ONC Seeks Synthetic Health Data Engine Capabilities to Support Researchers

ONC Seeks Synthetic Health Data Engine Capabilities to Support Researchers

A new competition encourages ideas to validate and innovate synthetic health data to improve research in quality of care.

A new challenge within the Department of Health and Human Services aims to explore advanced capabilities for a synthetic health data engine to strengthen how patients make smart and informed health care decisions.

HHS’ Office of the National Coordinator for Health Information Technology (ONC) launched the competition, called the Synthetic Health Data Challenge, this week to support the data infrastructure for the research it is doing around patient-centered outcomes. 

Improving patient-centered outcomes with research is critical to providing high-quality health care, ONC said, but gathering quality data can be difficult because of the cost of data and privacy concerns and restrictions around patient consent. Synthetic health data can support quality of care research, and the agency is looking to develop its synthetic health data engine, Synthea, to accelerate it.

“Synthea is an open-source synthetic patient generator that models the medical history of synthetic patients,” ONC said. “The resulting data are free from cost, privacy and security restrictions and have the potential to support a variety of academia, research, industry and government initiatives.”

Synthea’s software has a temporal model that covers a patient’s entire lifetime instead of focusing on singular instances of health complications or infections, ONC added, emphasizing that Synthea’s generated synthetic data must be validated to ensure that its samples are realistic. 

The competition is centered around developing tools that can support the validation of the synthetic data and find novel uses of this generated data for researchers and health IT developers. These two goals shape the two entry categories of the challenge.

“Synthetic data like those created by Synthea can augment the infrastructure for patient-centered outcomes research by providing a source of low risk, readily available, synthetic data that can complement the use of real clinical data," ONC Chief Scientist Teresa Zayas-Cabán said.

The challenge has two phases. In phase one, participants will submit a written proposal for innovative models describing their solution, including methodology and intended outcomes. Selected proposals will continue to phase two, which will launch once a minimum of four qualified phase-one proposals are selected. The second phase involves the development of participants’ prototypes or solutions.

ONC will select up to six winners with prizes ranging from $10,000 to $50,000, with a cap of $100,000 for the total award amount. 

ONC started accepting proposals for the first phase Jan. 19 and will close submissions March 2. The second phase submission period will open March 23 and close July 13. ONC will hold informational webinars Feb. 2 and April 6, and will announce award winners Sept. 21.

Standard