Beyond GMFCS: The Five Classification Systems and What They Mean for Your Supports

At some point, usually in a report you half-read years ago, somebody wrote down a Roman numeral next to your name. GMFCS Level IV. Or II. Or III.

Nobody sat you down and explained what it meant. It just turned up in paperwork and then kept turning up — in assessments, in funding requests, in the shorthand professionals use when they talk to each other about you.

Here’s what most people are never told: GMFCS is one of five classification systems used in cerebral palsy, and it is the narrowest of the five in what it actually describes. If the only number in your file is a GMFCS level, your support plan is probably being built on about a fifth of the available picture.

This guide covers all five — what each measures, what they genuinely tell you, what they don’t, and how they should and shouldn’t shape an NDIS plan.

Why classification exists at all

A reasonable first question: why label people this way?

The honest answer is that before these systems existed, cerebral palsy was described in language that was close to useless for planning anything. “Mild”, “moderate” and “severe” meant different things to every clinician using them. Two people described identically could have entirely different lives.

The classification systems replaced that with something consistent. They describe what a person does in everyday life — not what they can manage once in a clinic with maximum effort, and not what a scan shows. They’re designed to be stable over time, understandable to families, and consistent between practitioners.

That’s genuinely useful. The problem isn’t the systems. It’s that most people only ever encounter one of them.

The five systems

Each uses five levels, numbered I to V, and in each case Level I means the least limitation and Level V the most. They’re separate scales measuring separate things, so a person can be Level I in one and Level V in another.

GMFCS — Gross Motor Function Classification System

Measures walking and mobility.

• Level I: Walks and performs activities without limitations
• Level II: Walks and performs activities, but with some limitations
• Level III: Walks using a hand-held mobility device
• Level IV: Mobility is limited; may use powered mobility
• Level V: Transported in a manual wheelchair

In Australia, the distribution is roughly 63% at Levels I–II, 11% at Level III, and 26% at Levels IV–V.

MACS — Manual Ability Classification System

Measures handling objects — hands, in everyday tasks.

• Level I: Handles objects easily and successfully
• Level II: Handles objects with reduced speed or accuracy; may use alternative methods
• Level III: Has difficulty and needs assistance with activities
• Level IV: Needs continuous support; handles a limited range of objects
• Level V: Does not handle objects and needs total assistance

MACS is arguably the most under-used of the five in support planning, which is odd, because hand function determines an enormous amount about daily independence — dressing, eating, personal care, using a phone, handling money.

CFCS — Communication Function Classification System

Measures sending and receiving messages in everyday life.

• Level I: Effective sender and receiver with both unfamiliar and familiar partners
• Level II: Slower, but still effective, with unfamiliar and familiar partners
• Level III: Effective with familiar partners only
• Level IV: Inconsistent as a sender and receiver, even with familiar partners
• Level V: Seldom effective, even with familiar partners

Note what CFCS measures: effectiveness of communication, not speech, and not intelligence. Someone using an AAC device fluently may be Level I or II. The system is explicitly about whether messages get across, by whatever means.

EDACS — Eating and Drinking Ability Classification System

Measures safety and efficiency of eating and drinking.

• Level I: Eats and drinks safely and efficiently
• Level II: Eats and drinks safely, but efficiency is somewhat limited
• Level III: Eats and drinks with some safety limitations
• Level IV: Eats and drinks with significant safety limitations
• Level V: Cannot eat or drink safely; tube feeding is considered

EDACS is the one most directly tied to risk. Levels III and above indicate genuine aspiration and choking risk, which has direct implications for the competencies support workers need. Our high intensity supports guides cover dysphagia support and enteral feeding in detail.

VFCS — Visual Function Classification System

Measures how a person uses vision in everyday activities.

• Level I: Uses vision easily and successfully
• Level II: Uses vision successfully but needs compensatory strategies
• Level III: Uses vision but needs some adaptations
• Level IV: Uses vision only in highly adapted environments
• Level V: Does not use vision, even in highly adapted environments

Vision problems are common in cerebral palsy and routinely under-identified, partly because they can be hard to assess alongside communication and motor difficulties. VFCS is the newest of the five and the least likely to appear in an older file.

Why one number isn’t enough

Here’s the practical problem with GMFCS-only planning, and it shows up constantly.

Consider two people, both GMFCS Level IV — both with limited mobility, both likely using powered or manual wheelchairs.

Person A is MACS II, CFCS I, EDACS I, VFCS I. They have good hand function, communicate effectively with anyone, eat and drink safely, and see well. Their support needs centre on mobility, transfers and personal care. They might work full-time, live with a housemate, and need help with showering and getting out of the house. Their support is substantial but narrow.

Person B is MACS V, CFCS IV, EDACS IV, VFCS III. They need total assistance handling objects, communication is inconsistent even with familiar people, eating carries significant safety risk, and vision needs adaptation. Their support needs are comprehensive — full personal care, trained mealtime support, a communication system that staff must learn, an environment adapted for vision.

