Omnipod Insulin Pump for Type 1 and Type 2 Diabetes

The Omnipod is a tubeless insulin pump worn directly on the skin as a small adhesive pod, distinguishing it from traditional pumps that connect to the body through several feet of plastic tubing. Its latest version, the Omnipod 5, pairs the pod with a continuous glucose monitor and an algorithm that automatically adjusts insulin delivery throughout the day and night. Clinical trials in adults with type 1 diabetes show time-in-range improvements of roughly 16 to 18 percentage points and sustained reductions in HbA1c, and newer research is extending those findings to children, people with type 2 diabetes, and exercise scenarios.

How the System Works

The physical setup has two main pieces. The pod itself is a waterproof, disposable unit about the size of a small egg. It holds up to 200 units of rapid-acting insulin and attaches to the skin with built-in adhesive, typically on the abdomen, upper arm, or lower back. A short, automatically inserted cannula sits just beneath the skin and delivers insulin subcutaneously. Each pod lasts up to three days before it needs to be replaced.

The second piece is the controller, which in the Omnipod 5 system is a compatible smartphone app rather than a dedicated handheld device. The app communicates wirelessly with both the pod and a Dexcom continuous glucose monitor (CGM). Every five minutes, the algorithm reads the CGM value, predicts where glucose is heading, and adjusts basal insulin delivery up or down. Users still enter carbohydrate counts and confirm meal boluses manually, but the background delivery runs on autopilot. You set a glucose target, and the algorithm works to keep you near it.

Glucose Control in Adults with Type 1 Diabetes

The headline metric for any insulin delivery system is time in range, meaning the percentage of the day spent with blood glucose between 70 and 180 mg/dL. In a 12-month extension of a randomized controlled trial, adults using the Omnipod 5 increased their time in range by about 18 percentage points compared to baseline on their previous therapy, gaining roughly four extra hours per day in target. HbA1c dropped from an average of 8.3% down to 7.2%, and those improvements held steady through the full year.¹

A separate real-world study looking at a broader group of adults confirmed the direction of these results, reporting a 16-percentage-point increase in time in range and an HbA1c reduction of about 3 mmol/mol across the whole group. People who started with the poorest control benefited most, with HbA1c dropping by roughly 7 mmol/mol among those whose baseline was already elevated.² That pattern is consistent across automated delivery systems generally: the worse your starting control, the more room the algorithm has to make a difference.

Children and Very Young Users

Managing diabetes in toddlers and preschoolers is uniquely difficult. Tiny insulin doses, unpredictable eating, and the impossibility of a three-year-old reliably reporting symptoms make every tool more important. The Omnipod 5 is cleared for children as young as two, and a pivotal trial in very young children (ages two to under six) showed HbA1c falling from 7.4% to 6.9% after starting the system. Time in range rose from about 57% to 68%, and those improvements held for up to two years of follow-up.³

During the extension phase, participants spent a median of 97% of their time in the system’s automated mode, suggesting that families were able to keep the system running consistently rather than constantly overriding it. Serious events were rare: one episode of severe low blood sugar and one case of diabetic ketoacidosis over the entire follow-up period.⁴ For parents, the tubeless design has a practical advantage: there is no tubing for a child to catch on furniture or pull out during play.

Expanding to Type 2 Diabetes

Automated insulin delivery has historically been developed for type 1 diabetes, but a growing number of people with type 2 diabetes use insulin as well, and many of them struggle with the same glucose variability. A clinical trial published in JAMA Network Open tested the Omnipod 5 in adults with type 2 diabetes who had been using insulin injections or a standard pump. After 13 weeks, average HbA1c dropped from 8.2% to 7.4%, and time in range jumped from 45% to 66%, an increase of about 20 percentage points. Time spent in dangerously low glucose ranges did not increase.⁵

That last point matters. One of the persistent fears with tighter glucose control is that pushing numbers down will cause more lows, which can be dangerous. The trial data suggest the algorithm threads that needle reasonably well, tightening overall control without a corresponding rise in hypoglycemia. Whether these results hold up in longer trials and in more diverse populations is still being studied, but the early evidence is encouraging for the large number of people with type 2 diabetes who are already on insulin and not hitting their targets.

