Saving Time: How Innovations Are Transforming Diabetes Care

The initial situations differ: Type 2 diabetes often develops gradually over the course of years. In contrast, with autoimmune Type 1 diabetes, symptoms can appear rapidly, even though the autoimmune process began months or years earlier. It is precisely this symptom-free window that offers the opportunity to detect the early stages before a pronounced insulin deficiency sets in.

The Fr1da study, launched in Bavaria and led by Prof. Dr. Anette-Gabriele Ziegler, demonstrates how this opportunity can be harnessed. It screens children for islet autoantibodies, which indicate an autoimmune process targeting insulin-producing beta cells. If an early stage is detected, further testing, family education, and follow-up visits are conducted. The crucial step is combining blood testing with proactive support. “If we detect type 1 diabetes before the first symptoms appear, we can prepare families well and, in most cases, prevent severe metabolic complications. This makes an enormous difference for the children,” says Ziegler.

A Fr1da analysis from 2023 makes this difference tangible: 2.5 percent of the children previously identified through screening developed diabetic ketoacidosis—a dangerous metabolic crisis—at the time of clinical onset. Without early detection, the rate reported in Germany was more than 20 percent. Screening does not automatically prevent the disease, but it can make its clinical onset more manageable.

After diagnosis, early detection remains important—now in terms of daily glucose trends. Continuous glucose monitoring (CGM) systems provide readings every one to five minutes, depending on the system. A sensor under the skin measures glucose levels in the interstitial fluid. Instead of isolated readings and often incomplete logs, this provides a continuous trend that also reveals changes that occur overnight.

Dr. Guido Freckmann, M.D., Managing Director and Medical Director at the Institute for Diabetes Technology Ulm GmbH (IfDT), describes the benefits: “Standardized analyses such as the ambulatory glucose profile (AGP) with time in target range immediately reveal patterns—such as nocturnal hypoglycemia—and make consultations more focused.” The AGP aggregates measurements into a daily profile; time in the target range describes the proportion of readings within the specified glucose range.

But more data alone does not necessarily mean better treatment. “In my view, we don’t have a data problem when it comes to diabetes. We have a problem making something truly meaningful out of this data.” Matthias Beuth, Chief Strategy Officer at DextraData Healthcare, is thus identifying a central task of digital healthcare: information must be consolidated and organized in a way that makes sense.

Pump data on the most recent supplemental insulin dose—the bolus—and on the insulin still active in the system complete the picture. Artificial intelligence is not automatically required for standardized analyses. However, AI could identify additional patterns, predict glucose trends, or prioritize anomalies. Freckmann classifies AI-supported decision aids primarily as subjects of study or early-stage applications. Their benefits, safety, and traceability must be demonstrated: A plausible prediction does not yet constitute a reliable treatment decision.

In contrast, automated insulin delivery is already a reality in clinical practice. AID systems combine a CGM sensor, a control algorithm, and an insulin pump: Based on the glucose trend, the algorithm calculates the insulin dose, and the pump delivers it within the scope of its intended functions. “AID systems are well-established and incorporate predictive algorithms,” says Freckmann. These are not equivalent to every form of modern AI.

Nevertheless, automation has its limits. Most systems operate as hybrid closed-loop systems: parts of the insulin delivery are automated, but meals must still be entered. Managing physical activity also remains challenging. The next step in development, therefore, is systems that can reliably handle changing everyday situations with fewer user inputs. Full automation is not yet a reality in current care.

Progress in this area does not depend solely on software. Insulin must also take effect quickly enough. Ultra-rapid-acting analogs can reduce the postprandial glucose spike and are of interest for use with AID; research is underway on even faster-acting insulins. Freckmann takes a more cautious view of noninvasive glucose monitoring: A more convenient method must first achieve accuracy suitable for clinical use and, in his view, is not currently the next major breakthrough.

Whether new technology actually saves time is also determined during patient consultations. Structured glucose and pump data can reduce the need for manual entry and processing; automated preliminary analyses could take over additional routine tasks. Given the shortage of healthcare professionals, every opportunity this creates for consultation and treatment counts.

However, a continuous data connection can also create new work. Freckmann emphasizes: “Not every data point requires a response, and responsibilities as well as the time required must be clarified.” Who reviews which abnormalities, and when? Only clear responsibilities can make connectivity a viable part of the care process. For Beuth, it is therefore clear: “Digitization must reduce the workload. If it creates additional work, we’ve done something wrong.”

This shifts the focus from technical performance to practical application. “Innovations gain acceptance when their benefits, safety, and practicality are proven,” Freckmann emphasizes. Depending on the stage of development, preclinical and clinical studies are required, ideally under real-world conditions. Patients and treatment teams should be involved early on, for example in focus groups: the technology must be tailored to their needs and workflows.

For companies, the path also involves approval procedures under the Medical Device Regulation (EU) 2017/745, reimbursement issues, and the interoperability of sensors, pumps, and apps. Data protection, cybersecurity, and self-determined data use are just as important as the reliable interaction of the components.

Even if these prerequisites are met, one crucial step remains: users and care teams must be able to operate the devices and interpret the data meaningfully. “Without such training, even the best technology cannot realize its full potential,” says Freckmann.

While sensors and AID help compensate for impaired insulin supply, regenerative research focuses on its biological basis: the beta cells themselves. At the Institute of Diabetes and Regeneration Research at Helmholtz Munich, researchers are investigating approaches to preserve, regenerate, or replace these cells—with the goal of restoring the body’s natural function.

To this end, researchers are working on functional beta cells derived from stem cells and investigating processes that could enable regeneration within the body. The challenge goes beyond mere insulin production: cells must respond over the long term and in accordance with the body’s needs. However, these research goals do not indicate that a cure is already widely available. They offer a different perspective than digital therapy but do not replace its current benefits.

From early detection to cell research, innovations are making an impact in various areas. For these innovations to reach people, clinical research, sensor technology, data science, and biotechnology must work together with healthcare providers. In Bavaria, Bayern Innovativ facilitates the exchange between research, medical practice, startups, and established companies that is needed to bring new developments to where they can make a difference.

The Health Innovation Network at Bayern Innovativ operates at these intersections. It brings together expertise from medical technology, digital health, biotechnology, and pharmaceuticals, addressing issues ranging from research and development to financing, regulatory approval, and reimbursement. It connects research with medical practice, startups, and established companies, thereby creating the exchange needed to bring new developments to where they make a difference. Partners such as the IfDT contribute medical expertise to this effort. This allows technical possibilities to be aligned early on with healthcare needs and practical requirements.

In diabetes care, one goal remains: to buy time to prevent metabolic complications, identify risks earlier, and reduce the administrative burden of documentation and management. Progress begins in the laboratory and in development. Its value is evident in the daily lives of people with diabetes.