Orchestrate Identity API

Orchestrate Identity API, developed by CareEvolution, is a privacy-preserving record linking (PPRL) service designed to securely match patient data across disparate healthcare data sources. It is intended for healthcare organizations and developers who need to link patient records without exposing or centralizing sensitive demographic information in a recoverable form.

Traditional healthcare interoperability implementations rely on master patient indexes with centralized demographic data to perform record linking. Identity API improves on this approach by applying irreversible hashing to demographic data before storage, so that all matching operations use hashed values rather than plaintext. This means patient demographic data stored in a centralized index cannot be recovered, reducing the risk associated with centralized data exposure. The algorithms underlying Identity have been developed and refined by CareEvolution over more than 15 years and have been applied across 200 million or more lives with zero reported false positive errors.

Core Capabilities

• Irreversible hashing: Before a record is stored, Identity performs a one-way hash operation on demographic data, converting plaintext strings into long sequences of semantically meaningless characters. All subsequent matching processes use only these hashed values, keeping the original data "blinded."
• Multiple linking strategies: Identity determines a final link status by applying multiple record linking strategies, including both deterministic and probabilistic approaches.
• Record linking across demographic fields: The service performs matching across all demographic fields to determine which records belong to the same individual, and minimizes false positives by linking records only when there is high confidence of a match.
• Demographic data standardization: Demographic information is cleaned and standardized before comparison, so that matching operations produce meaningful results.
• Automated with manual review option: PPRL techniques handle the vast majority of linking automatically, while still allowing for manual review where appropriate.

Technical and Architectural Characteristics

• Cloud native architecture: The Identity API is built on modern, cloud native infrastructure.
• Proven algorithms: The underlying matching logic is based on a mature set of algorithms developed over 15 or more years by CareEvolution.

How the Matching Process Works

1. Demographic data from a patient record is subjected to an irreversible hash operation before being stored.
2. The service stores and operates only on hashed values, not the original plaintext demographic data.
3. Identity applies multiple linking strategies across all demographic fields to assess whether records belong to the same individual.
4. A final link status is determined, with links only confirmed when confidence of a match is high, minimizing false positives.

Orchestrate API Suite

• Terminology API: Standardizes and classifies healthcare codes and concepts.
• Convert API: Enriches and transforms patient data to support workflow optimization.
• Insight API: Computes hierarchical condition category (HCC) risk adjustment profiles.
• Identity API: Privacy-preserving record linking across data sources and use cases.
• Identity Lite: A simplified option for securely matching patient records in two steps.

Identity API is part of CareEvolution's broader Orchestrate technology suite, which provides APIs for liberating, aggregating, standardizing, enriching, and transforming data to support patient-centric use cases. Documentation and demo requests are available directly from CareEvolution.