I try to stay healthy. I eat a (mostly) vegetarian diet, prioritize daily movement and try to make time for sunshine. I also get annual blood work done that measures things like cholesterol, vitamin levels, thyroid health and glucose. That’s because I’m generally aware of the US problem with chronic diseases, including Type 2 diabetes and prediabetes, which affect millions of adults and can be prevented.
I note all of this to set the stage for how arrogant I was before I started testing a Stelo continuous glucose monitor, the first available without prescription in the US. Previously, CGMs have only been available as medical devices for people who use insulin and are required to track their blood sugar for medical purposes, along with a small number of wellness biohackers who’ve paid top dollar to get a CGM prescription off-label so they can better see how their body uses energy (food).
Given glucose, or blood sugar, may be the last basic health metric we can’t track with a conventional smartwatch, I didn’t think twice before stamping the quarter-sized biosensor to my upper arm and wading into deeper wellness waters, expecting the assurance from past bloodwork and the fact that I read about health stuff all day to shield me from the likelihood of anything scary popping up in this new wellness terrain. Plus, my Oura ring says my cardiovascular age is several years younger than my real age.
In other words, I got cocky.
Not only was I wrong about expecting glucose fluctuations to be straightforward — there is frustratingly little information on what continuous glucose levels should look like in people without diabetes, and mine seemed significantly different from the first week of my trial run to the last couple of weeks — I also realized I may’ve been underestimating the importance of other daily habits that have impacts on my health, like chronically skimping on sleep, sitting down for too long and not taking time to manage my stress. While I understood these things can affect my health, they were never truly visible to me until I allowed a CGM to get under my skin.
Here’s my experience wearing the Stelo and what you should expect when tracking your own glucose 24/7.
Sticking on a Stelo and Dexcom CGM basics
The Stelo glucose monitor is made by Dexcom, one of the two major diabetes care companies and maker of the Dexcom G6 and G7, which measure blood sugar continuously like the Stelo but require a prescription. (The other big diabetes company is Abbott, which announced the US availability of its own prescription-free CGM shortly after Dexcom had cleared its own.)
The Stelo was designed for people with Type 2 diabetes, but any adult who doesn’t use insulin and wants more insight into their blood sugar levels can use it. The thinking goes that spending more time in lower blood sugar zones will reduce the risk of health effects from high blood sugar — this is especially true for people with Type 2 diabetes, who have a higher baseline. As this is intended for informational purposes and isn’t the same as prescription devices, you shouldn’t make any medication decisions based on readings without consulting your doctor first.
Like other CGMs, the Stelo monitor doesn’t track your blood’s glucose through actual blood. Instead, blood sugar is read through a tiny thread-like needle that draws interstitial fluid, which surrounds the cells. This tiny needle is stuck to the back of a small sensor that sticks to the arm with adhesive tape.
I’d never used a CGM before, but I was happy to learn the Stelo instructions were easy to follow. The first step you need to take is download the Stelo app, which is compatible with iPhone or Android. Then, create an account.
One important thing to note is that you’ll need to select a “target range” for your glucose measurements to stay inside, based on whether or not you’ve been diagnosed with diabetes or prediabetes. The target range for Type 2 diabetes goes up to 180 milligrams per deciliter and appears to be the standard range doctors use for looking at glucose across all populations. People withoutType 2 diabetes or prediabetes will be given a lower target range to stay beneath during most of the day at 140 mg/dL.
Once you have the app and an account, you can attach your sensor by following the app’s instructions.
The biosensor attaches to the back of the arm, preferably to an area with less muscle. After choosing the fleshiest part of my upper arm and disinfecting it with an alcohol wipe, per the instructions, I unscrewed the applicator’s cap and pushed the button to eject the sensor into my arm. It was virtually painless, definitely a “less than one” on a 1 to 10 pain scale, but the sound it makes upon ejection is surprisingly loud.
