Sign up for a new account.
And get access to
The latest T1D content
Research that matters
Our daily questions
Sign up by entering your info below.
Sign in to your account with
Reset Your Password
We will email you instructions to reset your
The T1D Exchange Quality Improvement Collaborative (T1DX-QI) was established in 2016 — with the support of The Leona M. and Harry B. Helmsley Charitable Trust — in an effort to refine best practices and improve daily life for people with type 1 diabetes (T1D). Growth has been tremendous, with 54 endocrine clinics from across the U.S. participating in the Collaborative.
Fueled by top leaders in diabetes care, the T1DX-QI has become an engine of innovation and inspiration. By engaging with the shared, data-driven, and systematic methods of the T1DX-QI, clinics have seen unprecedented success in their approach to diabetes management.
With members working closely together to identify gaps in care, discover and refine best practices, and share research — the process has become knowledge-sharing at its very best. While collated data gives clinics a clear sense of “where they are,” it also demonstrates “where they can be,” by applying shared, evidence-based methods for improving care.
Interview with Nestoras Mathioudakis, MD, MHS
Nestoras “Nes” Mathioudakis, MD, MHS is an endocrinologist and Associate Professor of Medicine at Johns Hopkins University School of Medicine in Baltimore, Maryland.
Mathioudakis also serves as Co-Medical Director of the Johns Hopkins Medicine Diabetes Prevention and Education Program, Co-Director of the Multidisciplinary Diabetic Foot and Wound Clinic at Johns Hopkins, and is the former Clinical Director of the Division of Endocrinology, Diabetes & Metabolism. Additionally, Mathioudakis is the Diabetes Clinical Community Lead for the Armstrong Institute for Patient Safety and Quality.
“I got hooked on endocrinology in my third year of medical school,” said Mathioudakis. “Dr. David Cooper, a world-renowned thyroidologist, was my faculty attending, and we saw some of the most interesting cases. I was intrigued by the complexity of endocrine conditions and drawn to the fact that most diagnoses were treatable.”
Mathioudakis went on to explain that being an endocrinologist is rewarding on many levels — including identifying a condition, providing person-centered care, forming personal connections with patients, and witnessing improvements in their health.
“While I started out broadly in endocrinology, I quickly gravitated towards diabetes,” said Mathioudakis, whose interest in advancing tech solutions, as well as data and machine learning, have all led to his current focus areas: quality improvement (QI) and research.
What do you enjoy most about your work?
“One of the most rewarding things is helping patients who have been struggling with their diabetes management,” explains Mathioudakis.
“T1D is a tough, demanding condition, and it’s easy to lose faith and get discouraged sometimes. Developing strategies to make improvements that are followed by ‘aha moments,’ along with joy and relief, is incredibly rewarding for a practicing physician.”
What led you to an MHS Program?
“In my first two years on faculty at Hopkins, I was the Associate Director of the Inpatient Diabetes Service, along with my mentor, Dr. Golden, a distinguished diabetes researcher. During this time, I completed a six-month fellowship in patient safety and quality training.”
“As I was compiling inpatient diabetes management data, I began to generate questions for larger-scale research,” Mathioudakis explained. In turn, he realized that more formal training in this space would allow him to perform these analyses.
As a result, Mathioudakis pursued graduate training in clinical investigation, and upon completion of his Master of Health Science degree from the Johns Hopkins Bloomberg School of Public Health in 2015, his career path began to change.
“At that point, my trajectory shifted towards research,” said Mathioudakis. “The MHS degree provided me with a strong foundation in data analysis and research methods, so I was well prepared to answer the scientific questions generated from my clinical experiences.”
With his degrees stitched together, Mathioudakis now dedicates about 20 percent of his time to seeing patients and about 80 percent towards research — broadly in the areas of health informatics, clinical decision support, machine learning applied to diabetes management, complications, and prevention.
Johns Hopkins and T1DX-QI
Johns Hopkins is a relative newcomer to T1DX-QI, having joined in the Fall of 2022, with Mathioudakis serving as the PI for adult endocrinology, along with Risa Wolf, MD, as the PI for pediatric endocrinology. With data collection and mapping underway, early ideation on T1DX-QI projects has begun.
