Continuous Glucose Monitors for Athletes: Insight or Expensive Anxiety?
Glucose sensors are marketed to athletes as a fuelling gauge, but no trial shows they improve performance in people without diabetes, readings lag and drift during exercise, and normal glucose swings are easy to misread as a problem.
Sprint Summary
The short version — read this if you're short on time.
Continuous glucose monitors, small sensors worn on the arm that read glucose every few minutes, were built for people with diabetes and are now marketed to athletes as a fuelling gauge. The evidence doesn't yet support that use. No randomised trial has shown that CGM-guided fuelling improves endurance performance in people without diabetes, and sports nutrition reviews describe the idea as untested. The sensors are reasonable at showing trends, but in studies of runners and cyclists only about half of readings during exercise were within 10% of a finger-prick test, and errors grew during exercise and after sugary food.
Normal glucose also moves more than many people expect. Healthy adults without diabetes rise after meals and dip briefly every day, and sensors can overestimate highs. That makes it easy to read normal physiology as a problem and to start cutting carbohydrate, which can leave you under-fuelled. If you're curious, treat the data as a pattern, not a verdict, keep fuelling to evidence-based guidance and practice, and stop if it's making you anxious about food. If you have diabetes, your clinical team's plan comes first.
If you have diabetes, a continuous glucose monitor is a medical tool and your clinical team's plan for exercise, fuelling and insulin comes first. Nothing in this article replaces their advice.
This article deliberately gives no glucose targets or ranges to aim for, and doesn't suggest fasting, low-carbohydrate eating or cutting carbohydrate to flatten a glucose trace. Normal rises after meals and during hard exercise aren't a sign that you should eat less. If glucose tracking is making you anxious about food or pushing you to restrict, stop using it and speak to your GP or a registered sports dietitian; TriForward's "Under-Fuelled, Not Undertrained" covers low energy availability.
Don't use a sensor reading to decide whether symptoms are real. Shakiness, sweating, confusion or feeling faint during or after exercise mean stop, eat or drink something with sugar and get help if it doesn't settle quickly, whatever the sensor says. Repeated episodes, or very high readings confirmed by a finger-prick test, are a reason to see your GP rather than self-diagnose.
Sensors sit under the skin. Watch for redness, swelling, pain or discharge at the site, and check any race or event rules about wearing medical-style devices before relying on one in competition.
Full Distance
The complete research and analysis.
You've probably seen them: a small white disc on the back of a professional cyclist's arm, or a friend's phone showing a glucose curve after a long ride. Continuous glucose monitors (CGMs) were developed for people with diabetes, for whom they've transformed day-to-day management. Now they're sold to healthy athletes with promises of knowing exactly when to fuel.
The appeal is obvious. Fuelling is one of the hardest parts of long-course triathlon to get right, and a live number feels like it should take the guesswork out. This guide looks at what the evidence says about CGMs in athletes without diabetes: how accurate they are during exercise, what normal glucose looks like, whether they help performance, and the real risk that the data does more harm than good.
TriForward's "90–120g of Carbs an Hour" and "Gut Training" cover how much to fuel and how to train your gut to cope. This piece is about whether a sensor can tell you more.
What a CGM actually measures
A CGM has a fine filament that sits just under the skin, usually on the back of the upper arm. It measures glucose in the interstitial fluid, the fluid between cells, not in the blood, and sends a reading every one to fifteen minutes to a phone or watch.
That matters for three reasons:
- There's a lag. Glucose moves from blood into interstitial fluid with a delay, so the sensor trails behind your blood when glucose is changing quickly, such as after a gel or during hard efforts.
- The relationship can shift during exercise. Blood flow, temperature and muscle activity near the sensor all change during exercise.
- Most devices aren't calibrated by the user. The algorithm converts an electrical signal to a glucose number using assumptions built for typical users.
Accuracy is usually reported as the mean absolute relative difference (MARD): the average percentage gap between the sensor and a reference blood test. Lower is better. A MARD of around 10% is typical for modern sensors at rest.
What the reviews say
Two recent reviews by sports scientists set the scene.
A 2023 review by Bowler and colleagues in the International Journal of Sport Nutrition and Exercise Metabolism noted that manufacturers of athlete-targeted devices market them as a "fuelling gauge". The authors concluded that:
- Whether CGM-derived interstitial glucose has any meaning for fuelling or performance in sport is untested.