The Stelo monitor is sold in a two-pack box with two sensors in separate applicators. Each sensor lasts up to 15 days and a month’s supply costs $99. If you want a renewing subscription, it costs $89 per month. Insurance companies won’t cover CGMs for people who don’t need insulin in most cases, but you can use your FSA/HSA dollars.
While it’s definitely not cheap at $99 for a month’s supply of sensors, the Stelo is more affordable than other consumer CGM companies that use a workaround for the prescription requirement. These companies do so by getting their own doctors to write the prescriptions and using Dexcom and Abbott’s sensors for diabetics. Levels, for example, costs $199 for a month’s supply of sensors plus a separate $199 yearly fee for the app. Nutrisense costs $225 to $299 a month.
Abbott’s general wellness glucose monitor, Lingo, costs $89 for a month’s supply.
Hopefully, more availability of glucose monitoring in the wellness space will drive down prices and even influence insurance coverage so people with higher-than-normal blood sugar levels, or those with Type 2 diabetes who don’t need insulin, can afford to keep wearing a CGM like the Stelo.
Day 1: Getting comfortable with the miserable task of food tracking
I started using the Stelo monitor in the middle of a work day and was quickly met with my first head-scratcher blood sugar event: eating a vegan bowl of brown rice, grilled vegetables and vegan dressing shot my blood sugar up pretty high. I was made to know exactly what I ate because the Stelo app has an “events” tab where you can log meal descriptions (with carb counts in grams, if you want to), exercise or other activities and notes.
Stress, illness, hard exercise and even lack of sleep can also affect blood sugar and raise your levels.
At risk of sounding dramatic, I find the idea of logging the carbs in a salad dressing preposterous. I knew I couldn’t last a month doing it, so I didn’t. I instead logged my meals with the main ingredients. For example, one “meal” in the Stelo app is logged as a “black bean, cheese and vegetable empanada with a side salad of arugula and ranch.” Other “meals” are actually more like snacks (“popcorn with butter, parmesan and nutritional yeast,” “banana,” “carrots and hummus”) or drinks (“coffee with half and half and agave).”
I will note here that I eventually quit logging meals altogether because as I got more used to checking the app periodically throughout the day, narrowing down a “culprit” food or stressor causing a little bump or spike in my glucose graph became easier.
This isn’t to say you shouldn’t log any of your meals in the Stelo app — doing so in the early days really helped me get in the habit of associating certain foods with higher or lower glucose levels so I could eventually go “off book.” Making sure you note your meal is especially helpful if you get a “spike detected” prompt because the app will only notify you of significant blood sugar jumps. There does seem to be a slight delay in these notifications, though.
A salad tests my patience
It was the end of a work week and I was hurrying to finish an article that had a deadline while also eating a late lunch I thought was pretty healthy: a salad made of a bed of arugula, sauteed peppers and onions, seasoned black beans, cheddar cheese and croutons. I also had a small black bean empanada and put salad dressing on the salad. And I know croutons are mostly carbs and little substance but I like adding crunch to salads and still consider the meal overall to be fairly nutrient-dense and colorful.
I was surprised, then, to check the Stelo app and see a high spike in glucose. It was definitely above 140 and looking down at the graph doubled my stress. It became a chicken-or-the-egg situation — was my glucose spiking because I was stressed and tired and then ate a carb-y salad? Or was I feeling extra tired and stressed because I went a long time without eating and then ate a carb-y salad and my blood sugar is now crashing down?
Either way, life went on, I had plans with a friend to see a movie later that night and was then hit with a different curveball: The two glasses of red wine and buttered popcorn that I consumed almost in spite, assuming my biosensor was going to scream at me, didn’t raise my blood sugar in a noticeable way.
I later learned that alcohol actually lowers blood sugar and that not eating enough protein, even when there are plenty of fruits and vegetables involved, causes my glucose levels to go up. Still, the first few days of tracking my blood sugar served to tell me that foods that are full of nutrients and otherwise good for me may also cause higher spikes in glucose than foods that I feel confident aren’t doing me any health favors (wine or alcohol).
