Integrated care networks run on identity reconciliation. A patient seen by primary care, community nursing, mental health services, and an acute trust within the same Integrated Care System has to be recognised as the same patient across every system, or the joined-up care promise fails before it gets a chance to start. The five MPI tools below are the ones that hold up against the realistic identity workload of a UK Integrated Care System in 2026.
For broader context on master patient index selection and where Integrated Care System deployments sit within it, the FHIR primer hub is the right entry point.
What Integrated Care Networks Ask of an MPI
Three demands matter most:
- Reliable matching across the mixed quality of demographic data that comes out of primary care, secondary care, and community providers in a single ICS footprint.
- Configurable stewardship workflows that can route uncertain matches to the right operational team rather than dumping every edge case on a single queue.
- A standard FHIR API surface, because every modern care coordination application expects $match and Patient.link to behave as the specification describes.
A tool that hits all three becomes part of the shared infrastructure. A tool that misses any of them creates a parallel reconciliation process that drifts over time.
The 5 MPI Tools for Integrated Care Networks Worth Shortlisting
- HAPI FHIR EMPI. HAPI FHIR's EMPI module is the open-source baseline. The matching configuration is flexible, the integration with the wider HAPI stack reduces friction for teams already running HAPI, and the absence of recurring licence cost is attractive for budget-pressured ICS footprints. The honest cost is the operational ownership required to keep matching thresholds tuned.
- Smile CDR MDM. Smile CDR's commercial MDM module builds on the HAPI EMPI foundation with managed operations and a stewardship interface that ICS-scale teams find easier to staff. Useful when the ICS prefers a commercial vendor on the hook for matching accuracy and stewardship operations.
- NextGate MatchMetrix. NextGate is one of the longest-standing dedicated MPI vendors and has a deep track record in healthcare. The matching engine is mature, the stewardship tooling is well established, and the integration with FHIR-native systems has improved significantly over the past few years.
- Verato. Verato's referential matching approach (using a maintained reference dataset to support identity decisions) suits ICS footprints where demographic data quality varies sharply across providers. The trade-off is the operating model: referential matching is a managed service with its own data residency and privacy considerations that need to be worked through.
- OpenEMPI. OpenEMPI is the long-standing open-source MPI option, with a deployment footprint that suits smaller ICS pilots and community-led deployments. The community is smaller than HAPI's, but the core matching behaviour is sound and the software has matured into a stable platform.
How to Pilot These Across an ICS
The most useful pilot is a real one across the messiest data the ICS holds. Take a sample of records from each provider in the footprint, including the records the providers themselves know are imperfect, and run them through each candidate. Pay attention to four behaviours: the match rate against records that should match, the false-positive rate against records that should not match, the stewardship queue volume produced, and the API performance under realistic concurrency.
The candidate that wins on the first two is the one to shortlist. The candidate that loses on stewardship volume tends to be expensive in operations even if it looks fine on a feature matrix.
For broader strategic context on master patient index selection, the practical guide to master patient index for FHIR in 2026 is the right back-reference. For an adjacent shortlist tuned to FHIR-native care coordination workflows that often pair with an ICS MPI deployment, the top 7 MPI products for FHIR-native care coordination in 2026 is the natural companion read.
Sources
- MDM module overview - HTML docs, HAPI FHIR
- MDM module documentation - HTML docs, Smile CDR
- Patient Deduplication Architectures - HTML docs, Medplum
