The A-to-Z Guide to Healthcare Data Models

by | Nov 28, 2022 | Healthcare

A healthcare professional reviewing healthcare data models.

In January 2009, the U.S. Congress passed the Health Information Technology for Economic and Clinical Health Act (HITECH), which removed legal barriers to data interoperability in healthcare systems and financially incentivized the adoption of electronic health records (EHRs). Polls conducted in the same year showed that fewer than half of U.S. physicians – 48.3% – were using EHRs. By 2021, the adoption rate for EHRs had risen to 88% nationwide. While the digitalization of health data has significantly improved patient outcomes and streamlined operations, the rapid growth of health data – in both volume and type – has pushed organizations to adopt newer technologies for data management and to implement healthcare data models. 

In this guide, you’ll learn what data models are and how they structure healthcare data management systems. 

Key Takeaways:
  • The healthcare industry is moving towards the adoption of contemporary data management platforms employing data models.
  • Data models structure raw data and define the relationships between data elements to improve visibility and utility.
  • In recent years, more developers of healthcare data management tools have begun providing products and services featuring data models. 

What Is a Data Model?

A data model is an abstract structure that organizes and standardizes data objects and elements and defines their properties and relationships. Data elements may be abstractions, quantities, or representations of real-world objects such as a physical product’s type, size, or color. Data models specify what data information systems capture and how they attribute data elements to different objects. 

Depending on the context, the term “data model” can refer to two similar but distinct concepts. For example, in object-oriented (OO) programming, data models refer to the abstract representation of objects and their attributes in statements such as:

class Customer < ApplicationRecord
  has_many :purchases
class Purchase < ApplicationRecord
  belongs_to :customer


In OO programming languages, this kind of statement establishes a data model for the objects “customers” and “purchases,” in which customers will have tables for multiple objects called “purchases” and purchases will belong to a class of objects called “customers.” In social media applications, graphed data models look like the figure below. 


Example of a graph data model.
Image Source:

Alternatively, a data model may refer to the reverse-engineered properties and definitions abstracted from an existing relational database. In this context, data models help database programmers understand and modify existing databases. 

Data Models in Healthcare Applications

Although the industry has some unique obligations to data privacy regulations, data models in healthcare data management systems serve the same purposes as they do in other kinds of enterprise data management. They give structure to data captured in different component IT systems and define how individual data elements fit into the larger system. Thus, in healthcare applications, a data model refers to a structure for storing critical information about patient health, health systems operations, patient billing, and planning. 


Medical data collection sources in healthcare sectors
Image Source:

Healthcare data comes from many sources. These include:

  • Operations
  • EHRs
  • Public records
  • Provider data 
  • Lab and medical imaging data
  • Patient portals
  • Physicians and other care providers entering patient data
  • Government agencies
  • Payer records

In the IT systems that record these data types, data elements must have classifications and defined relationships. Data models provide this structure and determine how system objects such as patients, providers, care, locations, staff, and materials associate with this system. 

How Data Models Affect Patient Care and Operational Costs

To understand how data models ultimately affect patient care and operational costs, it helps to look at a specific example. When care providers write prescriptions, they must specify several critical data elements. 

  • Form: Pills, capsules, injections, and serums
  • Dose: Quantity and frequency
  • Route: Orally or by injection – plus location for injections
  • Duration: Ongoing, fixed duration, and with or without refills

In systems using data models, a prescription must match known criteria to be validated. Before the adoption of EHRs and integrated data management systems, approximately 20% of handwritten prescriptions contained errors such as requesting a form for medication that did not exist or confusing the names of similar medications. 


Systems governed by data models recognize these errors as they’ve entered and can prevent additional problems such as patients waiting unnecessarily long times for prescriptions because they have to contact their physicians again once the pharmacy identifies the error or – in a worst-case scenario – patients receiving incorrect and potentially harmful medications. 

In addition to preventing incorrect prescriptions, data models can also include automated alerts for adverse drug reactions, interactions, and allergies. The broader the network of data that IT systems and data models have access to, the more effective alert systems will be at catching errors or warning care providers about critical patient data. 

Open Standards and the Adoption of Healthcare Data Models

Out of data privacy concerns, the healthcare industry has traditionally resisted adopting technologies that facilitate interoperability and data exchange between organizations. Following a series of legislative efforts by Congress in the last decade to push healthcare organizations towards the use of open standards in digital records, such as EHRs, the industry has begun to shift towards replacing legacy IT systems with platforms built for standards compliance through data models. 

In 2017, the industry’s major EHR services – Allscripts, Cerner, and Epic – announced that they would offer clients access to application programming interfaces (APIs) to enable more effective interoperability and patient identification. With access to significantly larger data sets, data management service providers have begun developing the next generation of healthcare data management platforms and data models. These platforms increasingly rely on current open standards such as substitutable medical applications, reusable technologies (SMART), and fast healthcare interoperability resources (FHIR).

For organizations in the market for IT platforms employing data models, there are now three kinds of products and services to choose from. 

1. Cross-Industry Enterprise Data Management (EDM) and Analytics Technology

EDM platforms frequently offer data models as a feature of analytical and data management tools. 

2. Industry Clouds

Healthcare software-as-a-service (SaaS) clouds have begun providing pre-modeled content in their services. 

3. Healthcare-Specific Technology Vendors

Healthcare product vendors now deliver a broader range of options with industry open standards such as FHIR and the Observational Medical Outcomes Partnership (OMOP) built in. 

Experience the Next Generation in Healthcare Data Modeling with Coperor by Gaine

Coperor’s healthcare master data management platform offers organizations a comprehensive data model unparalleled in the market. With support for complex, interconnected data ecosystems, hundreds of objects, and thousands of healthcare-specific attributes, Coperor can create structure and order in even the most challenging data management environments. 

To learn more and schedule a live demo, contact Gaine today.


Opt-in with Gaine for More Insight

Keep ahead of the rest with critical insight into Healthcare and Life Sciences MDM and interoperability technique, best practices, and the latest solutions.