Standing in my bathroom, staring at the needle poking out of a continuous glucose monitor (CGM), I started having second thoughts.
CGMs measure glucose levels in interstitial fluid — the fluid between your cells. Unlike traditional fingerprick tests, they can give you real-time data about your blood sugar levels, and they’re typically used by people with Type 1 diabetes to manage their condition. But Nutrisense and other companies like Levels and January are proposing that nondiabetics can use blood sugar data to lose weight, optimize fitness gains, and improve their overall metabolic health. Of course, these startups still have a lot to prove. When The Verge looked into these devices last year, it was unclear whether CGMs were effective for people without diabetes, especially when it came to goals like weight loss.
But I happen to be the sort of person that companies like Nutrisense are trying to court. I don’t have diabetes, I eat well, and I exercise often. However, my family history and polycystic ovary syndrome (PCOS) diagnosis put me at a heightened risk of diabetes. My doctors say losing weight will help with the metabolic symptoms associated with PCOS. Fueling properly for races and workouts has always been a struggle for me. On top of real-time glucose monitoring, Nutrisense also pairs you with a dietician to make sense of your results. On paper, the math adds up to me potentially benefiting from CGM use — even if the evidence isn’t fully there yet.
Ultimately, I gritted my teeth and jammed a needle-bearing sensor into my arm.
How it works, setup, and removal
Because CGMs require a prescription, you have to take a questionnaire to see whether you’re eligible to use Nutrisense. If you are, a Nutrisense doctor will issue your prescription. After, you get sent a kit containing Abbott’s FreeStyle Libre CGM and two bandages. Separately, you download the Nutrisense app and then follow a series of helpful video tutorials that walk you through assembling the applicator and “installing” the CGM sensor on your body. When that’s done, you slap the bandage over the CGM and hover your phone near the sensor to take your first measurement. It uses NFC, so it’s the same motion as using your phone to pay for something at the store, just on the back of your arm.
There’s a helpful walkthrough in the Nutrisense app when you first get the CGM.
For the first day, the sensor needs to be calibrated. Afterward, you’re encouraged to log your meals and exercises so you can directly see how they impact your blood glucose levels. The sensor only retains about eight hours’ worth of information, so you’ll have to scan at least three times a day to ensure you don’t lose data. (This can be annoying, and I did lose a few hours here and there.) After 14 days, you replace the CGM and start all over again. For the first month, Nutrisense gives you complimentary access to a nutritionist who’ll review your data, point out trends, and answer any questions you might have. (After that first month, it’ll cost you $100 extra per month.)
Not me nervously laughing while reading the instructrions.
I hate needles. Needless to say, I was a little freaked out at the prospect of jamming this into the back of my arm.
This is easy to write out after the fact, but I was sweatin’ during the entire setup process. Unlike a noninvasive device like Lumen, which claims to measure your metabolism via your breath, CGMs pierce your skin with a small, thin needle. The CGM Nutrisense uses is no exception, and I hate needles the way Indiana Jones hates snakes. (I’m frankly not sure how I sat through three tattoos without fainting.) It doesn’t help that the applicator makes loud noises as you assemble it and contains a visible needle.
So I was shocked when I truly felt nothing as the needle went in. The sound of the applicator was more frightening than the actual motion of jamming the applicator into my arm. Both the covid-19 jab and the flu shot are exponentially more painful. This is likely because you insert the CGM into the fattiest part of your underarm, whereas a vaccine goes straight into muscle.
Our photographer documented my entire emotional journey of wearing the CGM.
I was most nervous about placing the CGM, but it was actually harder to remove at the end of my 14 days. There were no official instructions for this part, so I just... pulled it off. What made it hard was that the bandage Nutrisense gives you is quite strong, and I had trouble ripping it off my skin. Afterward, I had adhesive residue on my arm for two days even though I scrubbed real hard in the shower. You can also see trace amounts of dried blood on the used CGM, and there was a small puncture mark on my arm. It healed within a day, but Nutrisense recommends rotating arms to give your skin a couple of weeks to recover.
All about glucose
Using Nutrisense is similar to food logging. You’re supposed to scan the sensor a few times a day and log all your food and exercise. Scanning brings up a squiggly line of your blood sugar levels, and your meals / workouts are represented by red dots on that line. Other factors, like sleep, are shown as blue dots. Nutrisense interprets your data in two buckets. First, there’s your glucose data, and second, there’s your nutritional data. For you to get the most out of it, you really do have to be diligent about logging your meals.
I hate long-term food tracking. While keeping an eye on macros can be helpful and insightful for training and weight loss, no app I’ve ever tested does a good job of making this convenient. Nutrisense isn’t an exception, though it’s one of the better implementations I’ve seen in recent years. On top of searching a database or scanning barcodes like other calorie-tracking apps, you can write out a description or snap a picture of your meal. Nutrisense’s AI will then populate your log with those items. It’s not perfect — the AI will get ingredients wrong, and you’ll have to fix it manually. But it gives you a starting place. I was also impressed with the support for international foods. I could log gimbap, tteokbokki, and Indian takeout with relative ease. That isn’t always the case.
In the second shot, you can see my glucose spike after a run where I pushed myself.
Scanning and logging are the easy parts. Interpreting the data, however, comes with a learning curve. For blood glucose, there’s the self-explanatory daily average and a more complicated Daily Glucose Score. The latter comprises your highest daily glucose value, average value, adaptability (time spent above your glucose threshold), and variability (how intensely your glucose “swings”). I wish that were it, but there’s more. In the app’s Insights tab, you’ll find other values, like how long you spent within the ideal range, variability, average, min, max, sleep average, and morning average.
