Getting Started

CareEvolution manages an exceptionally comprehensive array of health data resources. These data assets are reconciled and enriched with a high performance data enrichment pipeline to provide applications with the analytics-ready information they require for insightful analysis.

The Bulk Data API provides access to these data assets using one of the most widely adopted and accessible big data technologies available: Amazon’s Redshift Cloud Data Platform. Redshift is optimized for storing and providing high performance access to large data sets. Any application or framework that can connect to Redshift (or generic PostgreSQL) can connect to CareEvolution’s data assets via our Bulk Data API.

What Data Concepts are Available?

The CareEvolution Bulk Data API provides access to several broad categories of data:

• Transactional data from the HIEBus™ integration platform including EHR, financial, and administrative concepts such as: encounters, claims, referrals, diagnoses, procedure, services, medications, lab results, care tasks, and provider relationships.

• Patient-reported and clinical study data from our RKStudio™ platform and MyDataHelps™ mobile application such as: survey results and connected device data.

• Data transformations computed and updated daily including:

  • Reconciled encounters from claim and EHR data, annotated with population health categories based on coded diagnoses and related utilization from any data source. Annotations include: ambulatory care sensitive condition (ACSC), behavioral health, elective vs. emergent, and 30 day readmission.
  • Utilization aggregates (i.e. hospitalizations, ED visits, cost per member per month, etc.).
  • Risk scores and disease rollups using the CMS-HCC model and HCUP CCSR.
  • Medication therapy rollups based on RxNorm classification.
  • Critical event models to explore pre- and post- utilization and clinical markers based on a reference event date such as hospitalization.
  • NCQA certified clinical quality measures.
  • Derived provider attribution based on a hierarchical provider master index and a prioritized algorithm using explicit assignments and recent utilization.
  • Customized transformations purpose-built for specific use cases.

See the data model documentation for detailed specifications and usage guidelines for the data concepts that are available.

How Does the Bulk Data API Compare to FHIR?

CareEvolution also provides a FHIR API (Fast Healthcare Interoperability Resources, based on the HL7 standard). This API is ideal for applications that can utilize the FHIR standard and acquire data for one patient in each query. FHIR can be used to obtain data for large sets of patients; however, it is not optimized for large data sets, the standard does not express many higher-order transformations, and it will require an intermediate integration step to connect with many analytics or data science applications.

The Bulk Data API is preferable to the FHIR API in use cases with one or more of these requirements:

• Bulk access to large data sets is required for analysis, query, or model-building.
• Direct connection to third-party tools (i.e. Tableau, Power BI, and most data science frameworks) that natively support Redshift but not FHIR.
• Access to concepts that are not widely supported by FHIR.

How Do I Get Started?

1. Choose an application or framework that can connect to Amazon Redshift. We have several examples, available in the use cases.

2. Contact CareEvolution to obtain the API endpoint and credentials that you will use to connect and authenticate.

3. Use our data model documentation to understand the concepts available and how to apply them in your first analysis.