An A1c tells you the average of the last three months. It doesn’t tell you about the overnight low the patient slept through, or the post-meal spike they never felt. Continuous data does, and it changes what you can actually adjust.
Remote patient monitoring for diabetes means collecting glucose data from a patient at home, from a cellular meter or a continuous glucose monitor or both, and putting it in front of a care team that reviews it monthly and acts on it. The clinical argument is that two patients with an identical A1c can have entirely different days.
One is steady. The other swings between overnight lows they sleep through and post-meal spikes they never feel, and averages out to the same number. Time in range distinguishes them; an A1c never will. That’s why time in range has largely replaced the single A1c as the number diabetes teams manage to.
The operational argument is simpler. Most of these patients also carry hypertension, and many carry a third condition, so the same panel supports RPM and CCM in the same month, and often MTM on top.
The reading nobody sees is the one that matters. Nocturnal hypoglycaemia is the classic example: the patient sleeps through it, a fingerstick regimen never catches it, and it changes the insulin plan the moment you can see it.
Glucose is the centre of it, but a diabetes programme that only tracks glucose leaves most of the risk unmanaged.
| Signal | Device | Common alert threshold | What the team does |
|---|---|---|---|
| Blood glucose | Cellular glucose meter | < 70 or > 250 mg/dL | Same-day contact on hypoglycaemia. Pattern review on repeated highs. |
| Continuous glucose | CGM | Time in range below 70%, or any nocturnal low | Regimen review. Nocturnal lows change the evening dose, not the daytime one. |
| Blood pressure | Cellular BP monitor | 7-day average above the patient’s goal | Comorbid in most of the panel and a major driver of downstream risk. |
| Weight | Cellular scale | Sustained trend in either direction | Tracks the lifestyle side and flags fluid changes where CHF coexists. |
| Medication adherence | Adherence sensor | Clustered missed doses | Separates “the regimen is wrong” from “the regimen isn’t being taken”. |
Alert thresholds are configured per patient by the ordering clinician and adjusted over time. The values above are common starting points, not clinical guidance, and nothing here establishes a standard of care. Reimbursement figures are approximate national non-facility Medicare averages and vary by locality, facility status, payer and calendar year: verify current rates at cms.gov or with your MAC before you bill.
Cellular meters for the whole panel, CGM data brought in for the insulin-treated group.
Results reach the care team seconds after the test, with no app and no pairing.
RPM 99454Full glucose curves, time-in-range reporting and overnight low alerts.
95249 · 95250For the hypertension that travels with most of this panel.
RPM 99454Step-on weights, no button to press.
RPM 99454These stack. The constraint is documenting the minutes separately, never counting them twice.
Meter supply plus monthly treatment-management time.
99453 · 99454 · 99457Diabetes plus a second chronic condition, which is the usual case.
99490 · 99439Pharmacist-led review of a regimen that is not producing control.
99605–99607When diabetes alone drives the care burden.
99424–99427Diabetes distress and depression, which materially affect adherence.
99484 · 99492Five steps, and only two of them involve the patient doing anything.
Identify diabetes patients by diagnosis code and payer, and flag who is already eligible for CCM alongside RPM.
A cellular meter for most. CGM where the patient is insulin-treated or the curve is what you need to see.
The patient tests the way they always have. Transmission is automatic; nothing gets written in a logbook.
Your team works time in range, nocturnal events and post-meal spikes rather than isolated numbers.
Time, consent and transmission days are documented as the work happens.
A well-built diabetes programme tracks five things:
Glucose alone is the most common programme design and the one that leaves the most risk unmanaged.
A glucometer gives a precise reading at a moment the patient chooses, which means it captures the moments they think to test. A CGM produces a continuous curve, which captures the overnight lows and post-meal spikes a fingerstick regimen would never see.
Many programmes use both: CGM for insulin-treated patients, cellular meters for the rest. They bill differently: meters under RPM device supply, CGM under 95249/95250.
Yes. Glucose data from a connected meter is billed under the RPM codes: 99453 for setup and education, 99454 for device supply with readings on at least 16 of 30 days, and 99457/99458 for monthly treatment-management time.
CGM services are billed separately under 95249 (patient-owned equipment) and 95250 (practice-supplied, minimum 72 hours of recording), with 95251 for interpretation. Verify current rates and payer policy before billing.
The widely used consensus target is more than 70% of readings between 70 and 180 mg/dL, with less than 4% below 70 and minimal time below 54. Targets are individualised, and older patients or those with hypoglycaemia unawareness are commonly managed to a looser range.
Time in range is more actionable than an A1c because it tells you when control is being lost, which is what determines whether you change the basal dose, the mealtime dose or the timing.
Yes, provided the time is tracked separately. CMS permits RPM treatment-management time (99457) and CCM care-management time (99490) in the same calendar month for the same patient, but the same minutes can never count toward both.
This is the most common source of denials in diabetes programmes run across two systems, because neither system can prove the minutes were distinct. One time log across all programmes is the clean answer.
Not for the cellular meter. It transmits over its own 4G connection with 2G fallback, so the patient tests exactly as they always have and the result reaches the care team without an app, a pairing step or a logbook.
CGM sensors vary by manufacturer, and some do use a phone as the reader. We confirm the specific device pathway during programme design.
Bring your patient count and payer mix. We’ll model the billable population across RPM, CCM and MTM before you commit.
Practical guidance on remote monitoring, devices and billing. Updated with every new post.
Compare RPM and CCM platforms on the five things that decide real-world performance, then implement an RPM program step by step with 2026 CPT rates.
Read articleCellular or Bluetooth for RPM blood pressure? What CPT 99445 changed in 2026, why 78% of over-65s owning a smartphone breaks the usual argument, and how to match devices per patient.
Read articleA measure-by-measure map of what RPM, CCM and APCM can actually influence, the CBP specification detail most programs miss, and why PDC targeting should start at 79%.
Read articleInterested in a demo or just general questions? Fill out the form below and a representative will respond shortly!