Each metric is color-graded. Green is good, orange is okay, and red indicates areas of improvement. According to Nutrisense, you want to stay within 70 to 140 milligrams per deciliter 95 to 100 percent of the time. You also want your variability to stay under 14mg/dL and for your morning and sleeping glucose averages to be between 70 and 105mg/dL.
Meanwhile, everything you eat is given an overall Meal Score that’s made up of four metrics: peak, exposure, stability, and recovery. Peak is your highest glucose value within two hours of eating; exposure is how strong your glucose response was; stability measures the highest versus lowest values within a two-hour eating window; and recovery is how close you get to your premeal levels after two hours. Nutrisense uses the same color-based grading system here, too.
You just hold your phone up to the CGM to sync data. Feels kind of like contactless payments.
Nutrisense has a crap ton of educational material within the app to explain all of this in further detail. I went through about two hours of it and barely scratched the surface. At the end of the day, I relied mainly on the colored grading scheme and would read descriptions to remind myself what each submetric was and why it mattered.
Ultimately, I didn’t learn a ton of new things about my blood sugar and eating habits, but I appreciated having visual confirmation. I know that I get cranky and sleepy if my blood sugar is low, but it’s vindicating to see the connection represented on a graph. I don’t need a CGM to tell me that tteokbokki — a dish that’s 90 percent rice cake — will spike my glucose levels, so I should probably have a protein along with it. It was more interesting to see how the same meal eaten at different times of day could induce different responses. For instance, my nutritionist pointed out that a helping of butter chicken curry produced a less intense spike when I ate it for lunch versus a late-night dinner. Perhaps the most fascinating thing to me was how my glucose skyrocketed right after a hard run — which makes sense because your body releases glycogen to fuel that kind of exercise. Conversely, a strength training session registered as a relatively flat line.
Nutrisense encourages experiments like these. That could be a lot of fun! Especially if your knowledge of nutritional science can be summed up as “low-carb diets are popular for weight loss.” But if you’re already familiar with all of this, it’s less exciting. More of a “huh, would you look at that” than an “oh wow, this is blowing my mind!”
I’m smiling, but I’m really thinking, “Holy shitballs, it actually didn’t hurt?!”
Case in point, my two weeks with Nutrisense indicated I was already doing everything right — which was then confirmed by my nutritionist. My average glucose score was 99. They told me my baseline glucose trends were well within the optimum range of 70–90mg/dL, my highest spike never breached 140mg/dL, and I had “excellent glycemic variability” with an average of 11mg/dL. That was nice to hear considering I ate worse and exercised less than usual in those two weeks. Still, it was neat to have someone interpret the data and be on hand to answer questions.
But the bottom line was there wasn’t much for me to improve. My glucose minimum was the only metric in orange because I skipped a few meals, and I had none in the red zone. If I weren’t a reviewer, my two-week experiment would’ve cost me several hundred dollars to learn I’m already managing alright.
Accessibility is the biggest issue
My biggest problem with Nutrisense is that it takes a certain income level just to try it out.
My eyes nearly popped out of their sockets like the Tex Avery wolf when I saw how much Nutrisense costs. It’s $399 for the no-commitment one-month plan, $299 monthly for the three-month plan, $250 monthly for the six-month plan, and $225 monthly for a 12-month plan. As mentioned earlier, all plans get you a free month of nutritionist support, but you’ll have to tack on $100 monthly after that. There’s no option for a free trial, and once your plan ends, it’ll automatically renew on a month-to-month basis unless you pause or cancel. At a minimum, you’re looking at $2,700 annually on the 12-month plan or $3,800 if you shell out for the nutritionist support. Personally speaking, I wouldn’t have been able to afford this experience if I weren’t a reviewer — there’s no free trial for most people.
That’s a lotta moola, though the bulk of the cost is the CGM itself. According to Nutrisense’s FAQ page, the CGM it uses is not covered by any insurance companies, though you can use an HSA or FSA to help with the cost. To be fair, other CGM startups are similarly priced, but that simply underscores that, at the moment, this is an incredibly niche product.
Nutrisense’s CGM program could be educational, but the price of entry is too steep for the average person.
From my own experience as a wearables reviewer and someone with PCOS, that’s so frustrating. Behavioral change is hard, losing weight is hard, and managing chronic conditions is really hard. We won’t have a better idea if CGM programs like Nutrisense are effective for nondiabetics if they aren’t used by a wider population. But realistically speaking, I don’t think the average person would pay this much and scan multiple times a day for several weeks or months without at least some expectation of capital-R results. What’s more likely is all that effort, time, and money helps you learn a little bit more about yourself, and you may see some results over an extended timeframe. Even then, there’s no guarantee it will give you the solution you’re looking for. Is that worth the price of entry?
If you have the means to try Nutrisense, I’m not going to stop you. Emerging tech is, by definition, unproven, but sometimes you just don’t question why something works for you. I firmly believe it’s your right to figure out what that thing is. But as with Lumen, I would say you should go in with tempered expectations. I was hoping to find out something I didn’t already know or had overlooked. What I got was an underwhelming but sobering reminder that there’s no magic bullet for chronic conditions.
Agree to Continue: Nutrisense CGM program
Every smart device now requires you to agree to a series of terms and conditions before you can use it — contracts that no one actually reads. It’s impossible for us to read and analyze every single one of these agreements. But we started counting exactly how many times you have to hit “agree” to use devices when we review them, since these are agreements most people don’t read and definitely can’t negotiate.
By setting up Nutrisense, you’re agreeing to:
Final tally: three mandatory policies and several permissions
If you want to delete your data after using Nutrisense, you can apply to do so here.