What is referential matching?
Referential matching is an identity matching technology that links patient records and provider records by checking each one against a large, continuously updated database of real identities. That database acts as an answer key, so records are never compared to each other. Referential matching connects records that deterministic and probabilistic matching methods miss, even when the data is old, sparse, or inconsistent.
How does referential matching work?
Referential matching looks up each of your records separately in a reference database of known identities, then links every record that points to the same person.
In the example below, Records A and B come from two different systems, such as an EHR and a CRM, and they share no fields at all. Both still match Rebecca’s one identity in Verato Carbon®, the reference database behind Verato Referential Matching®, which holds every name, phone number, and address she has used.
How referential matching compares to other patient matching methods
Referential matching checks each record against a reference database of known identities, so every system in your organization recognizes the same person. Deterministic and probabilistic matching just compare records to each other, where a single typo or missing field is enough to leave duplicate records behind.
Referential matching vs. deterministic matching
Deterministic matching, also called rules-based matching, is built into most electronic health records and declares a match only when two records’ demographic fields line up exactly. When Rebecca Smith remarries and becomes Rebecca Jones, deterministic matching sees two different patients, while referential matching already knows both names belong to her.
Referential matching vs. probabilistic matching
Probabilistic matching, also called fuzzy matching, scores how likely it is that two records belong to the same person, then routes anything uncertain to a data steward for review. When Rebecca moves across the state, probabilistic matching can only call her old and new records a maybe, while referential matching recognizes both addresses as hers.
How accurate is referential matching?
Accuracy depends on what each record is matched against. Research compiled by the Pew Charitable Trusts found match rates fall to about 80% inside a single organization and as low as 50% between organizations, even when both run the same EHR.1
Referential matching closes much of that gap. Verato Referential Matching® is 24% more accurate than competing MDM and EMPI matching approaches.² See how Verato verifies match accuracy.
Common referential matching use cases across healthcare
Because referential matching doesn’t depend on two records looking alike, it shows up anywhere identity has to hold up across systems that were never designed to talk to each other — inside a single health system, across a payer’s member and provider data, or between organizations that share patients but not software.
Identity verification
Digital front doors, online scheduling, and self-service registration all need to confirm a real person is who they claim to be before a record is created. Verato Verify™ checks identity in real time against the reference database, which cuts down on duplicate creation at the exact moment it usually happens — first contact — without adding friction to the sign-up itself.
Patient scheduling
The same real-time check that stops a duplicate record from being created also keeps a returning patient’s scheduling history, insurance details, and preferences attached to the right person, instead of splitting across a new profile every time she books online rather than by phone.
Patient referral
When a patient is referred between a primary care practice, a specialist, and a health system, referential matching lets every organization involved recognize that they’re looking at the same person, even if each one only has a partial record. That same principle scales up to health information exchanges: an HIE connecting dozens of participating organizations can only be as useful as its ability to tell them they’re all looking at the same patient.
Provider attribution
The same logic applies to clinicians. Referential matching links a provider’s affiliations, credentials, and claims history across every organization they touch — so the Dr. Jane Smith seeing patients at a satellite clinic is correctly tied to the same Dr. Jane Smith with hospital privileges — which keeps directories accurate and referrals routed correctly.
Patient personalization
Once identity is resolved, organizations can safely connect clinical data to consumer and behavioral data, so outreach reflects a patient’s actual situation instead of a fragment of it.
How does referential matching make identity data AI-ready?
Referential matching and AI support each other in three ways:
- AI-ready data. Duplicate and fragmented records become AI training data that teaches a model the wrong thing about who a person is, so referential matching resolves them first.
- AI-powered matching. AI handles nicknames, cultural name variations, and alternate spellings.
- Explainable decisions. A data steward can ask why two records did or did not match and get a plain-language answer.
What AI cannot do is invent the reference data. A model can tell you two records look similar. Only a reference database can confirm that Rebecca Smith and Rebecca Jones are the same person.
Verato’s solution for referential matching
Verato MDM Cloud™ gives you one trusted view of every patient, provider, member, and consumer across all your systems. The platform runs on Verato Referential Matching®, our patented third-generation approach to referential matching, which resolves your records against Verato Carbon®, a reference database of more than 300 million U.S. identities.
Not every duplicate patient or provider record looks like one. Book a strategy session to see what referential matching catches.
¹Bipartisan Policy Center (2012) and ONC/Audacious Inquiry (2014), as reported by The Pew Charitable Trusts, “Enhanced Patient Matching Is Critical to Achieving Full Promise of Digital Health Records” (2018).
2Verato’s Referential Matching® was independently verified by the Regenstrief Institute (2022) to be 24% more accurate than competing MDM/EMPI matching approaches.