Continuous Glucose Monitoring as a Safeguard

Continuous glucose monitoring as a safeguard: the alarm does the job the person cannot

Hypoglycaemia has one early warning system and it is the person’s own sensation. Shaky. Sweaty. Strange. A person who cannot produce that sentence has no warning system, and the first thing anybody notices is usually the stage where their brain is already affected.

A continuous glucose monitor does something no other intervention in this library does. It supplies the missing warning system from outside the body.

That is why CGM is not a convenience item for people with communication or cognitive disability. It is a safeguard — arguably the most effective single safeguard available for anybody on insulin who cannot report symptoms. And the subsidy that makes it affordable is tied to a diagnosis that many of those people do not have.

We support NDIS participants across New South Wales, including people using glucose monitoring technology under clinical plans from their treating teams. We are not a diabetes service, and nothing below is medical advice. Eligibility, prescription and clinical use are matters for the person’s diabetes team, and the National Diabetes Services Scheme Helpline is free on 1800 637 700.

What CGM actually is

A small sensor sits under the skin, usually on the upper arm or abdomen, and measures glucose in the interstitial fluid — the fluid between cells — continuously. A transmitter sends readings to a reader, a phone or a pump, typically every few minutes. The sensor is replaced every one to two weeks depending on the product.

There are two broad categories, and the difference matters enormously for this population:

Real-time CGM streams readings continuously and alarms when glucose goes below or above set thresholds, or when it is falling fast. Some systems predict a low and alarm before it arrives.

Flash or intermittently scanned monitoring stores readings and displays them when the reader or phone is scanned over the sensor. Depending on the product and configuration, alarm functionality may be more limited or absent.

For somebody who can feel their own symptoms, either is a large improvement on finger pricks. For somebody who cannot, the alarm is the entire point, and the distinction between a system that alarms and one that must be scanned is the difference between a safeguard and a record.

Why this matters more here than anywhere

Six reasons, in rough order of importance.

One: it replaces the warning system. An alarm at 4.0 falling does what the person’s adrenaline response cannot. For somebody with impaired awareness of hypoglycaemia — which, in a person who cannot self-report, is undetectable by history and therefore has to be assumed possible — this is the only reliable detection method.

Two: it shows direction, not just position. A finger prick gives a number at a moment. A CGM gives a number and a trend arrow. A reading of 5.5 holding steady and a reading of 5.5 falling rapidly are completely different situations, and only one of them needs acting on now. This is the single most useful operational feature for support workers.

Three: it finds the lows nobody would ever know about. Nocturnal hypoglycaemia in a house with no overnight testing is invisible, and somebody can have it repeatedly for months — showing up only as night sweats, morning headaches, unexplained fatigue, or high morning readings that are actually a rebound. CGM makes those visible, and the difference matters because the instinctive response to a high morning reading is more insulin, which makes an overnight low worse.

Four: it removes most of the finger pricks. Four tests a day is around 1,460 lancet punctures a year. A sensor changed fortnightly is around 26 applications. For a person who finds finger pricks painful or frightening — and for whom resistance to them may be heading toward a physical restraint nobody has authorised — that is not a lifestyle improvement. It is the removal of a daily distressing procedure.

Five: it gives the diabetes team data they can actually use. CGM produces an ambulatory glucose profile and a time in range figure — the proportion of time glucose sat within target. That is a far better basis for adjusting a regimen than a handful of scattered finger pricks, and for somebody whose glucose has been unstable with no obvious explanation, it frequently reveals the pattern within two weeks.

Six: it can be followed remotely. Most systems allow data to be shared with another person’s phone. In supported accommodation that means a family member, an on-call nurse or a team leader can see what is happening. For a person living alone with drop-in support, it means somebody can notice a problem without being in the room.

That last point needs a caveat about consent and privacy, below.

The subsidy, and the gap it leaves

In Australia, CGM products are subsidised through the National Diabetes Services Scheme, and the eligibility is specific.

Per the NDSS, subsidised CGM is available to people living with type 1 diabetes, including children and young people under 21, adults 21 and over, and women who are actively planning pregnancy, pregnant, or immediately post-pregnancy. Children and young people under 21 with conditions similar to type 1 diabetes who require insulin may also apply.

Access requires certification by an authorised health professional — an endocrinologist, a credentialled diabetes educator, or a physician or paediatrician specialising in diabetes. GPs and practice nurses cannot certify CGM access forms, which is a practical obstacle in itself for somebody whose diabetes is managed primarily in general practice. Credentialled diabetes educators cannot certify forms for young people under the “other conditions” category.

The form can be completed through the NDSS Health Professional Portal, where access is often confirmed within minutes, or downloaded, certified and submitted. Eligibility and product arrangements change, so confirm current details with the NDSS Helpline on 1800 637 700.

Now the gap.

Consider an adult with moderate intellectual disability, non-speaking, living in supported accommodation, with insulin-treated type 2 diabetes — which is a common picture, particularly where antipsychotic medication contributed to the diabetes developing. They cannot recognise or report a low. Their hypos present as agitation. They have been treated for behaviour at least once when they were actually hypoglycaemic.

Clinically, this person has about as strong a case for alarm-enabled continuous glucose monitoring as anybody in the country.

They are an adult with type 2 diabetes, so the type 1 subsidy pathway does not apply to them.

That is a real gap and it is worth naming plainly, because it is the kind of thing that gets discovered at the point of asking rather than in advance. The subsidy is organised around a diagnosis. The clinical need here is organised around a communication impairment, and the two do not line up.

