New USDHub Will Create Robust Data Resource to Advance Kidney Stone Research

USDHub will enable research that strengthens the evidence supporting the management of children with kidney stones.

Editor's Note: Holly Burnside is the Administrative Manager for Clinical Futures and PolicyLab, two Centers of Emphasis at Children's Hospital of Philadelphia. Lindsey Klinger-O'Donnell is the Communications Manager for Clinical Futures.

Urinary stone disease (USD) is a common and painful condition with significant morbidity; however, there have been limited advancements in its diagnosis and management. This is partly due to the lack of datasets containing important clinical information about individuals with urinary stone disease throughout their lifespan.

To help solve this problem, researchers from CHOP and Vanderbilt University Medical Center will create a novel research tool called the Urinary Stone Disease Hub, known as USDHub, using de-identified patient data from over 230,000 individuals with USD across nine health systems in the U.S.

This vital undertaking, supported by $7.37 million National Institutes of Health (NIH) grant, seeks to establish a robust data resource to bolster research efforts focused on USD. The project's leaders want to accelerate investigator-initiated research in USD to uncover novel insights that will positively impact the health outcomes of all individuals grappling with this challenging condition.

While urinary and kidney stones are traditionally thought of as an adult diagnosis, kidney stones in children are becoming more common, and CHOP clinicians encounter them daily. Among other questions, USDHub will enable research that strengthens the evidence supporting the management of children with kidney stones.

"USDHub creates a novel resource that will transform kidney stone disease research by bringing together experts in data science, machine learning, patient engagement, kidney stone disease, and learning health systems, as well as patients living with kidney stones and their caregivers," said CHOP pediatric urologist Gregory Tasian, MD, MSc, MSCE. "This community works together to identify and answer the most important research questions that can reduce children's suffering and support their healthy development."

The Urinary Stone Disease Hub, known as USDHub, is a novel research tool that uses de-identified patient data to help bolster research for USD.

USDHub will be led by Dr. Tasian and Vanderbilt urologist Ryan Hsi, MD. Dr. Tasian is an Associate Professor of Surgery and Epidemiology and serves as Director of the Center for Outcomes Research in Surgery (CORES) and the Associate Director for Clinical Trials at Clinical Futures.

The foundation for pediatric patients in USDHub is PEDSnet, a PCORnet Clinical Research Network of health systems, researchers and clinicians, and patients and families, led by CHOP Principal Investigators Christopher Forrest, MD, PhD, and L. Charles Bailey, MD, PhD. PEDSnet consists of 11 children's hospital members who provide care for 10% of the nation's children. Its mission is to conduct multicenter learning health system science projects that improve the care and outcomes of children and pediatric healthcare systems.

"Through this partnership with PEDSnet, the USDHub will finally provide us with longitudinal data and eliminate our reliance on single-center data sources that segregate pediatric and adult patients," Dr. Tasian said.

Developing a Novel Research Tool for Urinary Stone Disease Research

The USDHub's participating sites are part of PEDSnet and STAR, another PCORnet Clinical Research Network, that share a common data model of electronic health record data. Along with CHOP and Vanderbilt, these sites include University of North Carolina, Medical University of South Carolina, Duke University, Texas Children's Hospital, Nationwide Children's Hospital, Lurie Children's Hospital of Chicago, and Cincinnati Children's Hospital.  

To accomplish the first aim of the five-year project, researchers will identify children and adults with kidney and ureteral stones from four STAR and five PEDSnet sites, leveraging the size and geographic diversity of the health systems to ensure generalizability, scalability, and feasibility of future research studies. They will link extant structured data in the PCORnet common data model (e.g., procedures, anthropomorphic measurements, lab results, prescriptions) with clinically important external data, namely 24-hour urine chemistry results and kidney stone composition.

The project's second aim is to use advanced machine learning algorithms to augment the standard data model with data extracted from computed tomography images and unstructured clinical notes. They will integrate the expanded data back into the STAR and PEDSnet databases, which will establish sustainable access to USDHub data beyond the funding period.

A key third aim will be broad dissemination, making the USDHub's resources accessible to research communities at large. The USDHub team will partner with patient and caregiver stakeholders to create a public website with cohort information, data dictionaries, counts, and data access. To maximize impact, they also will:

• Offer pilot grants
• Provide clinics to orient researchers to USDHub data
• Partner with professional organizations to disseminate USDHub to USD research communities.

USDHub's overall goal is to establish a new research ecosystem that transforms the way USD research is conducted, enabling investigators to generate impactful real-world knowledge to improve clinical care and outcomes for the increasing number of both children and adults with USD.

The USDHub is funded by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK grant RC2DK140865). NIDDK is part of the NIH, the nation's medical research agency. They conduct and support biomedical research, disseminating research findings and health information to the public, and are part of the U.S. government under the Department of Health and Human Services.

Learn more in this CHOP press release.