- Whether CGMs will provide reliable and accurate information and improve sports nutrition practice is also untested.
- Practical problems include sensors dislodging, data loss when syncing fails, cost and possible bans in some sporting settings.
- Interpreting the data needs context, which is where sports nutrition professionals come in.
A 2024 current-opinion paper by Flockhart and Larsen in Sports Medicine reached a similar view. There is no consensus on how to interpret CGM data in athletes with normal glucose control, and well-defined approaches to use glucose monitoring to improve endurance athletes' performance and health are lacking. They also noted that glucose regulation in endurance athletes can differ from that in non-athletes, and is affected by hard sessions and by high or low energy and carbohydrate availability.
Neither paper is a trial. But when experts in sports nutrition say the central use case is untested, that's a meaningful starting point.
How accurate are CGMs during exercise?
Recreational runners on a treadmill (2024)
Mason, Hartwig and Greene, in a 2024 paper in the International Journal of Sports Physiology and Performance, tested a commercial CGM (a model on sale in 2024) worn on the upper arm and the chest in 13 well-trained recreational runners without diabetes (average age 40). They ran on a treadmill for 30, 60 and 90 minutes at increasing steady intensities, and sensor readings were compared with finger-prick blood glucose.
- Average error (MARD) was 13.8% on the arm and 11.4% on the chest.
- Only about 47% (arm) and 50% (chest) of readings were within 10% of the finger-prick value.
- Variation was large: the coefficient of variation for the error exceeded 70%.
- Accuracy was worse at lower glucose levels, which is when an athlete would most want to know.
- Trends in average glucose over time were tracked well.
The authors concluded that CGMs are not valid for point glucose monitoring during exercise but appear valid for trends during steady running. The sex split isn't reported in the abstract.
Para cyclists over two weeks (2024)
A 2024 observational study by Weijer and colleagues in the European Journal of Sport Science gave 13 para cyclists (eight hand cyclists and five cyclists) a CGM for two weeks, with finger-prick checks at rest and during a standard training session. The abstract doesn't give the sex split.
- At rest, MARD was 12%.
- During exercise, MARD rose to 34%, and was 41% in hand cyclists, whose sensor sat close to working arm muscles, versus 24% in cyclists.
Para athletes differ from most triathletes, but the message about sensors near active muscle is relevant to anyone who swims, a sport that works the upper arm hard.
Healthy adults and food (2025)
A 2025 randomised crossover trial by Hutchins and colleagues in The American Journal of Clinical Nutrition had 15 healthy adults (9 women, 6 men) drink or eat seven different carbohydrate challenges while glucose was measured by CGM and by capillary blood sampling every 15 minutes.
- The CGM read about 0.9 mmol/L higher than capillary samples, both fasting and after eating.
- The size of the error differed between people and between foods.
- The CGM overestimated time spent at high glucose about four-fold; adjusting for each person's baseline difference reduced this to about two-fold.
Over-the-counter sensors (2026)
The newest study here, by Prakash and colleagues in the Journal of Diabetes Science and Technology (2026, online ahead of print), was the first independent test of over-the-counter CGMs sold directly to consumers without diabetes. Thirty-nine young adults wore them during rest or exercise, with a sugary or placebo drink.
- Overall MARD was 12.9%, with about two-thirds of readings meeting a standard accuracy criterion.
- Agreement was worse after sugar (15.3%) than after placebo (10.6%), and worse when glucose was changing quickly.
- Exercise had no consistent effect in this study.
- Readings lagged by about 6 minutes.
- About 14% of the trend arrows showing which way glucose was heading were classified as moderate to extreme risk of misleading the user.
Inference: across these studies, current sensors are reasonable at showing the shape of a day or a session, but a single reading during a race, or right after a gel, could easily be 10 to 30% out, and slightly behind.
What does normal glucose look like?
A large part of the risk with CGMs is misreading normal physiology.
Healthy people without diabetes (2019)
A 2019 multicentre study by Shah and colleagues in The Journal of Clinical Endocrinology & Metabolism gave CGMs to 153 healthy, non-obese participants aged 7 to 80 for up to ten days.
- Average glucose was 98 to 99 mg/dL (5.4 to 5.5 mmol/L) in all age groups except those over 60 (104 mg/dL).
- Healthy people spent a little time above and below the thresholds used in diabetes care every day: a median of about 30 minutes above and 15 minutes below.
