I tested an AI-powered glucose monitor against a traditional monitor for two weeks. Here’s my verdict

ZDNET

Also: I tested a $49 OTC continuous glucose monitor for two weeks – here’s what I learned After applying, it asks you a few personal questions, and from there, gives you a target range of glucose.
Spike alerts My last trial using Abbott’s Lingo CGM helped me tackle broader eating habits.
However, there are many reasons why your Stelo CGM reading might not match up to a standard finger-prick monitor.
This is because Stelo reads the glucose levels from your interstitial fluid, while the finger-prick monitor measures glucose from your blood.
Altogether, I found that the Stelo monitor rarely dipped below or above the 20% margin.

POSITIVE

Since receiving a prediabetic and insulin-resistant diagnosis in middle school, I have been committed to managing my metabolic health and working to reverse the diagnoses that have followed me into adulthood. I tested Abbott’s Lingo earlier this year, and I’ve seen the positive effects a trustworthy CGM can have on my health.

In addition, this over-the-counter glucose monitor inspired me to alter my eating patterns. Here’s how.

I was therefore interested in seeing how Dexcom’s Stelo, the first over-the-counter CGM to be released last year, would affect my health and to compare it to the Abbott Lingo. During my two-week testing period, I contrasted the Stelo’s results with a conventional finger-stick monitor because Dexcom claims that the Stelo has an accuracy rate of 93%. This is how it transpired.

putting the sensor to use.

You will need to apply the sensor before you can start tracking. You simply stamp the app’s detailed instructions onto the back of your arm, in an area with adequate fat and three inches away from any injection sites.

To activate the sensor’s application and release a flexible needle, the Stelo uses a button. After a few minutes, the heat and adrenaline I had been experiencing subsided. The worst aspect of the procedure was this. Instead of feeling like a typical doctor’s office shot, the needle felt more like having your ears pierced with a piercing gun. I also felt that Stelo’s sensor fit on the back of my arm more comfortably than Lingo’s because it is slightly smaller.

Also, here is what I discovered after two weeks of testing a $49 over-the-counter continuous glucose monitor.

Following your application, you will be asked a series of personal questions before receiving a glucose target range. Your resting rate may be 70, but after eating, it should rise to >140 to reflect food intake. That’s what happened to me.

You have to wait around 20 minutes to get active readings from this sensor. The process then starts as you record your diet, stress levels, and physical activity for 15 days. The AI software in the app then uses the information you collected from your readings, events, and meals to provide you with tips on how to stabilize your blood sugar.

Additionally, the Oura 3 and Ultrahuman are surpassed by one of the greatest smart rings I’ve tested.

The Vertex AI platform on Google Cloud is used by the Gen AI software to customize the Stelo experience. For the time being, the AI only offers weekly insights based on the data gathered by the sensor and the activities, stress, and food you record. Dexcom is still working on the technology because it is still in its early phases.

alerts for spikes.

My previous experiment with Abbott’s Lingo CGM assisted me in addressing more general eating patterns. For example, Lingo’s weekly insights helped me prioritize protein, reintroduce breakfast, and gain insight into how stress impacts my body’s glucose levels. I also played detective to determine what foods were most effective for my body.

Additionally, for a few weeks, I tracked my blood sugar using the new CGM integration on Oura Ring. I learned the following.

Similar to the Lingo, the Stelo demonstrated to me the extent to which stress can impact glucose and, consequently, the body as a whole. My blood sugar levels seemed to rise most when I was driving and playing video games with my friends. After playing Fortnite with my friends, I realized that I need to manage my emotions during stressful situations and give my body plenty of nutrients, like protein, vegetables, and healthy fats, rather than high-sugar carbohydrates.

Keeping a careful eye on which carbohydrates had the biggest effects on my body was also crucial to my Stelo experience. I discovered that diet isn’t everything or nothing. As long as you combine them with nutrient-dense or activity-rich foods, occasionally indulging in your favorite high-sugar foods or snacks won’t harm you.

Reading the app’s instruction and monitoring your blood sugar levels in real time are two ways to accomplish this. The app’s spike alerts were helpful in preventing me from continuously monitoring my blood sugar levels.

Also read: Expert-tested smartwatches that are the best available.

For a well-informed opinion, you can also link your data to your medical professional. You can enter the code your provider has provided for their clinic in the Stelo app. Once pasted, Dexcom’s Clarity software automatically streams your data to your provider. Your provider may also provide you with a hard copy of your data; however, this cannot be done via the app.

My expectations for the upcoming model.