Bagels and blood sugar
This was the most dramatic (and, I thought, concerning) spike in my blood sugar since wearing the Stelo. Shortly before this, I had a Pilates workout and then ate a veggie omelet on a bagel with a beet, ginger and carrot juice. I’ve eaten a couple of other bagels since this incident in late August but the effect hasn’t even come close to this.
There’s a good bagel shop down the street from where I live in Brooklyn. One Saturday after a Pilates workout, I ordered a veggie omelet on a salt bagel with a beet, carrot and ginger juice.
In other words, it was a carb-heavy meal. And while it was delicious, I knew Big Daddy Blood Sugar was going to have something to say about it. And oh, did it. This remains my biggest glucose spike since I began tracking my blood sugar, and it was quite shocking to see how high it got, even though it fell back down fast.
This one wasn’t surprising — I know bagels have a lot of carbs, and while the juice was freshly blended and full of vitamins, it was processed juice and a lot of natural sugar. But it was around this point that I started to get genuinely concerned that I may be one of the 1 in 3 US adults who have prediabetes but don’t yet know it.
The Stelo starts to mellow out
A couple weeks into testing the Stelo, my glucose charts started becoming a little less spikey and my average glucose level is trending down. I’m able to do this in the glucose summary in the Stelo app, as well as a more comprehensive view in Dexcom Clarity, an app people can use to share their glucose information with a health care provider. My glucose levels have also been lower in the mornings when I wake up, suggesting a slightly lower fasting glucose level (or morning glucose level).
I think a few factors contributed to this. One is that I’ve been trying to add more protein to more snacks and meals, especially as the first food choice of the day when I tend to not be very hungry. Protein does not have the same glucose-spiking effect as carbs. I’ve also been trying to sit around less after eating, as walking or moving following a meal is a fast way to lower glucose.
But I haven’t been trying that hard to lower my glucose levels as I wanted to get a more accurate snapshot based on what my norm is. This leads me to other potential influences that are more speculative. One is that I think I was getting over some type of virus when I first started wearing the monitor. I felt fatigued enough to test for COVID a couple of times (negative), but my tiredness persisted for several days after a quick sore throat, suggesting my body was fighting something and (potentially) raising my glucose levels. At first, I didn’t think much of this because the information I’d seen suggested that more severe illnesses raised blood sugar, not minor bouts of fatigue.
The other factor that may have lowered my fasting glucose level is that I was a bit less stressed and may’ve been getting more restful sleep. For a big chunk of my Stelo review, I was spending time with family in Wisconsin. During this time, I actually expected my blood sugar levels to spike a bit more than normal considering I was eating desserts more often than I usually do and also exercising a bit less. But coming out of a stressful couple of weeks in New York City full of travel and poor sleep, could working in the peace and quiet and spending time with family improved my stress levels in ways that may have inadvertently lowered my blood sugar?
More days over the last two weeks wearing the Stelo have been more aligned with its “target” range for someone without diabetes. This is one of the days I wasn’t necessarily “eating healthier” than normal — I was visiting family and actually eating more desserts than normal — suggesting that blood sugar is affected by more than specific food choices.
What the doctors and research say
Much of my concern and confusion throughout this whole process comes to how often my glucose levels should exceed 140 mg/dL, which is the target limit for people without Type 2 diabetes. While the past couple of weeks I’ve been more aligned with the suggested goal of spending 96% of time in the “green,” I definitely can’t say it’s been that way from the jump.
Basically, I’ve been trying to find out whether I should be worried about insulin resistance, which is when the body doesn’t respond as well to the blood sugar regulating hormone. If left unchecked, insulin resistance leads to Type 2 diabetes, which is associated with other health problems.
Dexcom says that they selected their under 140 mg/dL based on research showing that most people without diabetes should stay below 140 mg/dL the vast majority of the day. The US Centers for Disease Control and Prevention says people without Type 2 diabetes should be under 140 mg/dL 2 hours after the glucose tolerance test.