“We’re so thrilled to be a part of collaborative,” said Mathioudakis. “One of the things I appreciate about QI-based research is how quickly it reaches patients. While other forms of research are equally as important, it can take years to translate into something beneficial. People living with diabetes need changes that can help them today.”
Reducing disparities, predicting glucose trends, and studying AI-based diabetes interventions
“A study we’re currently working on is an issue that’s near and dear to my heart — reducing disparities in access to T1D diabetes technology,” said Mathioudakis, who along with Wolf, recently published, “Racial Disparities in Access and Use of Diabetes Technology Among Adult Patients With Type 1 Diabetes in a U.S. Academic Medical Center” in Diabetes Care.
Mathioudakis recently discussed this research as a guest on the Diabetes Care On Air Podcast.
Their study demonstrated that CGM and insulin pump use is significantly lower for Black versus non-Black individuals with T1D. It highlights the fact that this disparity begins at the point of discussions with diabetes providers — a critical part of educating a patient about diabetes technology and writing the prescription.
“At Hopkins, qualitative work is underway, and we hope to have a trial in the coming year to evaluate the effectiveness of interventions to reduce these disparities. We’ve secured a grant and will be testing whether a diabetes navigator, with a personal connection to diabetes, can increase uptake and sustained use of CGMs in this population.”
Mathioudakis recently completed an NIH funded study for machine learning based decision-making in hospitalized patients at highest risk of hypoglycemia. He has another ongoing NIH-funded clinical trial comparing whether an AI-based diabetes prevention program is as effective as human-based interventions.
Mathioudakis explained that the majority of hospitalized patients aren’t wearing CGMs, and many competing conditions can affect glucose trends.
“We have rich data in EHRs, and when you utilize them to create a model, you can get high levels of predictive accuracy on where someone’s glucose is headed — in ways that human beings can’t.”
This even holds true for seasoned endocrinologists, explained Mathioudakis.
Next steps involve deployment back into the EHR as a form of extra decision support — taking into account insulin on board, glucose values over the past 24 hours, and other factors. While pilot studies are on hold for now, Mathioudakis is navigating FDA rules on regulated devices and decision-support models to push things forward.
Mathioudakis’ other research is supported by an NIH-funded R01, and that’s focused on testing an AI-based app for diabetes prevention.
“We’re randomizing people between our traditional diabetes prevention program and an AI-based app with push notification-based algorithms. In essence, comparing AI and human-based coaching with outcomes people are having — such as changes in weight, physical activity, and A1C values,” said Mathioudakis. “We’re close to completing enrollment, and we hope to have results by 2025, so stay tuned.”
Precision Medicine Initiative
“I’ve been co-chairing an International Precision Diabetes Medicine Initiative through the American Diabetes Association. There are 12 working groups, and we’ve just wrapped up one of the most comprehensive systematic reviews on T2D precision prognosis for cardiovascular disease, including over 10,000 articles.”
“It’s been a very fulfilling collaboration with international experts. We’re excited to see the culmination of these extensive reviews in this forthcoming consensus report on personalized recommendations, which we hope to share at the 2023 EASD meeting in Hamburg, Germany” said Mathioudakis.
What’s your hope for future diabetes-related tech?
While Mathioudakis emphasized how difficult it is to predict the development of diabetes tech.
“Having reviewed upcoming machine learning algorithms, I can say that in general, we’re getting better and better at it, the field is advancing, and it’ll reach patients in more meaningful ways.”
“Just look at the progress we’ve seen over the past decade,” he added. “The pace with which diabetes tech has evolved and the attention to machine learning in medicine is impressive — we’ll keep seeing improvements as algorithms in closed-loop systems are refined and become more rapidly adaptive.”
“We have more work to do,” said Mathioudakis. “One of the biggest challenges is streamlining how data is uploaded from patient devices to improve accessibility for clinicians, ideally in a universal platform.”
Outside of work, Nes Mathioudakis enjoys traveling, tennis, running, and playing the violin. While Nes chose a career in medicine, he almost became a professional violinist. He can often be found helping his children with their violin lessons, as they follow in his footsteps.
This site uses Akismet to reduce spam. Learn how your comment data is processed.
Meet the Expert: Asking the ‘Why?’ Behind Racial Disparities in Diabetes Tech Use
You must be logged in to post a comment.