Orchestrate

Orchestrate is a health data API platform from CareEvolution designed to ingest, aggregate, standardize, and transform disparate real-world health data into analytically-ready datasets. It is used by healthcare stakeholders ranging from national-scale organizations to startups, supporting data use cases across more than 200 million lives. The platform is aimed at developers building healthcare applications — from consumer apps to population health analytics — as well as organizations seeking to enable interoperability solutions and digital clinical trials.

Orchestrate operates through a structured pipeline that takes raw data from varied source systems and formats and progressively refines it into a consistent, usable form. It supports a cloud-native, full-stack platform deployment as well as a self-service API model for organizations that prefer to build their own solutions.

Data Processing Pipeline

1. Liberate: Collects disparate, non-standard real-world data from varied source systems including payer, provider, consumer, and social determinants of health (SDOH) sources, across formats and transports such as HL7, CCDA, FHIR, CCLF, and EHR systems.
2. Aggregate: Aggregates data according to unique individuals, providers, and clinical concepts. Uses identity privacy-preserving record linking to consistently link records belonging to the same individual across different sources.
3. Organize: Stores data in a consistent patient-centric model regardless of source or data type, while maintaining source data provenance in its original form for traceability.
4. Standardize: Automatically harmonizes and standardizes data, mapping uncoded or dirty concepts to standard code systems such as ICD-10, RxNORM, and SNOMED. Also maps concepts across multiple terminology standards, for example ICD-10 to SNOMED and NDC to RxNORM.
5. Group: Groups similar codes together — including therapeutic classes, drug ingredients, HCC categories, and AHRQ value sets — to make data analytically ready.
6. Enrich and Transform: Stacks, normalizes, and transforms unstructured data into APIs and data pipelines compatible with tools and platforms such as Tableau, SQL, Jupyter, Amazon Redshift, Snowflake, FHIR, and OMOP.

Core Capabilities

• No Data Left Behind: Raw data from various sources is loaded and aggregated according to unique individuals, providers, and clinical concepts, with the goal of capturing all available data regardless of origin.
• Lifetime Person Record: Data is cleaned and standardized automatically so that patient information is consistently tracked across data sources, with similar codes grouped together for analytical readiness.
• Flexible Architecture: Organizations can build their own solutions using self-service APIs or use the full-stack, cloud-native Orchestrate platform.

Developer APIs

• Terminology: Provides access to standard code systems and value sets to map and classify inputs to common references.
• Convert: Transforms input data from one standardized format to another, including parsing output, organizing it into appropriate data types, standardizing source codings, and outputting in another format.
• Insight: Computes advanced derived data elements — such as condition profiles, risk strata, and gaps in care — from FHIR bundles processed via the Convert API.
• Identity: Integrates privacy-preserving (blinded) patient matching across a spectrum of data sources and use cases.
• Identity Lite: Provides a simplified patient record matching process in two steps.

Orchestrate supports output to a range of downstream platforms including Snowflake, Amazon Redshift, Tableau, Jupyter, FHIR, and OMOP, making it suitable for integration into existing analytics and interoperability workflows.