Continuous glucose monitoring training: CGM vs SMBG, time in range, and clinic workflow

Primary clinical educator

Speaker 1 in Otter — clinical diabetes educator / GP-facing trainer. A colleague later addresses him as Jeff (surname not stated). Wore a sensor during the session; Tesla / heads-up-display teaching analogies.

Device / practice training

Tanya (Speaker 9) — Dexcom / AMSL Gold Coast representative; applicator demo, app alerts, Clarity platform, costing, and practice support. Speaker 14 — company colleague (self-described midwife background); accuracy (MARD), replacement policy, G7 vs One Plus app distinction.

Audience / panel

Multiple numbered Otter speakers (practice GPs and guests) — Q&A on funding, accuracy vs venous glucose, MRI/security, pacemaker, UK app region lock, live dinner CGM readings.

Session coverage (important)

Otter id g0Qh4f35W-qMoBrXmpljJPi9EzQ · title Continuous Glucose Monitoring Training · Otter session length shown as 2 hr 14 min, but the available transcript ends around 1:29:30 (~67% coverage; Otter transcription limit). This page summarises only the Otter-available portion (~first 90 minutes), not the full 2h14 recording. Do not invent the missing hour. Export date 2026-09-16 · visible date “Sep 8 at 1:25 am” on Otter.

Key messages for GPs

• SMBG (fingerstick) is a point-in-time snapshot; diaries can be incomplete or “too clean.” HbA1c is usually three-monthly and can mislead in anaemia, pregnancy, and extremes of age.
• CGM samples interstitial fluid via a tiny filament — not a deep venous needle — and streams a near-continuous profile to a compatible phone app (with optional clinic Clarity link).
• Teaching frame used repeatedly: CGM is like driving with a heads-up speed display — awareness changes behaviour around diet, activity, and meds (“therapeutic inertia” / “mental constipation”).
• Time in range (TIR) — often discussed with a 3.9–10 mmol/L band and a rough ~70% target for many adults — plus time high / time low pie views; tighten further in pregnancy / GDM contexts as taught.
• Practice path: put a sensor on → set a 10-day goal (often “live normally”) → review, don’t set-and-forget → adjust insulin / orals / GLP-1 / lifestyle with the AGP-style Clarity view.
• Australia: type 1 CGM commonly funded via NDSS pathways as discussed; type 2 largely self-funded at the time of the talk (DVA gold/white card with diabetes listed called out as an exception). Costs cited for single sensors / three-packs / monthly subscription.

Limits of SMBG and HbA1c

The educator opened with clinic reality: average diabetes consult time ~10–12 minutes, with high-turnover days seeing large volumes. Against that clock, self-monitoring blood glucose (SMBG / “old style glucometers”) fails on compliance — finger pricks, excuses, incomplete profiles — while HbA1c, though helpful, is intermittent (Medicare typically every three months), does not show glycaemic shape, and is less reliable with anaemia or pregnancy. Medication adherence is also hard to infer if monitoring itself is sparse.

A teaching slide equated approximate average glucose bands to HbA1c targets (e.g. ~7% mapping roughly to ~6–10 mmol/L profile language in the talk). The warning: a “good” 7% can still hide significant hypoglycaemia — especially in pregnancy underestimation and in older people — because a single clinic reading or a tidy diary is still only a point (or a curated set of points) in the day.

What CGM is (and is not)

CGM (continuous glucose monitoring) places a small sensor through the skin to read interstitial fluid. It is not a deep blood draw. Patients need a compatible phone; clinic software can be linked so the practice sees profiles between visits. The educator urged clinicians to wear one themselves before advising — car-salesman / Tesla analogy: you sell what you understand.

Glucose management was framed as revolving around nutrition, physical activity (stairs vs lift, post-lunch walks — distinct from formal “exercise”), and medication. CGM makes hidden sugars (e.g. cereal labels) and meal choices visible; self-realisation after seeing post-meal spikes was described as common. Fitness / weight-loss programs overseas often use CGM; in Australia the talk stressed following TGA / guideline framing — primarily type 1 and type 2 diabetes, with clinician use also for recurrent hypoglycaemia and “iffy” pregnancy sugars.

Time in range and targets

In the Dexcom app, TIR (time in range) — Otter often heard as “TIL/TIO” — is the share of time inside a set band (example given: 3.9–10 mmol/L). Out-of-range time splits into hyper- and hypoglycaemia; pie charts summarise the mix. A rough adult target of ~70% TIR was quoted, with context: a truck driver’s highs are higher-stakes than an office worker’s; aged-care / dementia patients may need fewer lows even if highs are slightly more tolerated — but prolonged values above ~15 mmol/L can be symptomatic.

Pregnancy needs tighter control; type 2 / gestational diabetes was described as needing more time in range than type 1 pregnancy allowances in the slide set. Wear-time matters: using CGM most of the time (talk cited >¾ of the time) associated with better HbA1c in presented trial summaries (DIAMOND / MOBILE-style naming in the ASR). Stopping CGM after a trial period was linked with HbA1c rising again — “remove the speedometer.”