Handling Exercise

Exercise is one of the trickiest scenarios for any insulin pump. Physical activity increases how quickly muscles pull glucose from the blood, and insulin that was delivered before the workout keeps working while this happens. The result can be a steep glucose drop that catches people off guard. The Omnipod 5 includes an “Activity Feature” that users can activate before a workout, raising the glucose target so the algorithm dials back insulin ahead of time.

A randomized controlled trial tested this feature by comparing the system’s standard automated mode to the activity setting activated either 30 or 60 minutes before exercise. Both pre-exercise activation windows reduced insulin delivery in the hour before and during exercise. Glucose still dropped during the workout in all groups, but the fall was notably smaller when the activity feature was engaged: about a 36 mg/dL drop with 30 minutes of lead time compared to 57 mg/dL in standard mode.⁶ The percentage of participants who developed actual hypoglycemia during exercise was lower with the activity feature, though the difference did not reach statistical significance in this particular trial. The practical takeaway: turning on the activity feature at least 30 minutes before working out helps, even if it does not eliminate exercise-related lows entirely.

Why People Stick with Tubeless Pumps

Choosing between a tubeless pod and a traditional tubed pump often comes down to lifestyle preferences, but real-world data suggest the choice also affects how long people stay on pump therapy. A large retrospective study using French national health data found that people who started on tubeless pumps had roughly 40% lower risk of stopping pump therapy compared to those who started on tubed pumps.⁷ Tubeless users were also less likely to switch between therapies within the first year.

Why the difference? The study did not dissect the reasons, but plausible factors include the absence of tubing (which can snag, kink, or disconnect), the discretion of a small pod under clothing, and the simpler setup process. None of this means tubed pumps are inferior in terms of glucose outcomes. Several tubed systems have their own strong clinical data, and some people genuinely prefer having a separate pump unit they can detach more easily. But persistence matters in diabetes management: a device you stop using delivers zero benefit, and the data here suggest that the tubeless form factor keeps more people engaged long term.

Quality of Life, Sleep, and Emotional Burden

Diabetes management is relentless. Counting carbs, checking glucose, dosing insulin, worrying about lows at night, and carrying supplies everywhere creates a psychological load that clinical trials often overlook. Newer research has started measuring these softer outcomes more carefully.

In a trial sub-study comparing the Omnipod 5 to sensor-augmented pump therapy (a pump plus CGM but without automated dosing), the automated group saw greater improvements in overall quality of life, perceived impact of diabetes on daily life, system usability, and sleep quality.⁸ The sleep finding is worth highlighting on its own: the algorithm adjusts insulin overnight without the user needing to wake up, which likely explains why participants reported sleeping better.

An earlier study in adults starting the system found improvements across diabetes distress, confidence in managing lows, and overall satisfaction with insulin delivery after just three months of use.⁹ And a randomized trial comparing the Omnipod DASH (the non-automated predecessor to Omnipod 5) against usual care found a large jump in treatment satisfaction in the Omnipod group, while the control group’s satisfaction did not change at all.¹⁰ Taken together, the pattern is consistent: automated tubeless delivery reduces the emotional weight of managing diabetes, not just the glucose numbers.

Pregnancy and Its Limitations

Pregnancy with type 1 diabetes demands much tighter glucose targets than usual. Guidelines generally recommend keeping glucose below 140 mg/dL after meals, well below the standard 180 mg/dL ceiling used in non-pregnant adults. The Omnipod 5’s algorithm was not designed for those tighter targets, and its lowest glucose target setting is still higher than what pregnancy guidelines call for.¹¹

That does not mean the system is never used during pregnancy. Some people use it with close clinical supervision and frequent manual overrides. In a small case series of six pregnant women with type 1 diabetes using different automated systems, the two on Omnipod 5 achieved average time-in-range values of 68% and 82% across trimesters.¹² Those results varied widely between the two individuals, and with only two data points it is impossible to draw conclusions. A review of the field noted that the need for lower glucose targets and frequent dose adjustments throughout pregnancy represents a genuine limitation of all current FDA-cleared algorithms, not just Omnipod’s.¹³ Pregnancy-specific algorithms are in development, but none have been approved yet.