These were not athletes and not exercising hard. But they show that brief "highs" and "lows" on a sensor are part of normal life, not a diagnosis.
Elite endurance athletes on a controlled diet (2025)
Bowler and colleagues followed up in a 2025 paper in the Journal of Diabetes Science and Technology. Twelve elite racewalkers (7 men, 5 women; average age 22) completed two four-day blocks of standardised training on a high-energy, high-carbohydrate diet while wearing CGMs.
- Day-to-day glucose variability was similar to that reported in healthy non-athletes, despite hard daily training and a high-carbohydrate diet.
- Average glucose was higher across 24 hours than overnight, and tended to be slightly lower in the women.
Athletes who looked "abnormal" (2016)
A 2016 study by Thomas and colleagues in the Journal of Diabetes Science and Technology fitted CGMs to 10 sub-elite athletes for about six days of normal life. The researchers used a narrow glucose band to define where they expected values to sit.
- Four of the ten spent more than 70% of the time above that band, even excluding two hours after meals.
- Three of the ten had fasting values in a range that would be labelled prediabetes by a common definition.
- Only one spent much time low, and that athlete was eating well below recommended energy intake.
This is a small, older study using an earlier sensor generation, and a narrow band chosen by the researchers. It illustrates how easily athletes' normal values can be read as a problem, and why Flockhart and Larsen argue that athletes' glucose patterns need their own interpretation. It does not show that these athletes had a health problem.
Does CGM-guided fuelling improve performance?
No randomised trial found for this guide has tested whether athletes without diabetes who fuel according to a CGM perform better than those who follow standard guidance. Both reviews above describe the question as open.
There are also physiological reasons to be cautious:
- Glucose often rises during exercise. In the para cyclists, high readings were seen almost only after meals and during exercise. Inference: a high reading mid-session doesn't mean you've eaten enough.
- Blood glucose isn't the same as the fuel stored in your muscles. Inference: a normal reading late in a long race doesn't tell you how much muscle glycogen is left, and fatigue has many causes. TriForward's "Why Watts Fade: It's Not Just an Empty Fuel Tank" covers the mechanisms.
- A lagging, inaccurate reading during a fast-changing effort is a weak basis for a fuelling decision.
Inference: a CGM may show that a pre-race breakfast or a carbohydrate drink raises your glucose, which you'd expect anyway. Whether that information changes a decision for the better is unknown.
The expensive-anxiety problem
There's no published study of anxiety from CGM use in athletes. But two patterns from adjacent evidence are worth noting, and both are inference here:
- Normal rises look alarming on a graph. Post-meal peaks and exercise-driven highs are normal, and sensors may overestimate them. Some people respond by cutting carbohydrate, which can undermine training and race fuelling.
- More numbers can mean more worry. TriForward's "Tracking Everything: Does More Data Help or Hurt Motivation?" covers evidence that heavy tracking can reduce enjoyment and, in some people, is linked to compulsive exercise and disordered eating.
For athletes, the cost of under-fuelling is well established. TriForward's "Under-Fuelled, Not Undertrained" covers low energy availability, which can affect health, bone and performance in both women and men.
Practical application
None of this has been tested as a package in triathletes. It's built from the studies above and is inference.
If you're thinking about buying one
- Ask what decision it would change. If the answer is "I'd eat less when the line goes up", the evidence doesn't support that and it carries real risk.
- Keep your fuelling plan based on evidence and practice. Carbohydrate guidance for long sessions and races comes from performance trials, not glucose traces.
- Skip it if tracking food or numbers already makes you anxious.
If you use one anyway
- Look at patterns, not single readings. Trends were tracked reasonably well in runners; single values during exercise often weren't.
- Expect lag. A reading taken soon after a gel or a hard effort reflects where glucose was several minutes ago.
- Don't compare yourself with diabetes targets. These were designed for people with diabetes. Healthy people and athletes cross them briefly every day.
- Check a surprising reading with a finger-prick meter before acting on it, and see your GP if repeated checks are worrying.
- Mind sensor placement and fit if you swim, since the error was larger with the sensor close to working arm muscle in para cyclists.
A triathlon scenario
The following example is illustrative, not drawn from a study. A 44-year-old age-grouper training for a full-distance race buys a CGM after seeing professionals wear them. On his first long ride, his glucose climbs during the hill efforts despite not eating, then dips slightly late in the ride after a gel. Over breakfast the next day, he sees a sharp peak after porridge and toast and starts wondering whether to cut carbohydrate.