Same GMFCS level. Completely different lives, completely different support plans, and dramatically different funding requirements.

If the only classification in the file is GMFCS IV, a planner has no way to distinguish them. In practice that means Person B gets under-funded — because the plan gets built around the thing that was measured.

What the classifications don’t capture

Even all five together leave out things that matter enormously for adults.

Pain. Not measured by any of them, and extremely common in adults with CP. Pain changes what someone does far more than any level describes.

Fatigue. Also unmeasured, and arguably the single biggest determinant of what an adult with CP can sustain across a week. The classifications describe what someone does; they don’t capture what it costs them.

Variability. Each system gives one level. Real function fluctuates — across a day, across a week, with fatigue, with pain, with illness, with stress. A person who functions at Level II on a good morning and Level IV by Friday evening is classified at neither accurately.

Cognition. None of the five measure it. This matters enormously, because people routinely infer cognitive ability from communication or motor levels, and the inference is frequently wrong.

Epilepsy, continence, respiratory health, mental health, sleep. All common, all absent from the classifications.

What the person actually wants. Obviously, but worth stating. A classification describes function. It says nothing about goals, preferences or the life someone is trying to build.

The adult problem

Something worth knowing if you’re an adult reading your own file.

These classification systems were developed primarily for children, and the published descriptor sets are banded by age — the commonly used version covers age ranges up to 18. The level you were given may have been assigned when you were eight, in a developmental context, by a clinician who hasn’t seen you in twenty years.

Two consequences.

Your level may be out of date. Function changes in adulthood for many people with CP, and classifications are not routinely re-assessed once someone leaves paediatric services. A level assigned in childhood may substantially under-describe current support needs.

The descriptors assume a developmental trajectory. They were designed to describe children growing up, not adults living with decades of cumulative joint loading, pain and fatigue. They remain useful shorthand; they are not a complete adult assessment.

If your classifications are old, it’s worth asking your OT or physiotherapist whether a current functional assessment would better reflect where you are — particularly before an NDIS plan reassessment.

How to use classifications in an NDIS plan

Used well, these are genuinely helpful. Used badly, they flatten a person into a number.

Get all five documented, not just GMFCS. If your reports only carry a GMFCS level, ask your OT, speech pathologist and physiotherapist to record the others where relevant. MACS from the OT, CFCS and EDACS from the speech pathologist, VFCS where vision is an issue. This costs nothing extra at an appointment that’s happening anyway, and it materially improves the picture a planner sees.

Use them as a starting point, not a substitute for description. A plan request that says “GMFCS IV” and stops has told the NDIA almost nothing. A request that says “GMFCS IV, MACS V, CFCS IV, EDACS IV” and then describes an actual morning — the two-person hoist transfer, the 40 minutes of personal care, the modified-texture breakfast with supervision, the communication system staff have to learn — has told them everything.

Make sure the plan addresses every domain with a limitation. This is where the five-system view pays off directly. An EDACS IV should generate mealtime support with assessed worker competency. A CFCS IV should generate communication support, AAC review and staff training. A MACS V should generate personal care hours that reflect total assistance with objects, not just mobility support.

Record what the classifications miss. Explicitly. “GMFCS level does not capture the fatigue that limits Sarah to approximately four hours of activity per day” is a sentence that does real work in a plan request.

Reassess when things change. Classifications aren’t permanent. If function has changed, the documentation should change with it.

Getting all five documented, practically

This is easier than it sounds and almost nobody does it. The practitioners who can record each one are usually people you already see.

GMFCS — physiotherapist or rehabilitation physician. Most likely already in your file, possibly from years ago.

MACS — occupational therapist. Ask at your next OT appointment. It takes minutes for someone who knows your hand function.

CFCS and EDACS — speech pathologist. Both, in the same appointment. If you don’t currently see a speech pathologist and either communication or swallowing is relevant, that’s a referral worth requesting from your GP regardless.

VFCS — requires input from someone assessing functional vision, which may be an optometrist or ophthalmologist experienced with disability, or an OT. The newest of the five and the least commonly recorded.

A practical approach: before your next NDIS plan reassessment, ask each practitioner you see to include the relevant classification in their report, with a sentence on what it means for daily function. You’re not asking for extra appointments — you’re asking for an extra line in a report that’s being written anyway.

Then hold them together in your own health summary, with dates. When a new practitioner asks about your CP, handing over one page with all five classifications and a short functional description saves an enormous amount of explaining and gets you taken more seriously from the first minute.

What this looks like in a plan request

To make the difference concrete, two versions of the same person in an NDIS plan request.

Version A, as most requests read:

James has cerebral palsy, GMFCS Level IV. He requires assistance with daily living tasks including personal care, meals and community access.