The Environmental Footprint

Every Omnipod gets thrown away after three days. Unlike a traditional pump that lasts for years with replaceable infusion sets, the pod contains its own motor, battery, electronics, and plastic housing, all of which go in the trash. An estimated 43.8 million Omnipod units are discarded globally each year.¹⁴

A lifecycle analysis comparing the resource loss from various insulin delivery devices found that the waste footprint of the Omnipod dwarfed that of standard infusion sets used with tubed pumps. The study measured “loss of resources through waste,” a metric that accounts for the embedded materials and energy in discarded components, not just their physical weight. By that measure, the resource loss from Omnipod use was orders of magnitude greater than from traditional infusion sets, driven largely by the electronics and batteries discarded every three days.¹⁵ Insulet, the company behind Omnipod, has launched a recycling program in some markets, but the scale of the problem is substantial given tens of millions of units discarded annually. For people who weigh environmental impact alongside clinical benefit, this is a genuine trade-off that the tubeless design imposes.

Cost-Effectiveness

The Omnipod 5 costs more upfront than multiple daily injections, and the ongoing pod supply adds recurring expense that pen needles do not. But diabetes costs extend far beyond the price of the delivery device. Poorly controlled blood sugar leads to complications like kidney disease, nerve damage, and cardiovascular events, all of which are expensive to treat and devastating to quality of life.

A cost-effectiveness analysis from a Swedish payer perspective modeled the long-term economic picture. It found that the Omnipod 5 was cost-effective compared to multiple daily injections, primarily because the improved glucose control led to fewer diabetes-related complications over time. The model projected an additional 3.8 quality-adjusted life years for Omnipod 5 users. Higher device costs were partially offset by savings from avoided complications and reduced work absenteeism.¹⁶ These models depend heavily on assumptions about how well people maintain glucose improvements over years and decades, but the direction of the finding aligns with what health economists have consistently found for pump therapy more broadly: the upfront cost premium shrinks or disappears when you account for avoided complications.

Insurance coverage varies widely. In the United States, most private insurers and Medicare cover insulin pumps, but the specifics of copays, prior authorization requirements, and supply coverage differ by plan. In many European countries, national health systems cover the Omnipod 5 for people with type 1 diabetes who meet clinical criteria, though access for type 2 diabetes is still limited in most markets pending additional regulatory approvals.

Occlusion Detection and Hardware Quirks

One less-discussed aspect of any pump is what happens when the cannula gets blocked. An occlusion means insulin stops flowing, and blood glucose can rise quickly if the blockage goes undetected. Lab testing of tubeless pumps found that the time to detect an occlusion varied considerably depending on the flow rate. At a moderate basal rate of 1 unit per hour, detection happened within roughly three to seven and a half hours. At very low rates of 0.1 units per hour, detection took more than 24 hours or did not occur at all. This is not a flaw unique to the Omnipod; it reflects a physics problem shared by all small pumps delivering tiny volumes. But it means that people on very low basal rates, including young children, should be especially attentive to unexplained glucose rises that might signal a delivery problem.

Site adhesion is another practical concern. The pod stays on for three days through showers, sweat, and sleep. Most users find the adhesive holds well, but in hot climates or during heavy exercise, edges can peel. Skin irritation at the adhesive site is a known issue with any wearable medical device, and rotating pod placement between different body areas helps minimize it. Some users apply additional adhesive patches or skin-prep wipes to improve durability, though these are not officially part of the system.