Dementia and the NDIS: Access, Evidence and the Progressive-Condition Problem

Dementia and the NDIS: access, evidence and the progressive-condition problem

The NDIS was built for conditions that are stable.

Not deliberately, and not in the legislation — but in its machinery. Annual plan cycles. Capacity building aimed at increasing independence. Goals set for the year ahead. A workforce whose experience is mostly people with lifelong intellectual disability, autism or physical disability, whose support needs next year will look much like this year’s.

Dementia is the opposite of that. It is progressive, the rate is unpredictable, capacity declines rather than builds, and in younger onset dementia cognitive and language function can decline faster than in older people. A plan written in March can be wrong by October.

This guide covers how to get into the scheme, how to evidence it, and — the part that matters more over time — how to run a plan for a condition that will not hold still.

One thing first, because it is time-critical: you must become an NDIS participant before your 65th birthday. Not diagnosed before 65 — a participant. Everything else in this guide is secondary to that.

The age-65 deadline

You cannot become an NDIS participant after you turn 65. Nothing about the strength of the need changes it: there is no exception written into the Act and nothing to appeal.

For dementia this is harsher than for almost any other condition, because of the arithmetic. Younger onset dementia typically presents in the 50s and early 60s. The average diagnostic delay is around four years. So a person whose symptoms began at 58 and who was told it was depression until 62 may have a year or two of window left, and nobody will mention it.

If you are under 65 and have a dementia diagnosis — or are being investigated for one — apply now. Not when needs increase. Not when the diagnosis is refined. Now.

Some people find that uncomfortable, as though it were gaming the system. It is not. The scheme has an arbitrary hard boundary and a progressive condition does not. You still have to meet the disability test, so this is not applying speculatively — but “we’re managing at the moment” is not the test.

If you are already a participant

Much better placed.

• A participant who turns 65 keeps their plan, indefinitely.
• The scheme is left behind in just two circumstances: a permanent admission to residential aged care, or a first permanent home care package taken up once the person is past 65.
• You can choose to stay in the NDIS rather than move to aged care, and for most people with significant disability that is the better option.

So someone diagnosed at 59 who joined the scheme and is now 71 remains an NDIS participant. The clock only matters at the door. We set out what sits on the other side of that boundary in dementia, the NDIS and aged care.

Where dementia sits on the lists

The NDIA maintains two administrative lists.

List A sets out the conditions the NDIA accepts as almost certainly satisfying both halves of the disability test at once, permanence and substantially reduced function. Dementia is not on it.

List B is conditions likely to result in permanent impairment, which settles permanence only. Under the category of conditions primarily resulting in neurological impairment, List B names:

• Alzheimer’s dementia
• Vascular dementia
• Multi-infarct dementia
• Huntington’s disease
• Parkinson’s disease

So the common dementias are there. The practical consequence is the one that recurs across every condition we write about: the diagnosis establishes permanence and does no work at all on functional capacity. An access request consisting of a specialist letter confirming Alzheimer’s disease and an MRI report is a diagnosis, not an application.

A note on the less common causes. Frontotemporal dementia is not named on List B by that name, which does not prevent access — it means the permanence argument has to be made rather than assumed, and a neurologist’s statement that the condition is progressive and irreversible does that easily. Nobody seriously disputes that dementia is permanent.

Which is the unusual thing about dementia applications: permanence is the easy part. For most conditions it is the battleground. Here, the entire weight of the application sits on functional capacity.

Evidencing function

The six domains are mobility, communication, social interaction, learning, self-care and self-management. For dementia the ones that carry an application are self-management, learning, social interaction and self-care, usually in that order. Mobility is frequently unaffected for years, which is precisely why physically-focused assessments understate the need.

Self-management

The domain covering planning, organising, decision-making, managing money, running a household and judging risk. For dementia it is almost always the most disabling area and it is the one most often under-documented.

Write it in dates, numbers and consequences:

• “Cannot plan or sequence a meal. Has left the stove on three times since June, twice overnight.”
• “Requires prompting for every step of the morning routine; takes 90 minutes with prompting and does not start without it.”
• “Cannot manage money. Paid the same invoice four times in August. Now has a limited-balance account with a second signatory.”
• “Missed five of seven medical appointments between May and September despite reminders.”
• “Became lost walking to the local shops in July, a route walked for eleven years. Returned by a neighbour.”

That last one is the register that works. Not “has difficulty with orientation” but what happened, when, and who found her.

Learning

New learning is severely impaired in most dementias while established knowledge persists. Say so specifically: that the person retains their trade or professional knowledge and cannot learn a changed routine, a new phone, or a different route.

Social interaction

Where behaviour and personality change belong, and where applications are weakest because families find this material painful to write down. Disinhibition, loss of social judgment, withdrawal, apathy, suspicion and accusation, and lost friendships are functional impairments with large consequences. In frontotemporal dementia they are frequently the presenting problem. See changed behaviour and dementia.

The evidence that carries weight

Two of those matter disproportionately.

The occupational therapy functional assessment is more useful for dementia than a cognitive test score, because the question is not what the person scores but what happens when they try to make lunch. Ask specifically for a functional cognitive assessment conducted in the person’s own home.

