HealthRecordCommunity
FHIR🌏 USA🏛 AWSEnriched

How to Resolve Duplicate Health Records with AWS HealthLake

Resolve duplicate health records with AWS HealthLake

August 20, 2026AWS

Summary

AWS has introduced resource matching for HealthLake, addressing the critical issue of duplicate patient identities derived from multiple data sources. This feature automatically creates FHIR links, enabling the construction of an accurate, unified view of patients and related entities.

Details

Duplicate records pose a significant challenge in healthcare data management, often stemming from misspellings, entry errors, or different presentation locations. This fragmentation complicates clinical care—making it harder to identify drug interactions or care gaps—and increases operational costs through billing errors and manual reconciliation. Historically, building longitudinal patient records from diverse sources (EHRs, HIEs, claims systems) required deploying separate Master Data Management (MDM) or Enterprise Master Patient Index (EMPI) systems. AWS HealthLake's resource matching capability simplifies this process by automatically identifying FHIR resources representing the same real-world entity. Available for seven key FHIR R4 resource types (Patient, Practitioner, Organization, etc.), the feature evaluates data written to the datastore against existing records using high-confidence healthcare identifiers (like SSN or MRN). When a match is found, it creates an immutable 'FHIR Linkage' resource, linking the originals without modification. This allows for the creation of a unified patient view without relying on a separate MDM system. The matching process is deterministic and recognizes multiple identifier formats, including those appearing under different system URIs (e.g., SSN written under http://hl7.org/fhir/sid/us-ssn). The feature runs asynchronously after data ingestion and is currently available in gated preview.

📰
Read Original Article
aws.amazon.com
→

Original content copyright by respective publishers