What is Open-Source?

Open-source technology is increasingly shaping the landscape of diabetes care. Yet many healthcare professionals, especially those outside technology-focused practice areas, remain uncertain about what “open-source” means and how it applies to their clinical practice.

At its core, open-source software is code that is publicly available for anyone to inspect, modify, and share. In diabetes care, this movement began as a grassroots, consumer-driven effort: people with diabetes and caregivers wanted timely access to continuous glucose monitoring (CGM) data to improve safety and quality of life.

By 2014, individuals in the #WeAreNotWaiting community had developed methods to intercept the CGM signal and securely transmit it to smartphones. Soon after, they created algorithms to automate insulin delivery through compatible pumps. These efforts evolved into open-source automated insulin delivery (AID) systems, now used worldwide.

Understanding “DIY” AID Systems

Open-source systems are often called “DIY” (Do-It-Yourself) because they rely on user-built applications. Importantly, “DIY” does not mean starting from scratch. Most open-source AID apps—such as OpenAPS, Loop, Trio, and AndroidAPS—are assembled using detailed, step-by-step instructions.

Users do not need advanced coding skills. Instead, they follow structured directions to load the prebuilt software onto compatible devices (specific FDA-approved pumps and CGMs). While the apps themselves are not FDA-approved, they are in use by tens of thousands globally, with strong safety records and real-world outcomes that compare favorably to many commercial systems.

How Open-Source AID Systems Work

At its core, an open-source AID system brings together three components: a continuous glucose monitor (CGM), an insulin pump, and a smartphone app that runs the open-source algorithm. The CGM provides real-time glucose readings, which the app uses to calculate insulin adjustments. The app then communicates directly with a compatible pump via Bluetooth or a small radio “bridge” device, creating a continuous feedback loop:

CGM → App Algorithm → Pump → Patient → CGM.

Pump Compatibility

Unlike commercial systems, open-source apps only work with certain insulin pumps:

• Older Medtronic pumps (such as the 515/522/722 models) are widely used because their communication protocol can be accessed.
• Omnipod Eros (the original pod system, not DASH or Omnipod 5) is supported by Loop.
• Dana Diabecare RS and Insight pumps (Europe) are compatible with AndroidAPS.
• Accu-Chek Combo and Spirit pumps (Europe) are also used with AndroidAPS.

Patients typically obtain these pumps through existing prescriptions, second-hand sources, or regional availability. While the pumps themselves are FDA-approved devices, the open-source software that controls them is not.

For clinicians, it is important to understand that patients using open-source AID may be on older or less common pumps, but the safety and functionality of the system rely on the integration of hardware and software. Asking which pump and which app a patient is using provides valuable insight into the system’s safety features, data-sharing capabilities, and limitations.

Safety, Oversight, and Support

A common concern among clinicians is whether open-source use places patients at risk due to the absence of formal regulatory approval.

Key Considerations Include:

• Community-driven safety: Development is transparent and collaborative. Updates are tested extensively by experienced-user panels before broader release.
• Adverse events: Reported issues have typically stemmed from device misuse (pump/CGM) rather than the software algorithms themselves.
• Support: While no formal customer service exists, robust peer-to-peer networks provide near-immediate guidance and troubleshooting. These communities include engineers, programmers, caregivers, and highly experienced users.

The Clinician’s Role

For DCESs and other clinicians, the question is not whether to endorse non-FDA-approved systems, but how to engage with patients who are using them.

• Many people with diabetes are highly engaged, data-literate, and motivated to optimize their care.
• Dismissing or ignoring their choices risks undermining trust and therapeutic relationships.
• Instead, clinicians can ask open, nonjudgmental questions about device function, settings, and outcomes.
• Supporting patient safety does not require prescribing these tools—it requires awareness, curiosity, and a willingness to partner in the patient’s self-management journey.

As with other complementary approaches in medicine, staying informed helps clinicians provide balanced guidance. Transparency is critical: when people with diabetes are well-informed, they are better positioned to make decisions aligned with their health and quality of life.

Key Takeaway

While open-source technology is not FDA-approved, it is an important and growing reality in diabetes care that you as a clinician need to know about. Until regulatory pathways and commercial options can fully match the speed of innovation, clinicians familiar with open- source play a vital role in supporting, educating, and safeguarding individuals who choose to use these systems.

For More Information on Open-Source technology in Diabetes Management

• LoopDocs: Introduction to Open-Source AID
• https://openaps.org/