Using the evidence above:
- The hill-climb rise matches the rises seen during exercise in studies, and isn't a sign he's well fuelled.
- The late dip is within the error range of sensors during exercise, and the reading lags several minutes behind his blood.
- The breakfast peak is normal in healthy people, and sensors may overstate peaks.
- He keeps his fuelling plan, based on what he's practised and on carbohydrate guidance, and treats the sensor as a curiosity for the rest of its two-week life.
Common mistakes
- Treating a single reading as precise. During exercise, about half of readings were more than 10% out in runners.
- Fuelling only when the number drops. By then you may be behind, and the reading lags anyway.
- Cutting carbohydrate to flatten the line. That can leave you under-fuelled for training and racing.
- Comparing yourself with diabetes thresholds or with a friend's trace.
- Using the sensor to self-diagnose diabetes or "pre-diabetes". That needs a proper medical test.
- Ignoring how you feel. Symptoms come first (see the safety notes).
What we don't know
- Whether CGM-guided fuelling improves performance in triathletes or any athletes without diabetes. No trial was found.
- How accurate current sensors are in long races, in the heat, in the water or after hours of exercise.
- What "normal" looks like for endurance athletes in daily life. Reference data are limited to small groups such as 12 racewalkers on a controlled diet.
- How often CGMs lead athletes to restrict food, and who is most at risk.
- Women and masters athletes. Most validation samples are small and few report results by sex or age.
References
- Bowler AM, Whitfield J, Marshall L, Coffey VG, Burke LM, Cox GR. The use of continuous glucose monitors in sport: possible applications and considerations. International Journal of Sport Nutrition and Exercise Metabolism, 2023;33(2):121–132. doi:10.1123/ijsnem.2022-0139
- Flockhart M, Larsen FJ. Continuous glucose monitoring in endurance athletes: interpretation and relevance of measurements for improving performance and health. Sports Medicine, 2024;54(2):247–255. doi:10.1007/s40279-023-01910-4
- Mason LJ, Hartwig T, Greene D. Validating the use of continuous glucose monitors with nondiabetic recreational runners. International Journal of Sports Physiology and Performance, 2024;19(11):1307–1313. doi:10.1123/ijspp.2024-0102
- Weijer V, van der Werf R, van der Haijden M, Jeukendrup A, van Loon LJC, van Dijk JW. Continuous glucose monitoring in para cyclists: an observational study. European Journal of Sport Science, 2024;24(12):1809–1819. doi:10.1002/ejsc.12220
- Hutchins KM, Betts JA, Thompson D, Hengist A, Gonzalez JT. Continuous glucose monitor overestimates glycemia, with the magnitude of bias varying by postprandial test and individual: a randomized crossover trial. The American Journal of Clinical Nutrition, 2025;121(5):1025–1034. doi:10.1016/j.ajcnut.2025.02.024
- Prakash N, Barrows AH, Stults-Kolehmainen M, Loza AJ, Kilpatrick MW, Ash GI. Agreement of over-the-counter continuous glucose monitors with capillary-meter glucose in young adults without diabetes. Journal of Diabetes Science and Technology, 2026 (online ahead of print). doi:10.1177/19322968261482150
- Shah VN, DuBose SN, Li Z, Beck RW, Peters AL, Weinstock RS, Kruger D, Tansey M, Sparling D, Woerner S, Vendrame F, Bergenstal R, Tamborlane WV, Watson SE, Sherr J. Continuous glucose monitoring profiles in healthy nondiabetic participants: a multicenter prospective study. The Journal of Clinical Endocrinology & Metabolism, 2019;104(10):4356–4364. doi:10.1210/jc.2018-02763
- Bowler AM, Burke LM, Coffey VG, Cox GR. Day-to-day glycemic variability using continuous glucose monitors in endurance athletes. Journal of Diabetes Science and Technology, 2025;19(6):1538–1545. doi:10.1177/19322968241250355
- Thomas F, Pretty CG, Desaive T, Chase JG. Blood glucose levels of subelite athletes during 6 days of free living. Journal of Diabetes Science and Technology, 2016;10(6):1335–1343. doi:10.1177/1932296816648344
Frequently asked questions
Will a glucose monitor help me fuel better in races?
