SSMT Claim Register
SSMT-2026-0009v12026-07-26 Type O · Outcome Wearables & Recovery draft
Claim, as marketed
“Fuel smarter. See how your body responds in real time.”
Canonical claim
Continuous glucose monitoring in athletes without diabetes improves fuelling decisions and thereby performance.
Scope
Athletes without diabetes. The type-1 diabetes evidence base is deeper and does not transfer — different population, different decision.
Verdict
Unevaluated Bordering Partially supported for the narrower claim that CGM tracks glucose trends during exercise.
No independent evidence located as of the card date.
Claim drift — 4 of 7
–Population–Setting▲Comparator▲Outcome▲Magnitude⋯Reproducibility▲Independence
Population and Setting are aligned — validation was done in the right people, actually exercising. Independence splits: the accuracy work is 4/5 independent, but the optimistic interpretive layer clusters in vendor-affiliated authorship.
Evidence located
5
exercise-accuracy studies
4/5
independent
0
trials with a performance endpoint
0
position stands from nutrition bodies
Findings

Accuracy fails the standard. Regulatory expectation for CGM is around ≤10% MARD. In non-diabetic athletes during exercise, independent studies report 13.8% to 34% — with 41% in hand cycling and 24.3% post-exercise. One well-designed validation concluded the devices are 'not valid for point glucose monitoring' and arguably valid only for trends. The only study reporting under 15% during exercise is a three-page brief report with three vendor-consultant authors.

The lag is not a fixed offset. Interstitial-to-blood delay runs 5–25 minutes and is directionally asymmetric — after carbohydrate intake interstitial glucose lags, whereas during exercise it appears to change faster than blood. It cannot be corrected with a single constant.

There is no validated target range. Independent authors state repeatedly that no consensus exists on how to interpret these measurements in athletes. Elite athletes routinely excurse above 140 and below 70 mg/dL without apparent performance detriment. The core mechanistic objection: CGM does not measure muscle glycogen or carbohydrate flux, which limits its use as a fuel sensor.

Tested once as a decision tool. A randomised crossover (n=12) compared CGM-informed refuelling against fixed 15-minute interval dosing over 75 minutes of cycling. Glucose variability fell (CV 8.82% vs 11.95%, p=0.013) but there was no difference in heart rate, RPE or blood lactate — and no performance endpoint was measured. Notably the CGM arm consumed 18.1 g carbohydrate per hour versus 51.2 g/h, i.e. CGM guidance drove intake well below conventional fuelling guidance.

Governance. The UCI banned in-competition use of devices capturing metabolic values in 2021, with medical derogations. Independent authors have also named 'glucorexia' — athletes becoming hypersensitive to normal, non-relevant glucose fluctuations.

What would change this verdict
A randomised trial in non-diabetic athletes, n≥40, with a real performance endpoint (time trial, mean power, time to exhaustion), comparing CGM-guided fuelling against a competent gram-and-clock protocol at recommended intake rates — with CGM winning by more than the smallest worthwhile change. Plus independent replication showing exercise MARD at or below 10–12% with per-intensity reporting.
Practitioner read
During exercise a CGM on a non-diabetic athlete is off by roughly 14–34% against a real blood measurement, and the reading reflects your blood 5–25 minutes ago — it is a trend line, not a live fuel gauge. Nobody has established what number an athlete should aim for, and no study has shown that steering fuelling by the sensor makes you faster; the one trial that tried found no physiological benefit and had the CGM group eating a third of the carbohydrate the control group did. Use a grams-per-hour plan and a stopwatch.
Could not be verified
Listed rather than dropped. Counts are floors, not censuses.
Sources
  1. Weijer V, et al. (2024). Eur J Sport Sci — elite para cyclists ↗
  2. Matzka M, et al. (2024). Eur J Appl Physiol 124(12):3557–3569 ↗
  3. Bauhaus H, et al. (2023). IJERPH — trained athletes ↗
  4. Poon ET-C, et al. (2025). JISSN — the one decision-tool trial, n=12 ↗
  5. Bowler AM, et al. (2023). Int J Sport Nutr Exerc Metab 33(2):121–132 ↗
  6. Flockhart M, Larsen FJ (2023). Sports Med — interpretation review ↗
Every source must resolve at a DOI or PubMed ID.
Changelog
Cite this card
SSMT. "Continuous glucose monitoring in athletes without diabetes improves fuelling decisions and thereby performance." Claim SSMT-2026-0009 v1, graded 2026-07-26. https://sportssciencetech.com/c/SSMT-2026-0009
Disagree with this grade? Every card carries “what would change this verdict” so the specification is explicit. Send evidence and it will be logged and the card re-versioned in public — including when it moves a grade in a manufacturer's favour. Submit evidence →