About the Core 5.1 Data Model
The Data Model Interface Guide describes the name, purpose, recommended use, and schema of data tables that may be present in data marts that are available when you make a connection to the Bulk Data API.
Depending on the intended use case, a specific data mart may not include all tables that are available in the model. Additionally, a specific data mart may have additional tables that are not described here. The data model behind the Bulk Data API is highly extensible and many data marts will have supplemental tables and columns. The core tables that form the foundation of the API are documented in this guide.
Tables in the core data model can be grouped into categories:
Contents
HIEBus Model
These tables reflect the comprehensive data model used by the HIEBus™ data platform to store transactional health data. These tables are organized in a highly normalized and highly expressive structure that captures an enormous portfolio of health data concepts and relationships. As a result, many common queries will require multiple unions and joins. Most analytics use cases will rely on the tables of enriched concepts represented in the simplified models described below. The HIEBus model is provided to ensure that all data concepts that have been integrated into HIEBus can be made accessible via the Bulk Data API.
Topics
Data Tables
Unified Model
These tables represent a transformation of the HIEBus model that unifies claim-sourced and EHR-sourced data and denormalizes the HIEBus transactional model to make data more readily accessible for analytics use cases via simpler queries. All data tables reference a master HIEBus personid (recordgroup) that links records across data sources into a unified person. Terminology concepts are represented with mapped primary terms in reference terminologies. These table concepts roughly align with FHIR resources that are available via the FHIR API.
Topics
Data Tables
Annotations are computed and provided as columns in order to simplify access to commonly used value set references (i.e. the HCC and CCSR condition category for diagnoses, RxNorm drug class for medications, etc.) and key derived concepts (i.e. normalized care setting computed for encounters, claims, conditions, procedures, etc.).
Population Health Model
These tables represent complex transformations of the HIEBus and Unified models into enriched, higher-order concepts that are useful for population health-related analyses. These transformations include concepts such as condition profiles, medication therapies, certified measures, and risk factors. Several complex algorithm-based computation subsystems are implemented within the high performance data pipeline to perform these transformations, including:
Topics
-
Condition Profiles and Risk Score computes a real-time implementation of the CMS HCC RAPS Score and HCUP CCSR. Condition profiles based on these algorithms can be used to create disease registries, sub-populations with multiple chronic conditions, utilization breakdowns by chronic conditions, and to identify persons with chronic conditions that have not been documented recently.
-
Galileo Encounter Reconciliation computes primarycaresetting columns for claim-sourced and EHR-sourced data, annotates tables in the Unified Model, and then creates a comprehensive, reconciled set of derived encounters to represent hospitalizations and ED visits. Derived encounters readily enable accurate utilization trends and utilization-based risk profiles using all available data sources.
-
Galileo Risk Stratification Framework (GRSF) computes historical risk factors in a vectorized model that can be used to relate current risk profiles to prospective utilization that is targeted by care management programs. GRSF performs whole-population risk stratification and assigns risk levels based on real-time risk factors detected from claim-sourced and EHR-sourced data for a population.
-
Galileo Measures Engine computes certified measures from claim-sourced and EHR-sourced data in order to report on up-to-date population measure performance and to identify real-time care gap opportunities.
Data Tables
RKStudio Model
These tables represent patient-reported survey data and personal device data that is captured via mobile devices using CareEvolution’s MyDataHelps™ application and the RKStudio™ platform, including concepts as: survey results, participant activity, and participant recruitment.
Topics
Data Tables
Terminology Model
These tables provide access to terminology concepts, mappings, and value sets used across the data model. In addition, these tables can be used to explore mapping effectiveness of the Rosetta™ terminology platform and identify mapping opportunities.
Data Tables
Platform Administration Model
These tables provide access to audit concepts and high level summaries of the data that is available from all integrated data sources.
Data Tables