Case vignettes: Alfred, Jonathan, Maria

Alfred — 91, MDI, hypo anxiety, poor dexterity

Multiple daily insulin injections; fear of hypos; arthritis limiting fingersticks; son nearby who wants to help. CGM (Dexcom) led to insulin de-escalation in the vignette (basal reduced, e.g. 40→32 units; rapid reduced) with improved TIR and fewer dangerous lows. Alternative raised: swap to a weekly basal insulin option (product names ASR-garbled; “Glutec” / “bougie” — confirm locally when available). Message: in the very elderly, hypo avoidance can matter more than chasing a perfect number.

Jonathan — 62, type 2, heart failure history

Unusual oral mix (simvastatin still listed — audience groaned); suspected post-dinner hyperglycaemia on CGM. Plan: stop two orals, start an SGLT2 (jar / “Himalayas Africa” ASR) while continuing a HF-appropriate agent; diet improvement and weight loss; HbA1c 7.9→6.2 and weight down in the story (e.g. BMI to 27).

Maria — 38, Pacific Islander, BMI 34

Metformin + lifestyle; high CV risk framing. SGLT2 + activity/diet left TIR still low (~68% at 12 weeks with ongoing hyperglycaemia). Escalate: replace SGLT2 with a GLP-1 (semaglutide / tirzepatide named) — semaglutide also framed for CV risk reduction licence. On 1.5 mg path in the vignette, TIR rose to ~90%.

Dexcom One Plus: apply, alerts, Follow, Clarity

Tanya demonstrated a three-step applicator: clean site → unscrew → push down → press button. Sites: upper arm (often preferred for comfort/accuracy in the talk) or abdomen (useful for bag/strap-heavy work). Filament remains under skin after needle retracts. Usability claim: patients >65 applied sensor + app in ~12.6 minutes from box+book; educators ~7 minutes — train patients to self-apply.

App features emphasised: current glucose, trend arrows (including double-up / double-down for rapid change), high/low alerts with vibration (Apple Watch mentioned), optional delayed first-high so brief post-meal rises do not alarm if they settle, diary tags for meals/meds, optional fingerstick calibration only if the reading “feels wrong” (factory calibration otherwise). Follow app: up to ~10 followers for hypo alerts — carers, adult children, farmers living alone, truck drivers (Siri voice query while driving cited). Sensor wear: 10 days + ~12-hour grace; waterproof for usual swimming/showering per Q&A.

Data access named three ways: phone Dexcom One Plus app, Clarity app/website for clinic linking, and patient-brought reports. Clarity AGP / overlay / daily / compare reports show patterns (e.g. “every Monday is bad — Indian takeaway?”). Patients must grant access via a Clarity clinic code. Critical: download the One Plus app, not G7 — devices do not cross-connect. Country-specific apps: UK sensors/apps are not interchangeable with Australian storefronts without region change.

Clinic workflow and nurse role

Start with easier wins: insulin users, newly diagnosed (avoid early fingerstick aversion), GDM where remote titration is attractive. Practice nurses: identify suitable patients aligned with the GP plan, set up Clarity on a nurse PC, teach application with a demo sensor, troubleshoot. Pre-visit review of Clarity data can shrink consult time — a COVID-era Paris GDM remote-clinic analogy was shared (login + phone titration in minutes vs waiting room).

Intermittent CGM before each HbA1c review (e.g. three 10-day wears across ~9 months using a three-pack) was offered as a cheaper pattern than continuous subscription for some type 2 patients. “Any visibility is good visibility.”

Australian access, cost, NDSS / RACGP

Slides referenced NDSS and RACGP type 2 guidance; international guidelines were framed as treating CGM as gold-standard when available. Funding caveat repeated: in Australia at the time of the talk, type 1 funded / type 2 generally self-funded; private health funds said not to cover type 2 CGM; DVA gold/white card with diabetes listed described as subsidised. Trial sensors via a practice “GADA”-sounding program (ASR) were offered so patients can experience benefit before paying.

Evidence snippets shown: HbA1c reductions in CGM trials including older adults (DIAMOND / MOBILE naming); assumption that “60+ won’t use tech” challenged. Continuous use outperformed sparse use in presented graphs.

Q&A pearls from the room

• Accuracy: interstitial lag vs venous; when in doubt / first hypo symptoms, still fingerstick. Rough ±5–10% variability spoken; site (muscle vs poor interstitial) matters; sensor flags poor placement.
• Sampling: measures frequently (30-second language) and pushes a reading about every 5 minutes — not a Libre-style on-demand scan model.
• MRI / big scanners can affect the sensor; walk-through metal detectors described as OK in personal travel anecdote. Pacemaker: speakers said no interference concern in-session.
• Replacement: if sensor fails early despite correct use / overpatch, company replacement free of charge per rep (even day 9).
• Injections nearby: keep other SC injections away from the sensor site (≈ three-inch rule mentioned).
• Bluetooth range ~10 m — keep phone nearby.
• Pipeline / MARD accuracy research encouraged when comparing brands; cheap non-validated “watch” gadgets dismissed in audience discussion.

allcpd.drkotha.com · directory.drkotha.com