There’s also newer research getting into the nitty-gritty of what glucose spikes and dips should look like in non-diabetic people as a way to optimize health. One small study from 2018, for example, found there are differences in the way people without diagnosable diabetes process blood sugar, suggesting there may be different “glucotypes” with higher risk of health problems, even if they don’t meet the medical criteria for prediabetes or diabetes.
But according to Dr. Scott Isaacs, a board-certified endocrinologist and president-elect of the American Association of Clinical Endocrinology (endocrinologists are specialists in hormone-related bodily functions such as diabetes and blood sugar control), the evidence for a specific “target range” in those without diabetes may not be solid yet. Hearing my stress about going above 140 mg/dL, he pushed back on the idea that it’s not healthy for people without diabetes or prediabetes to rise above 140, or even up to 200.
“When it spikes up like that, that can be completely normal.” He told a story about being at a conference with other endocrinologists a couple of years ago when CGMs were still new and many doctors were testing one out.
“Everybody was wearing one, and everybody was freaking out,” he said, particularly when blood sugar spiked after lunch.
“It happens,” Isaacs said. “If you eat a certain meal, it’s gonna go up to 200.” While those spikes should come down quickly, it’s one metric that may be “misleading” about wearing a CGM, Isaacs said.
“You can eat something and get a spike that goes really high on your CGM, and then it comes right back down,” he explained. At the same time, another person could eat a higher-fat food that’s not necessarily healthier, “but the sugar never spiked.”
Dr. Thomas Grace, head of clinical advocacy and outcomes at Dexcom, took a look at my Stelo graphs through the Dexcom Clarity app (which anyone with a Dexcom CGM can download and use to share their information with a doctor) and told me my swings didn’t elicit any worry. But he acknowledged there’s a research gap in CGMs being worn more generally, and that those spikes can be used as a learning experience.
“There’s still some work being done in this field,” Grace said. “Before the advent of having continuous glucose readings, we had a lot of assumptions that numbers should run in certain areas, and so we’re very early into this space, and there’s work currently being done to define that.”
Grace also called my attention back to Stelo’s real purpose: It’s aiming to help people with Type 2 diabetes manage and have agency over their own blood sugar levels so they feel better, or potentially reverse their need for medication or more medical interventions.
“That’s what we’re looking to do, is kind of give people gentle nudges to help them with what we know is one of the most impactful things on their diabetes, which is their food choices,” Grace said.
“Even though people have been told to do those things for years, it’s really eye opening to see that you can see the direct causation of things happening,” Grace said. “And I think that there’s such a huge benefit there for people to be able to see these things.”
According to 2024 information from The Commonwealth Fund, the US ranks last compared to other wealthy countries such as Australia, the UK and Canada, when it comes to weighing health care performance with the amount of money spent. This leaves the door wide open for more innovation in preventive care, which can stop diseases from progressing and help cut costs.
A1C vs. CGM: Is one more accurate than the other?
One of the benefits to wearing a CGM, according to Isaacs, is that it may be more telling than an A1C test drawn in a doctor’s office. An A1C is an average of your blood glucose levels over the last three months. It’s used for diagnosing diabetes and prediabetes, but it may be affected by small variabilities or abnormalities in red blood cells.
And it can’t track blood sugar in real time like wearing a CGM can, so people who have significant lows and significant highs may have a misleading A1C because it’s just averaging out, according to Isaacs.
“The CGM helps you figure out sort of more precisely what’s going on compared to the A1C,” Isaacs explained, specifically through a metric called a GMI. I could see mine once I downloaded Dexcom Clarity, which was easy to do and something I recommend because it filled out some of the long-term data missing from the Stelo app.
