US Accountable Care Organizations have an MPI problem with a specific shape. The ACO has to identify patients across multiple participating practices, attribute care to the right ACO under CMS attribution rules, and reconcile identity across the multiple EHRs that participating practices use. The choice between a commercial MPI and an open-source MPI for an ACO deployment is sharper than the same choice for a single health system. This comparison walks through how that decision tends to land for US ACOs in 2026.
For the broader buying frame, see the cornerstone Complete US FHIR Master Patient Index Buyer's Guide for 2026. For our FHIR knowledge desk, the rest of the desk follows on.
What Each Side Is Actually Offering
Commercial MPIs for US ACO use come from vendors like NextGate, Verato, Smile Digital Health, and a handful of others. The pitch is a matching engine plus operational layer plus support contract sized for ACO requirements. Open-source MPIs for US ACO use include OpenEMPI, Mirth Match, and patterns built around HAPI patient match extensions. The pitch is full control over matching logic at zero per-record licensing cost.
Both routes can hit the matching accuracy floor a US ACO needs. The shape of the cost and the shape of the operational risk are different.
Where Commercial Wins for US ACOs
Commercial wins for US ACOs that need to attribute care across multiple EHRs without building an identity engineering function. The ACO's IT team is usually small, the participating practices each have their own systems, and the commercial product takes the identity surface off the ACO's roadmap. Vendor-managed matching tuning, vendor-handled audit logging, and vendor support for CMS attribution edge cases all matter.
The other place commercial wins is when ACO performance bonuses depend on attribution accuracy. The vendor's reputation is on the line for matching outcomes, and the contract gives the ACO a target to hold the vendor to.
Where Open Source Wins for US ACOs
Open source wins for US ACOs with sufficient engineering bench, often through a parent health system or a long-term IT vendor. The ACO gets full control over matching logic, can tune for the specific patient populations participating practices serve, and avoids per-record licensing costs that scale with ACO size.
The other place open source wins is for ACOs that want strategic flexibility in how they evolve identity matching over time. The source tree is theirs, and the ACO can adapt the engine as participating practices change.
How Cost Compares for US ACOs
Commercial MPIs carry license fees that typically scale with patient volume. For an ACO with hundreds of thousands of attributed patients, the licensing line is meaningful. Open-source MPIs carry zero license cost and substantial operational cost: identity engineering staff, matching tuning, audit pipeline development.
For most US ACOs, the total cost lands in the same neighborhood once headcount is included. The deciding factor is which line item the organization is better at managing.
Picking for a US ACO
The honest rule of thumb for an ACO is similar to the rule of thumb for any US healthcare identity procurement. If the ACO has, or can lean on, identity engineering capacity, open source pays off over a five-year horizon. If the ACO needs the operational layer handled and wants the vendor accountable for matching outcomes, commercial fits.
The Top 5 open-source MPI solutions for US FHIR stacks covers the open-source side of the comparison in detail. The right MPI for a US ACO deployment is the one that supports clean attribution under CMS rules without becoming the engineering project that pulls the ACO's IT team away from clinical operations.
Sources
- Patient $match operation definition - Spec page, HL7 FHIR build, 2026 v6.0.0-ballot4
- Trusted Exchange Framework and Common Agreement (TEFCA) - Policy page, ONC ASTP, 2026
- The Tide and the Speedboats: TEFCA and CMS-aligned Networks - ONC blog post, 2025