Version B, with the full picture:

James has cerebral palsy. GMFCS IV (uses a powered wheelchair, requires hoist transfers). MACS V (no functional object handling; requires total assistance for all tasks involving hands). CFCS III (communicates effectively with familiar partners using a combination of speech and an eye-gaze device; unfamiliar partners require support to understand him). EDACS IV (significant safety limitations; follows a speech pathologist’s mealtime management plan with modified textures and full supervision).

In practice: morning routine takes approximately 2.5 hours and requires two workers for transfers. All meals require preparation to a specified texture and supervision throughout. Support workers require assessed competency in dysphagia support. New workers require training on his communication system before working alone with him. James directs his own care in detail and makes his own decisions; his communication requires time rather than interpretation.

Version A generates a support plan built around mobility. Version B generates a plan that funds what the person actually needs — including the two-worker transfers, the dysphagia competency, and the communication training that determines whether any of the rest works.

Same person. The difference is entirely in what was written down.

Where the systems came from

Brief context, because it explains both their strengths and their limits.

The GMFCS was developed in Canada in the 1990s, by a research group looking for a way to describe gross motor function in children with cerebral palsy that different clinicians would apply consistently. It worked, and it changed the field — suddenly research could compare like with like, and families could be given a description that meant the same thing in two different hospitals.

The others followed over the next two decades as it became obvious that mobility alone was describing too little. MACS for hand function, CFCS for communication, EDACS for eating and drinking, and VFCS most recently for vision.

Two things follow from that history. First, these are research tools adopted into clinical practice, which is why they’re precise about what they measure and silent about everything else — they were never designed to describe a whole person. Second, they were developed for and validated in children, with age-banded descriptors, which is why their fit for adults is imperfect and why an adult assessment needs more than a level.

Knowing this helps you use them correctly: as a shared shorthand that saves explaining the basics, sitting underneath a proper functional description rather than replacing one.

When a classification is wrong

It happens, and it’s worth challenging when it does.

Classifications are assigned by observation and clinical judgement. They can be wrong because the assessment was brief, because the person was having a good day, because the assessor didn’t see them in their usual environment, or because the person masked difficulty — which many adults with CP do habitually, having spent a lifetime being assessed.

Signs a classification may not fit:

• It was assigned years ago and nobody has revisited it
• It describes function in a clinic that bears no relation to function at home
• It was assigned without input from the person or the people who know them
• Everyone around the person disagrees with it
• Support based on it keeps falling short

What to do: ask the relevant practitioner for a reassessment, ideally in the person’s usual environment, with someone present who knows their typical function. Bring a log of a representative week if you have one. And be explicit about variability — a single level for someone whose function swings significantly is always going to be a compromise, and the report should say so.

Common misunderstandings

“Level V means the most severe cerebral palsy.” It means the most limitation in that specific domain. Someone can be GMFCS V and MACS I — unable to walk, with good hand function. Describing a person as “severe” based on one scale flattens them.

“The levels change as you get better or worse.” They’re designed to be relatively stable, and in childhood they rarely change. In adulthood, function can change enough that a level no longer fits — but that’s a reason for reassessment, not evidence the system is broken.

“GMFCS tells you about intelligence.” It does not. None of the five measure cognition. The inference from motor or communication level to cognitive ability is one of the most common and most damaging errors in cerebral palsy care.

“A higher level means more NDIS funding.” Funding is based on assessed need, described concretely. A classification on its own establishes very little; the description of what someone actually requires establishes everything.

“If I’m Level II, I don’t need much support.” GMFCS II describes walking with some limitation. It says nothing about fatigue, pain, hand function, communication or continence. Plenty of people at Levels I–II have substantial support needs that have nothing to do with mobility, and they’re frequently under-served precisely because they walk.

A note for families and support workers

If you’re supporting someone with CP, the classifications are useful to you in one specific way: they tell you what to pay attention to.

A CFCS IV tells you communication will take work and that the person may be inconsistently understood even by people who know them well — which means never assuming silence is agreement, and never assuming a failure to respond means a failure to understand.

An EDACS III or above tells you mealtimes carry real risk and that a speech pathologist’s plan should exist and be followed exactly.

A MACS IV or V tells you that tasks involving hands need assistance — and, crucially, that the person may still want to direct exactly how that assistance happens. Needing someone’s hands is not the same as needing someone’s decisions.

What the classifications should never do is tell you what someone is capable of understanding, deciding or wanting. None of them measure that, and treating them as though they do is the single most common way people with CP get underestimated.

---

We support adults with cerebral palsy across NSW in supported independent living homes, including people with high support needs across multiple domains. Our companion guides cover what changes for adults with CP and moving into supported living. If your classifications are old and your support no longer fits, call 1300 271 824.

General information only, current at the time of writing. Classification descriptors are summarised in plain language and the published systems contain fuller definitions, including age-banded versions. Classification should be undertaken by appropriately qualified practitioners who know the person. Australian distribution figures are from the Australian Cerebral Palsy Register 2023 Report. This guide is not clinical advice. In an emergency, call 000.