.
The Five Tools Running in Refugee and Migrant Clinic Production
- NextGate Patient Identity. Adopted by several urban refugee health programs because of strong probabilistic matching tuned to handle transliteration variants and approximate birthdates.
- Verato Universal MDM. Referential matching reduces duplicate creation when a refugee patient appears under variant name spellings across multiple touchpoints in the US health system.
- JEMPI. Open-source. Picked by smaller refugee health programs and by FQHCs serving migrant populations because of the budget fit. Name-variant configuration is flexible enough to handle the linguistic diversity these clinics see.
- Rhapsody EMPI. Used by refugee health programs already running Rhapsody for HL7 v2 integration with the affiliated hospital network. Hybrid matching handles the mix of known and minimally-known records.
- OpenEMPI. Long-running open-source MPI. Adopted by some refugee clinics that need MPI capability with no commercial license; requires more local engineering capacity.
The five cover the realistic FHIR-aligned options for refugee and migrant health clinics in 2026.
Three Capability Tests Worth Running
A refugee health clinic evaluating a patient-matching tool should put three scenarios through a pilot. Transliteration tolerance: the same patient's name may arrive as Mohammad, Mohammed, Muhammad, or Mahmoud, and the MPI should handle these as candidate variants of the same record rather than four distinct entries. Approximate-date matching: a patient who reports "around 1985" for a birthdate should still match a record with a specific 1985-04 entry. Cross-program linkage: a patient seen at a resettlement agency clinic should link to subsequent records at the FQHC primary-care home and at any specialty consultations within the network.
A three-month pilot against the clinic's real patient mix surfaces most operational issues. The long-term care cornerstone covers the broader selection framework, and the school-based health walkthrough covers identity resolution in a related minimally-known-record context.
Why Refugee and Migrant Health Fit Differs
Refugee and migrant health clinics work with patient records that violate many of the assumptions general MPI products are tuned for. Names are linguistically diverse, dates are approximate, prior care happened outside the US health system, and identity attributes can shift legitimately across encounters. The MPI has to handle this without either creating duplicates or merging incorrectly. All five tools above clear that bar in 2026 refugee and migrant clinic deployments, which is the practical reason they made the list ahead of other MPI products that work well in mainstream US settings but stumble on the additional variability.
Refugee and migrant clinics that pilot an MPI against their actual patient mix see the operational realities surface quickly. The transliteration tolerance, the approximate-date matching, and the cross-program linkage together separate the MPIs ready for refugee health from those that handle mainstream US identity well and stumble on the additional variability.
Beyond the five tools above, refugee-resettlement-agency software vendors are starting to embed MPI capabilities directly into the resettlement case-management platform. That tighter integration reduces the duplicate-record creation that happens at the agency-to-clinic handoff and is worth evaluating for new refugee-health-program deployments.
Refugee health networks that share notes with other resettlement-focused programs often discover their peers run one of these five tools. The practical reason is real-world performance against the refugee patient mix, and the five tools above clear that bar consistently in 2026 deployments.
Sources
- Optimizing Patient Record Linkage in a Master Patient Index Using Machine Learning - PMC
- Patient Matching section - HL7 FHIR Identity Matching IG
- HL7 FHIR Patient/$match operation specification