What can actually be done about it:

• Confirm current eligibility properly rather than assuming, via the NDSS Helpline, because arrangements do change and the “similar conditions” category exists for under-21s.
• Ask the endocrinologist to make the case in writing. Where there is documented severe or unrecognised hypoglycaemia in a person who cannot self-report, a specialist letter setting out the clinical risk is worth having on file regardless of the outcome — and it is the document any future application or appeal rests on.
• Ask about the NDIS, while expecting the answer to be no. The NDIS does not fund diabetes treatment, and glucose monitoring technology is health-system and NDSS territory. The honest position is that this is not an NDIS support, and a request framed as diabetes management will be declined. Where some families have had more success is framing the need around the disability rather than the diabetes — the person’s inability to communicate symptoms being the disability-related functional impairment that creates the risk. We would not promise that this works, and we would not want anybody to rely on it as a plan.
• Price the private option, which some families do. It is an ongoing consumable cost rather than a one-off.
• Push the clinical alternative. If CGM is genuinely unavailable, the diabetes team should be asked what the alternative risk mitigation is — a higher glucose target, a less hypo-prone regimen, more frequent testing including overnight, or a different insulin. “No CGM” should not mean “no change.”

The practical problems, and what works

CGM in supported accommodation has specific operational failure modes. These are the ones we see.

The person removes the sensor. Common, and the responses in order: placement where it is less noticeable or reachable — the back of the upper arm rather than the abdomen; an over-patch or adhesive overlay, which the manufacturers supply and which genuinely helps; a fabric arm band or sleeve over it; and giving the person something to do with the sensor change so it becomes a routine they participate in rather than something done to them.

Where somebody persistently removes it, the question of whether to physically prevent that arises — and an arm restraint or a device the person cannot remove would be mechanical restraint, one of the five regulated restrictive practices requiring authorisation, a behaviour support plan and reporting to the NDIS Quality and Safeguards Commission. A simple sleeve that the person can take off is a different proposition from a device designed to stop them. If it is heading in that direction, involve the behaviour support practitioner before it becomes practice.

Adhesive reactions. Genuinely common and genuinely uncomfortable. Barrier films, different placement sites, and rotation all help, and the diabetes team can advise. A person with a painful itchy rash under a sensor has a good reason to pull it off, and that should be investigated before it is treated as behaviour.

Nobody responds to the alarm. This is the most consequential failure. A CGM is only a safeguard if a trained person is within hearing, has the phone or reader, knows what the alarm means, and knows what to do. In practice: the phone is in another room, or the alarm has been silenced because it was going off too often, or the worker on shift was not shown how the system works, or the phone that was paired belongs to a worker who is on leave.

What this needs is a written protocol, which almost nobody has:

• Which device receives the alarm, and who holds it on each shift
• Who is set up as a follower, and whether that is reliable overnight
• What each alarm threshold means and what the response is, in numbers
• That alarms must not be silenced or disabled without the diabetes team’s agreement
• What happens at shift handover — the device, the charge, the pairing
• A charging routine, because a flat phone is a disabled safeguard
• What happens if the sensor fails mid-wear, including reverting to finger pricks

Alarm fatigue is a real problem and it has a real solution. If alarms are going off constantly, the thresholds are probably set wrong or the regimen needs adjusting — and the answer is a conversation with the diabetes team, not turning the alarm off. An alarm that is ignored is worse than no alarm, because it creates the appearance of a safeguard.

Interstitial lag. Sensor glucose reflects interstitial fluid, which trails blood glucose by several minutes. During a rapid fall the sensor reads higher than the blood actually is. Workers need to know this, because it means a falling trend arrow is more urgent than the number suggests — and it is another reason the arrow matters as much as the figure.

Compression lows. Lying on a sensor can produce a falsely low reading, which is a common cause of unexplained overnight alarms. Worth knowing before a regimen is changed on the strength of it.

Confirmatory finger pricks. Depending on the product and the situation, a finger prick may be required to confirm before acting on a reading — particularly where symptoms and sensor disagree, or during rapid change. The plan should say when.

Scans and procedures. Most sensors have to come off before an MRI and certain other scans — worth raising when the appointment is booked rather than discovering it at the scanner.

Consent and data, which nobody thinks about

CGM generates a continuous stream of physiological data about a person, and in supported accommodation several people may be able to see it.

That deserves a proper decision rather than a default. Who can follow this person’s glucose? Does the person know and agree? What happens to the data? Is a provider using it for anything other than clinical response?

For someone with capacity, this is their decision and their data. For someone without, it is a supported decision-making question — and the answer is usually that sharing is appropriate and proportionate, because the purpose is their safety. But the fact that the answer is usually yes is not a reason to skip asking.

The specific thing worth avoiding: glucose data being used as an accountability tool against the person. A pattern of high readings discussed in a house meeting, or described to the person as a failure, turns a safety device into surveillance. The data exists to adjust the regimen, not to judge the person.

What to ask the diabetes team for

If somebody is on CGM, these are the things worth requesting explicitly:

• The ambulatory glucose profile and time in range report, reviewed at each appointment
• Alarm thresholds set deliberately for this person, with the reasoning recorded
• A written protocol covering every point in the alarm section above
• Training for the support team, on this system — a credentialled diabetes educator will often do this if asked
• A review of whether the regimen itself can be made less hypo-prone, now that there is data showing what is actually happening
• Clarity on when a confirmatory finger prick is required

And if somebody is not on CGM but cannot report symptoms, the thing worth requesting is a documented clinical opinion on whether they should be, and what the alternative risk mitigation is.