There's no trial evidence that it does in athletes without diabetes. Reviews describe CGM-guided fuelling as untested, and the readings lag behind blood glucose and become less accurate during exercise. Evidence-based carbohydrate guidance and practising your fuelling in training remain the better-supported approach.
How accurate are glucose sensors during exercise?
Good enough for trends, not for single readings. In 13 recreational runners, only about half of sensor readings during steady running were within 10% of finger-prick values, and accuracy was worse at lower glucose. In para cyclists, the average error rose from 12% at rest to 34% during exercise.
My glucose spikes after meals and during hard sessions. Is that a problem?
Usually not. Healthy people without diabetes have rises after meals and brief highs and lows every day, and athletes' readings often rise during exercise. One study also found sensors read higher than finger-prick samples, overestimating time at high glucose about four-fold. If you're worried, see your GP for a proper test rather than relying on the sensor.
Who might find a CGM useful?
People with diabetes, under their clinical team, and researchers or sports dietitians who use them to look at patterns alongside other information. For a healthy age-group triathlete it's an optional curiosity, and best avoided if tracking food or numbers tends to make you anxious.
The evidence behind this guide
1 graded principle, 9 sources. Open one to see the sources.
Continuous glucose monitors show glucose trends reasonably but are imprecise during exercise, and no trial shows they improve fuelling or performance in athletes without diabetesLimited evidence9 sources
In practice: For healthy triathletes, a CGM is at most a way to view glucose patterns, not a fuelling gauge: fuel to evidence-based carbohydrate guidance and practice, don't act on single readings during exercise, don't compare yourself with diabetes thresholds, and never cut carbohydrate to flatten a trace. People with diabetes should follow their clinical team. Benefits for performance are unproven and the advice here is inference from small validation studies.
Review of CGMs in athletes: athlete-targeted devices are marketed as a 'fuelling gauge', but whether interstitial glucose has meaning for fuelling or performance, and whether CGMs give reliable and accurate information in sport, is untested. Caveats include sensor dislodging, data loss, cost, possible bans and the need for professional interpretation.
Current opinion: no consensus on how to interpret CGM data in athletes with normal glucose control, and well-defined approaches to use glucose monitoring to improve endurance performance or health are lacking. Endurance athletes' glucose regulation can differ from non-athletes' and is affected by demanding training and by energy and carbohydrate availability.
During 30-90 min of steady treadmill running, CGM error versus finger-prick glucose (MARD) was 13.8% on the arm and 11.4% on the chest; only about 47% and 50% of readings were within 10%, variability was large and accuracy was worse at low glucose. Trends in mean glucose were tracked well: valid for trends, not point values.
High readings occurred almost only after meals and during exercise. CGM error versus capillary glucose (MARD) was 12% at rest but 34% during exercise, 41% in hand cyclists (sensor near working arm muscle) versus 24% in cyclists. Para athletes; small sample.
CGM read about 0.9 mmol/L higher than capillary samples fasting and after eating; bias varied by food and by individual; CGM overestimated time at high glucose about 4-fold, about 2-fold after adjusting for baseline differences. Healthy non-athletes at rest.
First independent test of over-the-counter CGMs: MARD 12.9%, 66.5% of readings within 15%/15 mg/dL; worse after sugar (15.3%) than placebo (10.6%) and during rapid change; exercise had no consistent effect; about 6-minute lag; 13.8% of trend arrows classed as moderate to extreme risk. Online ahead of print.
Mean glucose 98-99 mg/dL in all age groups except over-60s (104 mg/dL). Healthy participants spent a median of about 30 min a day above and 15 min a day below the thresholds used in diabetes care. Normative data; not athletes, not exercise-specific.
On a high-energy, high-carbohydrate diet with hard daily training, day-to-day glycaemic variability was similar to values reported in healthy non-athletes; 24-hour mean glucose was higher than overnight and tended to be lower in women than men. Small elite sample under controlled conditions.
Over about 6 days, 4 of 10 athletes spent more than 70% of monitoring time above a narrow researcher-chosen glucose band even excluding post-meal periods, and 3 of 10 had fasting values in a prediabetes range by a common definition; only 1 spent much time low, with low energy intake. Small, older sensor generation; shows how athletes' values can be read as abnormal.
Get The Forward
Research-led triathlon guidance, straight to your inbox. No spam, no filler — unsubscribe any time.