But learning this was concerning to me, as I was leaning on my last couple of years’ worth of A1C numbers to feel confident I was firmly below a cut-off that would suggest diabetes, prediabetes or insulin resistance. But my GMI estimate from Dexcom is definitely higher than my past two A1Cs. (My GMI has decreased a bit over time since I started using the sensor, as my average glucose levels have been lower, but still.)
I have nothing to do with this information now other than speculatively sift through other blood work findings. I did find out I had a vitamin B12 deficiency in 2023 — could this affect my blood results or be a factor? I have zero idea. But I absolutely plan to bring it up at my next doctor appointment.
If I were to wear a Stelo again, I’d invest in some type of extra patch like Skin Grip or GrifGrips, which I learned from Reddit and Taking Control of Your Diabetes may help keep your CGM attached. This photo was taken when there were four days left of the sensor. Glucose readings were still going strong and remained unaffected, though.
What are the benefits and risks of tracking your blood sugar?
Continuous glucose monitoring gives people direct insight into their blood sugar levels, which can indicate diabetes, a preventable disease that can lead to nerve damage, vision problems, heart problems and more.
Without glucose monitoring, detection of diabetes relies on people getting regular blood work or having diabetes progress enough to where people actually develop symptoms. These can include increased thirst and tingling in the hands and feet.
In addition to being a real-time alternative to traditional bloodwork, Isaacs says glucose monitoring is beneficial for people who want more information about how their diet and other lifestyle factors affect blood glucose. So long as they understand the limitations of wearing a CGM, which includes not overreacting to what could be “possibly normal.” (Guilty.)
“On the other hand, I’m sure there are people that have diabetes that don’t know it, and they put out a sensor and it shows their diabetes,” Isaacs said. “It can go both ways,” he said, but CGMs do tend to motivate people to take health into their own hands.
From a public health standpoint, more people wearing glucose monitors will encourage more research and help inform what it means to be “metabolically healthy” — a goal post I still don’t feel confident is sturdy enough at the end of writing this review.
What tracking my blood sugar with the Stelo taught me
I learned there are still a lot of questions we don’t have answers for when tracking glucose in the general population. For example, how meaningful are the differences in terms of long-term health or overall wellness between people with varying glucose levels or spikes? Official health guidance around blood sugar seems to revolve around “disease” vs. “no disease.”
On the other hand, will glucoses monitoring monitoring give people a false sense of health security? What if their blood sugar stays low, but their diet is generally lacking in nutrients that tend to lead to positive long-term health outcomes?
On a personal level, and despite my early frustrations with the ups and downs, I did learn a lot from a month spent tracking my glucose. (I did miss a few days when switching sensors.) Mostly, I learned that making very small changes like prioritizing eating just a little bit more protein each day, and taking a short little walk after a meal does seem to keep my glucose from rising high.
Importantly, none of these changes required me to take on habits I find unsustainable or even harmful in the long run — such as calorie counting, applying moral values to food as “bad” in order to restrict them or over-exercising.
It also highlighted unhealthy habits of mine I’ve been sweeping under the rug. For example, I am a chronically poor sleeper who will stay up very late or get up very early in the name of getting things done, usually more creative or fun things I don’t have time for in the regular weekday. But because my poor sleep hygiene hasn’t been affecting my health in obvious ways, and I feel rested enough most days, I’ve been treating it more as the ghost of Christmas future that will only haunt a future version of myself. And by then, I’ll have a healthier sleep pattern, right? This is even with the knowledge that research has directly pointed to short sleeping as a potential cause of insulin resistance in women.
On a follow-up question about stress, Isaacs told me that stress can make blood sugar higher, but there are factors that affect how much. He added that the impact is more pronounced in people with diabetes. Whether or not my poor sleep and perceived high stress in my Stelo testing was a real reason for the subtle positive glucose shift from start to finish, I’ve decided it doesn’t matter much. Because if there are two patterns that do seem to solidify time and time again in the wellness world, it’s that healthy habits inspire other ones, and that you feel a greater sense of well-being when you have agency and choice over them.
