UF Health researchers use big data to improve care for people with rare diseases

Researchers at University of Florida Health and the OneFlorida+ Clinical Research Network are using big data analytics to scale up research and boost care and treatment options for people affected by three potentially debilitating rare diseases.

Study teams at UF Health will conduct research on preserving kidney function in children with chronic kidney disease and treating rare neuroendocrine cancers in adults. A team of scientists at Nicklaus Children’s Hospital in Miami, a OneFlorida+ clinical partner, will evaluate treatment and care options for children with a rare form of epilepsy known as Lennox-Gastaut syndrome.

All three studies are funded by the nonprofit Patient-Centered Outcomes Research Institute, or  PCORI, which has made rare disease research one of its funding priorities.

Rare diseases are often chronically debilitating and sometimes life-threatening.

“Treating and caring for people with rare diseases poses many challenges,” said François Modave, Ph.D., an associate professor and director of artificial intelligence and decision-making in the department of health outcomes and biomedical informatics at the UF College of Medicine. “Because the number of people affected by any given rare disease is so small, research funding and other resources are often in short supply,” he said.

The small number of affected patients also makes it hard for researchers and clinicians to develop and test new treatments. Clinical trials needed to test the safety and effectiveness of a new treatment typically require large numbers of study participants to produce reliable results.

About 25 million Americans are affected by rare diseases — medical conditions that affect fewer than 200,000 people nationwide, according to the U.S. Centers for Disease Control and Prevention.

 Although the number of affected individuals for any particular rare disease may be small, the total number of Americans diagnosed with rare diseases is nearly 10% of the U.S. population — about the same as the number of Americans with diabetes.

“Most rare diseases are not studied enough,” said Modave, who is also a member of the UF Health Cancer Center.

As a result, patients with rare diseases often undergo years of testing before they are properly diagnosed and many receive ineffective treatments that haven’t been adequately tested.

The three OneFlorida+ studies are part of PCORI’s $48 million 2021 Conducting Rare Disease Research initiative. It will use PCORnet’s national repository of electronic health records for 66 million patients as tools to identify and enroll patients as well as analyze data.

PCORnet’s national network of eight regional clinical research networks — including OneFlorida+ — gives researchers the ability to study conditions affecting even small numbers of people while maintaining confidentiality of protected health information and patient privacy.

Modave said these studies provide just a few examples of how data science, big data analytics and artificial intelligence are fundamentally changing the way scientists conduct health research. By pooling electronic health records and other data into a central location and applying data analytics and AI tools, scientists can now conduct health research on a scale that simply was not possible even 10 or 15 years ago — and at a much faster pace.

In the past, locating the small numbers of patients with rare diseases to participate in clinical trials may have involved using advertisements, fliers and other methods to get the word out. Trained medical staff would be paid to spend countless hours manually searching tens or even hundreds of thousands of medical records.

“The work was labor intensive, extremely slow and very expensive — like looking for a needle in a haystack,” Modave said. “Now, we can develop sophisticated search queries known as computable phenotypes that enable computers to comb through the electronic health records of millions of patients with incredible speed and help us locate the small number of patients affected by these rare diseases in a matter of hours.”

By tapping into the OneFlorida+ Data Trust, a repository of secure electronic health records for millions of patients in Florida and select cities in the Southeast, researchers at UF Health and OneFlorida+ hope to quickly identify and enroll patients in these studies. The researchers also will use electronic health records to evaluate the effectiveness of various treatments on health outcomes.

UF Health serves as the coordinating center for the OneFlorida+ Clinical Research Network, which includes 12 academic centers and health systems across Florida, together with Emory University in Atlanta and the University of Alabama at Birmingham. OneFlorida+’s network of 22 hospitals, 1,240 practices/clinics and 4,100 providers care for about 74% of Floridians in all 67 counties and in select metropolitan areas of Georgia and Alabama. OneFlorida+ is also one of eight clinical research networks nationwide participating in PCORnet, the national patient-centered clinical research network.

Learn more about the individual studies.

Media contact: Ken Garcia at kdgarcia@ufl.edu or 352-265-9408.