Entity Resolution: Definition and Healthcare Context
Full name: Provider Entity Resolution
Entity resolution is the process of determining when records across different datasets refer to the same real-world entity — a single provider, facility, or organization. In health care data, the same physician can appear under different name spellings, addresses, and identifiers across NPPES, PECOS, Open Payments, and exclusion lists. Entity resolution links those records to one resolved identity, usually anchored on a stable key such as the NPI for individuals or the CMS Certification Number for facilities, so a fact in one source can be joined to a fact in another.
How it’s used
- CMS NPPES NPI Registry: the NPI is the deterministic anchor key that joins individual-provider records across federal datasets.
- CMS Provider of Services (POS) File: the CCN is the facility identity backbone that links Care Compare, HCRIS, and ownership data.
- CMS Open Payments: physician payment records are resolved to NPPES identities so payments can be read against a provider's full profile.
- Fonteum's resolution layer produces one record per provider, with each linked field carrying provenance back to the source that asserted it.
Frequently asked questions
- What is entity resolution?
- Entity resolution determines when two or more records — often from different datasets — describe the same real-world person, facility, or organization, then links them into a single resolved identity.
- Why is entity resolution hard in healthcare data?
- The same provider appears with different name spellings, addresses, and identifiers across NPPES, PECOS, Open Payments, and exclusion lists, so matching requires more than an exact string comparison.
- What key does Fonteum resolve provider records on?
- Individual providers are anchored on the 10-digit NPI; certified facilities are anchored on the 6-digit CMS Certification Number, which serves as the join key across CMS facility datasets.
Related terms
Explore in Fonteum
How Fonteum sources, resolves, and publishes data tied to this term.