The family statement is free and changes outcomes. A written account of a real week — who prompted what, what was forgotten, what was left on, what was said to a neighbour, how long each routine took, what happened when the usual carer was away — makes the aggregate visible in a way five clinical reports do not.

Say the quiet part about the assessment day

Get it into the reports that presentation on the day is not representative, and say why:

Social conversation on familiar topics is often preserved long after memory and judgment are not. Many people with dementia present well for an hour and cannot run a day. Fluctuation is real, particularly in Lewy body dementia, where a good day and a bad day can look like different people. Fatigue means afternoon function differs sharply from morning. And insight is frequently impaired, so the person will genuinely under-report their difficulties — sometimes emphatically, sometimes while their partner sits beside them unable to contradict them politely.

What to do: bring a support person, submit a written statement in advance, request the person’s better hours, and ensure a report states plainly that self-report is unreliable because of impaired insight. That one sentence prevents a great deal of harm.

The progressive-condition problem

Now the part that matters beyond access.

Getting a plan is the beginning. Keeping it matched to a declining condition is the actual work, and the scheme’s machinery does not help.

The specific frictions, which the research on younger onset dementia identifies directly: annual reviews are too infrequent for needs that can change in months; few disability providers have dementia expertise, especially in rural and remote areas; most of the workforce’s experience is with stable conditions; and families end up training their own providers.

What to do about it:

Use the change-of-circumstances mechanism, and know it before you need it. A plan reassessment can be requested when circumstances change significantly. For dementia they will. Learn how to lodge it, what evidence it needs, and how long it takes — then use it on the first clear step down rather than waiting for the annual review.

Document function continuously. A dated record of what the person could do and when is what makes a change-of-circumstances request credible. Without it you are asserting decline; with it you are demonstrating it.

Never let a plan be built on an assumption of improvement. Capacity building in the NDIS is framed around increasing independence. For dementia the honest goal is frequently maintaining function and delaying decline, and a plan whose goals assume progress will under-fund reality and will be judged to have failed. Write goals as maintenance where that is the truth: “maintain independent showering with prompting for as long as possible” is legitimate and fundable.

Build in flexibility. Core supports are more flexible than capacity building. A plan weighted toward core gives you room to increase support hours as needs change without a fresh application for every adjustment.

Ask for more review than you think you need, and put the reason in writing.

What a dementia plan should contain

With 2026–27 NDIS price limits for reference.

Support work with the right training. Dementia support is prompting, structure, orientation, de-escalation and consistency — not primarily personal care. The service agreement should specify dementia-specific training.

Occupational therapy at $193.99 per hour, for functional cognitive assessment, environmental modification and equipment. Environmental change is one of the highest-yield interventions in dementia and it is under-used.

Speech pathology at $193.99, where language is affected — particularly in primary progressive aphasia — including the non-face-to-face item for communication profiles and team training.

Psychology at $252.99 per hour. Depression and anxiety are common after a dementia diagnosis, particularly in younger onset where insight is preserved, and adjustment support matters for the person and the partner.

Assistive technology for cognition, badly underclaimed: automatic medication dispensers that alarm until the dose is taken, GPS location sharing where getting lost is a risk, smart speakers for voice reminders, large-display clocks showing day and date, sensor lights, and door alarms. Most sits under the low-cost threshold of roughly $1,500 and can come from core or consumables without a formal assessment.

Support coordination, which for dementia is close to essential, because the coordination load is high and rising while the person’s capacity to manage it is falling.

Overnight support planning. The biggest cost lever and the one that arrives with night waking or wandering. A sleepover at $311.79 per night covering up to two hours of active support is a very different proposition from active overnight support at $82.57 per hour — across a year, roughly $163,000 of difference. Know which you will need and get ahead of it.

Respite, through Short Term Accommodation, used before exhaustion.

The 2026 and 2027 changes

From 1 October 2026 for new and reassessed plans, and 1 February 2027 on renewal: Specialist Support Coordination and Capacity Building Choice and Control reduced by up to 50%, Capacity Building Daily Activities by around 10%, with a protected supports list.

From 1 December 2026: providers get 90 days to claim rather than two years.

From 1 January 2027: support coordination from unregistered providers cut 10% and de-indexed.

For dementia plans, which lean heavily on coordination, these matter. The response is to document outcomes — admissions avoided, a tenancy sustained, a carer who did not collapse, function maintained longer than expected. Outcomes are the argument against a cut, and “maintained” is a legitimate outcome in a progressive condition.

If the answer is no

Internal review within 3 months, decided within 90 days. Most dementia refusals turn on insufficient functional evidence rather than on permanence, and that is fixable in the window — usually with a neuropsychological assessment, an in-home occupational therapy functional assessment, and a rewritten specialist letter.

Administrative Review Tribunal within 28 days of the internal review decision, no fee, with Commonwealth-funded NDIS Appeals advocates available through disability advocacy services.

The specific thing to check: whether the decision relies on the person’s own account of their function. If the reasons say the applicant reported managing independently, and you have documented impaired insight, that is the most commonly reversible error here.