In 2020, Children’s Mercy Hospital received an $8.5 million grant from the Leona M. and Harry B. Helmsley Charitable Trust to lead the Rising T1DE Alliance (RTA), an initiative to improve Type 1 Diabetes care by leveraging predictive analytics with the support of data analytics firm Cyft to improve inform behavioral care interventions.

Improving diabetes care in children is a pressing health need, not only due to the scale of the problem among children today, but also because of the way that poor glucose management early in life can lead to elevated risks of serious complications throughout an individual’s lifetime. According to data aggregated from across the country, only around 20% of youth in U.S. diabetes centers are meeting their goals for glucose control, demonstrating the need for improved care.

Currently, physicians have a limited number of treatment options for pediatric diabetes care. Most of these options tend to focus on the use of new pharmaceuticals or new devices, with relatively little attention paid to how behavioral and care delivery interventions could help improve diabetes management. These alternative intervention options offer a way to ensure that personalize healthcare is tailored not just to an individual’s biological and genetic profile, but also their behavioral profile.

The vision of RTA is to deliver these interventions based on predictive models that can forecast patient outcomes, allowing care providers to segregate patients according to their risk profile and provide them with personalized care plans. For example, the team is already using a model that can predict 90-day changes in blood glucose control to deliver target home telehealth monitoring, peer mentorship, and digital therapeutics to high risk patients.

One of the primary advantages of this approach is the way it allows care providers to undertake rapid testing of care options and iteratively improve the interventions they offer. In contrast to traditional clinical trials, which are slow, expensive, and provide limited information about what interventions work best for different groups of patients, behavioral and care delivery interventions are low risk and can be rolled out without needing formal trials. By collecting data about the targets of these interventions, it becomes possible to constantly monitor for whether a given care plan is improving the patient’s outcomes. Based on this information, physicians can decide whether to continue or expand a given intervention, or pivot to something different.

The data used to for predicting patient risk and evaluating intervention outcomes is taken from a variety of sources, including clinic notes, personal diabetes self-management devices, consumer wearables, and patient-reported outcomes. The RTA team has created a data flow that automates the collection and sharing of this data with the support of Cyft. Ultimately, the team hopes that this work can help create a virtuous cycle where health data generated by patients’ digital devices can be used to continuously assess their individualized risks, inform remote care interventions, and measure outcomes. While work so far has focused on diabetes management, the team believes that this model should be replicable by clinics focused on other kinds of chronic conditions as well.

The ultimate goal of this project is to form a sustainable entity that can help diabetes centers and other chronic care delivery services across the country use predictive analytics and rapid testing to improve the care they provide to patients. If you are interested in learning more, visit

Dr. Mark Clements, professor of pediatrics at Children’s Mercy Kansas City, presented to KC Digital Drive’s Health Innovation Team on August 25, 2021.

Further Reading

AR/VR Heartland Developer Challenge Demo Day

On February 3rd, KC Digital Drive hosted a demo day at the T-Mobile campus for the Augmented Reality/Virtual Reality (AR/VR) Heartland Developer Challenge, a year-long program to bring together entrepreneurs, hackers, makers, developers, and other creative minds with the goal of building AR applications that solve local problems. The full list of developers who presented […]

Read More

MoodSpark’s digital companion combats loneliness in seniors by detecting signs of sadness and uplifting with emotional memories and personalized interventions

MoodSpark was founded in 2020 to develop a digital companion for seniors suffering from loneliness, anxiety, and depression. MoodSpark’s companion is designed to detect signs of sadness and distress in elderly patients and respond with emotional memories, personalized conversations, and video visits with friends and family that can help to distract, re-direct, and uplift.

Read More