CGM Interpretation — Quick Reference

Nutritional Therapists · Prepared for Spencer Martin (Syai Ultra) · 16 April 2026 · v1.0
How to use this page. A 1-2 page desk reference for reading a CGM report in a non-diabetic / metabolic-health context. Click a pill in the header to jump to a section, or use Expand all to print. Every number is referenced - follow the number in square brackets to the source in section 6. Companion to Spencer's existing material on meal logging, sensor artefacts and practitioner workflow.
🎯 1. Priority read order — why SD first, not TIR The hierarchy that surfaces early dysfunction

In non-diabetic populations, mean glucose is usually unremarkable. Variability and fasting regulation are where early dysfunction shows up. Read in this order:

  1. Standard Deviation (SD) - narrow corridor vs wide swings around the mean.
  2. Fasting baseline - overnight profile and dawn rise.
  3. Delta peak - post-meal excursion above pre-meal baseline.
  4. Coefficient of variation (CV) - normalised variability signal.
  5. Average sensor glucose - last to deteriorate, useful as sanity check.
📊 2. Non-diabetic reference values — Shah 2019 [1] n = 153 healthy adults, ages 7-80, 10-day Dexcom G6
Metric Non-diabetic reference Early-signal zone
Mean glucose5.4–5.5 mmol/L (5.8 if >60 y)Rising trend over months
TIR 3.9–7.8 mmol/L96% (IQR 93–98)<90% sustained
TAR >7.8 mmol/L2.1% (~30 min/day)>10% sustained
TBR <3.9 mmol/L1.1% (~15 min/day)>3% sustained
Within-subject CV17 ± 3%>25%
SD (derived from mean + CV)~0.9 mmol/L>1.2 mmol/L
Post-meal delta peak1.5–2.5 mmol/L above pre-meal; peak <7.8Delta >3.5 or absolute peak >8.5
Honest confidence note. Shah's N is modest (153), single-device (Dexcom G6), and sensor accuracy is weakest in the 4–7 mmol/L range. Treat 95% TIR 3.9–7.8 as a defensible working reference, not a regulatory standard.
🧭 3. Methodology note — the Battelino / Shah distinction The single most common error in NT-facing CGM material

The single most common error in NT-facing CGM material is applying Battelino's diabetes targets to non-diabetic clients.

  • Battelino 2019 [2] provides the metric definitions (TIR, TBR, TAR, GMI, CV) and the diabetes targets (TIR 3.9–10.0, CV <36%).
  • Shah 2019 [1] provides the non-diabetic reference values.
Use Battelino for how to read the report. Use Shah for what normal looks like. Mixing the two hides the early-signal zone between 17% CV (Shah) and 36% CV (Battelino diabetes).
🚨 4. Referral triggers — GP investigation Screening signals, not diagnoses
SignalThreshold
Fasting glucose≥ 6.1 mmol/L (IFG) or ≥ 7.0 mmol/L on two occasions (diabetic range)
Any sustained reading≥ 11.1 mmol/L
14-day mean with symptoms≥ 6.5 mmol/L
Persistent hypoglycaemiaTBR <3.9 mmol/L exceeding 3% with symptoms
Scope of practice. CGM in non-diabetic coaching is a screening signal, not a diagnosis. These thresholds trigger a GP conversation - they do not replace one.
🐦 5. Lactate — the canary in the coalmine (teaser) The future metric. Full framework in v1.1.

A client can have pristine CGM metrics - TIR >95%, CV 17%, flat overnight - and still have a poor lactate profile. The pancreas is coping. The engine underneath is not.

  • Fasting lactate can rise before fasting glucose in early metabolic dysfunction [3].
  • Mitochondrial dysfunction is detectable in lean, insulin-resistant individuals before any glucose abnormality [4]: ~30% lower mitochondrial ATP synthesis, ~80% higher intramyocellular lipid, with normal fasting glucose, A1c and BMI.
THE FUTURE METRIC

% time in normal lactate range - the lactate equivalent of TIR. As wearable continuous lactate matures [5], continuous glucose + continuous lactate becomes the next monitoring frontier.

GNL Grace is building the full lactate knowledge base and practical application now. v1.1 of this guide will add the operational lactate framework once the remaining five anchor papers land (Monnier 2017, Kelley 2002, Danne 2017, Zhang 2019, San-Millán & Brooks 2018).

📚 6. References 5 primary sources
[1] Shah VN, DuBose SN, Li Z, et al. Continuous Glucose Monitoring Profiles in Healthy Nondiabetic Participants: A Multicenter Prospective Study. J Clin Endocrinol Metab 2019;104(10):4356–4364. DOI: 10.1210/jc.2018-02763. PMID: 31127824.
[2] Battelino T, Danne T, Bergenstal RM, et al. Clinical Targets for Continuous Glucose Monitoring Data Interpretation: Recommendations From the International Consensus on Time in Range. Diabetes Care 2019;42(8):1593–1603. DOI: 10.2337/dci19-0028. PMID: 31177185.
[3] Broskey NT, Pories WJ, DeMaria EJ, et al. Fasting Plasma Lactate as a Possible Early Clinical Marker for Metabolic Disease Risk. Diabetes Metab Syndr 2024;18(2):102955. PMID: 38382370.
[4] Petersen KF, Dufour S, Befroy D, Garcia R, Shulman GI. Impaired Mitochondrial Activity in the Insulin-Resistant Offspring of Patients with Type 2 Diabetes. N Engl J Med 2004;350(7):664–671. PMID: 14960743.
[5] Guzzi J, Falter F, Kumar AB, Perrino AC. Mind the Gap: Wearable Lactate and Glucose Monitors for Hospitalized Patients. Cureus 2025;17(2):e78536. DOI: 10.7759/cureus.78536.