Dr. Clements Receives $8.5 Million Grant to Improve Treatment of Type 1 Diabetes

Mark Clements, MD, PhD, Endocrinologist and Director-Endocrine/Diabetes Research, has received an $8.5 million grant from The Leona M. and Harry B. Helmsley Charitable Trust to develop a Diabetes Rapid Learning Lab, which combines data and implementation science to rapidly innovate and quality-improve diabetes care. The project is co-directed by Dr. Leonard D’Avolio, PhD, Assistant Professor in Data Sciences at Harvard University and founder of Cyft, Inc.

More than 18,000 cases of T1D are diagnosed among young people under age 20 in the U.S. each year, according to the American Diabetes Association. Studies show that poor disease control at any age significantly raises lifetime risk of serious complications, such as heart disease and kidney disease.

The foundations of this research project include data science and implementation science. Specifically, “machine learning,” a set of mathematical tools that make it possible to detect patterns in large collections of data, is used to predict clinically important outcomes. Then the model for improvement in health care is used to rapidly identify interventions that improve care.

“We want to take all the data captured during routine care and make it work harder for us in a way that hasn’t happened before,” Dr. Clements said. “In the past, we’ve gleaned basic insights based on looking back at patient histories. Now, we can take retrospective data to predict future outcomes. And we can do that routinely, every week.”

For example, data can be used for predictive models that identify those patients most at risk for blood sugar control issues so that new treatments can be initiated. This is needed in pediatric diabetes care, where fewer than 20% of youth meet their goals for disease control. Statistics show that 6% of the population experiences re-hospitalization for diabetes complications within 6 months. The efforts of this team can help predict who these kids are and deploy cost-effective measures to keep them out of the hospital.

“Up until this point we have delivered the same care to everyone who walks through our doors—a one-size-fits-all approach,” Dr. Clements said. “Now, using big data approaches, we can create a new paradigm to treat diabetes—the right intervention for the right person at the right time. We’re giving ourselves the opportunity to work smarter to identify the type of care needed for each patient.”

The grant will further develop a project originally launched in 2017 at Children’s Mercy under the direction of Drs. Clements and D’Avolio, as well as Dr. Sanjeev Mehta, MD, MPH (Joslin Diabetes Center). Dr. Clements was aided by Ryan McDonough, DO, Children’s Mercy, and Susana Patton, PhD, CDE (currently Nemours Children’s Health System). All three continue to collaborate with Drs. Clements and D’Avolio to make this project a success. Additional collaborators from Children’s Mercy include Dr. Mark Hoffman (Data Science), Dr. Emily Hurley (Formative Research), David Williams (Biostatistics), Brian Rivera and Avinash Kollu (Research Compute Team), and Krista Nelson and Sally Guezuraga (Innovation Center).

The team expects this actionable information can inform which interventions—behavioral, digital and health care delivery—could be good candidates for larger clinical trials and to disseminate these learnings to other diabetes centers. Ultimately, the vision is to pair predictive analytics with the right interventions to improve outcomes for pediatric T1D patients.