Leveraging De-Identified Clinical Data Sets for Research and Development: Three Use Cases

Leveraging De-Identified Clinical Data Sets

Healthcare providers and health information exchanges (HIEs) possess vast quantities of valuable clinical information—robust, realistic clinical data sets that are in high demand. These data sets not only have tremendous value for internal purposes — such as quality benchmarking and user acceptance testing of new functional capabilities — but also power a variety of use cases for outside data consumers, such as public health research, clinical trial research, and health economics and outcomes studies.

However, making meaningful use of all this data — de-identifying it to protect patients while still retaining the internal linkages between records that make it useful for analysis — is not so simple.

In order to create de-identified population health data in accordance with government policies and safe harbor requirements, these organizations need a capable partner who can handle the complexities of transforming large volumes of clinical information into anonymized data sets that support actionable knowledge.

J2 works closely with providers, HIEs, and other healthcare organizations to turn their clinical data into a valuable asset for managing population health and conducting critical research. The result is an expanded definition of which data consumers can derive value from the information and how they can benefit from it. We’ve provided three examples below.

Data Fit for Research

Healthix, one of the largest public HIEs in the nation, stands ready to support researchers with aggregated data of more than 20 million patients from across the New York region.

Recently, with the support of New York City’s largest public health agency, Healthix and J2 built a longitudinal research data warehouse, which was deployed in July 2023. The data warehouse, which complies with state policies for de-identified data standards, has been certified to meet those criteria with de-identified PHI / PII fields, and modified subsequent inbound data ingestion so that future data is loaded without PHI and PII.

The resulting data contains de-identified clinical information spanning more than 15 years’ worth of care delivered since the inception of New York State HIEs in 2005, allowing researchers to look for and analyze patterns in the data pertaining to population health. Learn more here.

Clinical Trial Populations

A non-profit academic medical center in Miami, Florida, and a nearby university wanted to analyze clinical data stored on their data platform to identify patient cohorts eligible for clinical trials. They needed a solution for de-identifying historical data and providing ongoing weekly updates.

Using a J2-developed anonymization utility that significantly reduced the amount of custom work required, J2 built a HIPAA-compliant data store,allowing the joint team’s analysts to accurately identify patient cohorts that meet requirements for clinical trial studies. J2 also constructed a tailored workflow that enables authorized researchers to tie anonymized records back to individual patients. Using this feature, each organization has the ability to identify and approach potential study participants, but only with the proper safeguards and approval process.

Advanced Analytics and Machine Learning

In addition to research, providers use clinically representative data sets to develop advanced analytics and machine learning algorithms. J2 enabled a large, not-for-profit health system in New England to generate these types of data sets for use by licensed technology partners in their own product development and testing efforts.

J2 de-identified clinical data for the provider’s community of patients while maintaining logical links between patient records so that algorithms could track clinical data longitudinally for a given patient without providing any patient-identifying characteristics.

The Bottom Line

In all three cases, experts at J2 applied their deep understanding of clinical data, HIPAA data privacy regulations, and extensive technical subject matter expertise to deliver unique solutions that address significant data de-identification challenges. In doing so, we enabled our clients to open up new avenues for deriving value from their large volumes of clinical data — for the benefit of their respective communities, while protecting the privacy of each individual patient in those communities.

Let’s Talk

J2 is committed to applying our experience to put your clinical data assets to work for the benefit of your organization, stakeholders, and patients. Drop us a line to learn more about how we can help you achieve the goals that matter most to you.

Contact Us →

J2 Interactive

J2 Interactive is an award-winning software development and IT consulting firm that specializes in customized solutions for healthcare and life sciences.