Finally, I discovered that exercise, even a brief burst during the day or after a large meal, is essential for stabilizing blood sugar levels. The data from the app allowed me to see the effects, even if I only took a daily walk of fifteen minutes. My average blood sugar level when I first started the sensor was in the 60s, slightly below the desired resting range of 70. The physical effects of lower blood sugar were noticeable to me, even though it is not harmful.

Also: 5 professional-recommended sleep-improvement strategies, particularly if you have a wearable health monitor.

Following a week of tracking my meals and stress levels, the app ultimately recommended incorporating a modest amount of exercise to help stabilize my blood sugar levels. I started walking 15 minutes before work and 15 minutes after dinner the following week. My blood sugar level gradually increased to the 70–80 range before meals, and it continued to rise steadily afterward with few peaks or troughs.

But when my birthday weekend arrived, I ate birthday cake, drank alcohol, and ate hot chicken without exercising, which completely undid all of my progress. I could actually see the graph showing that I was experiencing more spike and crash alerts than I had when I was eating mindfully.

Does the Stelo have accuracy?

I came upon some Reddit posts while researching the Stelo that reported readings that were erroneous in comparison to those of other readers. Since this is the product’s initial iteration, some errors are inevitable. A typical finger-prick monitor may not display your Stelo CGM reading for a variety of reasons.

First, there is a 20 percent margin of error when comparing your Stelo readings to a traditional finger-pricking monitor. The finger-prick monitor measures glucose from your blood, whereas Stelo reads the levels from your interstitial fluid. Naturally, when it comes to glucose levels, your interstitial fluid lags behind your blood. It particularly lags after a meal high in carbohydrates or sugar.

Also: How I came to love my health tracker and quit worrying.

There is also a healthy margin of error in the majority of glucose monitors available on the market. This is due to the fact that a number of external factors, including temperature, user error, interference from test strips, and more, can affect pinpoint accuracy. In the end, the majority of glucose monitors have an error margin of 10%.

I compared my readings to a conventional finger-prick monitor during the course of the two weeks. I took a glucose reading as soon as I woke up, after my first meal, and after dinner. I discovered that the Stelo monitor hardly ever fell below or rose above the 20 percent margin overall. If anything, the one-time readings on my finger-prick monitor were more erroneous. Here are just some of the readings I compared.

I had several occasions where I ate a dinner high in carbohydrates and noticed that the CGM lagged behind the finger-prick monitor. However, the Stelo’s readings were only a few points off, demonstrating how accurate it is.

Consequently, I would rate my monitor as accurate and, in certain situations, reflecting the latency of my blood glucose’s migration to my interstitial fluid. If you’re concerned about the precision of your CGM, I advise wearing the Stelo while taking margin error, the glucose’s path of travel, and other outside variables into account.

minimal initial iteration.

It was easy for the over-patch that held the gadget to my arm to get dirty. A few days later, my first sensor started to hurt really bad, so I took it out. It wasn’t a problem with the new sensor I used. When I used the product for the second time, I also experienced a defunct CGM, despite applying it and waiting a full day to see if it would ever turn on. Ultimately, I gave up and switched to the other one.

The Stelo has a lot of potential for future iterations, especially with the AI coaching and its partnership with Oura, even though its initial functionality is as limited as that of the Lingo.

Both Lingo and Stelo make use of generative AI software, which offers tailored advice and insights based on each person’s requirements. However, the AI features stop there. Dexcom has ambitious plans to add more AI-powered features to the device, such as a potential new AI feature that lets you snap a photo of your food and get its nutritional data. Because tracking food in this way can be time-consuming, this feature could revolutionize dieting apps.

Additionally, the top-rated Apple Watch has been expertly tested.

Dexcom needs to develop Stelo’s generative AI feature to give users reliable, thorough, and helpful guidance, even though it sounds alluring. My weekly insights didn’t recommend adding exercise for a week.

Having used a CGM before, I knew what I needed to improve. But if a person only has a sensor for a month, they might not have enough time to comprehend their metabolism completely. More recommendations and changes from the software, like daily insights, would be fantastic.

ZDNET’s purchasing recommendations.

A pair of 30-day sensors for the Stelo by Dexcom costs $99 at retail. This gadget helps users manage their blood sugar levels more effectively by providing an intuitive user interface. People who suffer from obesity, prediabetes, or Type 2 diabetes (not on insulin) may find this device to be very helpful. Because it requires a lot of work, education, and time to prevent metabolic diseases like Type 2 diabetes, Stelo may be a helpful tool.

With the use of this technology, even dedicated athletes could optimize their performance to the fullest extent possible. Despite its $99 monthly price tag, the Stelo may help people gain control over their bodies. If you choose a smaller sensor, a faster method of recording events, and a simpler procedure for tracking your blood sugar, I suggest the Stelo over the Lingo.

scroll to top