From 1 April 2027, a new tool will determine the NDIS funding for approximately 800,000 Australians with disability. The government calls it the gold standard. Independent researchers are raising serious questions about whether that description is warranted.

This week, an investigative report by Michael West Media — drawing on a detailed Senate submission (no. 1567) from a psychologist and PhD candidate who requested anonymity — set out a series of pointed concerns about the scientific foundations of I-CAN v6: the Instrument for Classification and Assessment of Support Needs, version 6.

These are not fringe concerns. They have been formally put to parliament. A clinical psychologist who independently reviewed the submission described the findings as "deeply concerning" and said they should worry every clinician and every person with a disability.

This article sets out what the concerns are, what the government has said, and what participants should understand as they prepare for a system that is coming — whatever its imperfections.

---

What Is I-CAN v6?

I-CAN v6 is a structured assessment tool originally designed as a qualitative care planning instrument. Under the NDIA's new assessment framework, it has been adapted to serve a very different purpose: generating a budget figure that determines how much NDIS funding a participant receives.

The process works like this: a participant sits through a structured interview of up to three hours. Their answers to questions about daily functioning across key domains are entered into a computer. An undisclosed algorithm converts those answers into a dollar figure. According to the NDIA's own staff briefings, once I-CAN generates a budget, NDIA staff have no authority to change it.

The rollout begins 1 April 2027 for participants aged 16 and over, phased over five years. For new applicants, this process begins from January 2028.

---

The Concerns Raised

Validation: Only Versions 2.0–4.2

The most significant concern in the Senate submission is about validation. Published validation studies of I-CAN cover versions 2.0 to 4.2 only. Version 6 — the version being deployed — has not been independently validated.

The researchers who conducted those earlier validation studies are the same group who developed I-CAN: the Centre for Disability Studies (CDS). The lead author holds a personal royalty agreement on the tool — a fact disclosed in their own published research. CDS is also a consortium partner in the NDIA's implementation of I-CAN.

In 20 years, no independent researcher outside CDS has been asked to check whether the tool works as intended.

Two further findings from the submission compound the concern. First, the developers' own published data show that I-CAN scores explain less than 4% of what participants actually receive in funding — meaning the other 96% is determined by factors the tool does not measure. Second, the reliability studies underpinning the tool were conducted on a sample of no more than 20 people, far too small to draw meaningful conclusions from. The developers themselves acknowledged in 2009 that using I-CAN to determine funding would require significantly more rigour than this. That additional rigour never happened.

The 94% Accuracy Figure

The CDS's published research headlines a 94% accuracy figure. The Senate submission examined how this figure was calculated.

It was achieved by selecting 49 of 186 real-world assessment reports and discarding the remainder as too inconsistent. When all 186 reports are included, accuracy drops to 65%.

The tool's own literature acknowledges that 65% accuracy is "acceptable only for low-stakes research." Determining whether a participant receives $40,000 or $400,000 in annual funding is not low-stakes.

The Algorithm Has Never Been Independently Reviewed

I-CAN provides scores across functional domains. A separate, undisclosed algorithm then converts those scores into a dollar figure. That algorithm — the mechanism that translates assessment into funding — has never been independently reviewed.

Participants who receive less than they need have no way to know whether the problem lies in how they answered the questions, in the I-CAN scoring, or in the algorithm that converts scores to dollars. There is no way to identify where an error occurred.

The Senate submission also raised the possibility that the software may meet the legal definition of a software medical device under the Therapeutic Goods Act — a category that would ordinarily require TGA approval. The government has not approached the TGA.

Assessors Without Professional Registration

Job advertisements published for Support Needs Assessors confirm that "active professional registration is not required" for the role. A participant's NDIS funding could be determined by someone with a counselling degree or an education qualification — with no professional disciplinary pathway if something goes wrong.

This is connected to a legal change buried in the new Act that has received very little public attention:

• Old law: Assessors were required to "identify" a participant's disability and support needs — a clinical determination requiring professional judgement.
• New Act: Assessors are now required merely to "include information about" those needs.

The effect is significant: the assessor is entering data. The algorithm makes the determination. This single word change allows an unregistered workforce to conduct what is, in every practical sense, a clinical assessment — without being legally required to answer for it.

The PECQ — and the Pipeline Problem

I-CAN v6 does not operate alone. Alongside it sits the PECQ — the Personal and Environmental Circumstances Questionnaire — as part of a broader assessment pipeline. The Senate submission found no published study examining how I-CAN and the PECQ work together or what happens to accuracy when both are applied. Three tools contribute to one funding outcome. If any one fails, there is currently no way to identify which one did.

The ART Cannot Override the Outcome

Under the previous NDIS framework, the Administrative Review Tribunal changed the outcome in 73% of NDIS decisions it reviewed in 2024–25. These were cases where independent review found the NDIA had got it wrong, in nearly three of every four cases examined.

Under the NDIS Amendment Act 2026, the ART can no longer alter a participant's plan or reinstate funding. It can only order a reassessment — by the same algorithm that produced the contested result. The system that corrected nearly three in four decisions is now effectively unable to correct them.

---

What the Government Says

The NDIA describes I-CAN v6 as a "gold standard" assessment tool. In response to questions about its validation, the agency has pointed to the Centre for Disability Studies' own published research and the tool's 20 years of development.

NDIS Minister Senator Jenny McAllister has acknowledged that some uncertainty is inevitable during the transition. The government's position is that I-CAN v6 represents a substantial improvement on the current system — which lacks any standardised assessment framework and has produced highly variable outcomes across participants with similar support needs.

The argument for standardisation is real. The current system's inconsistency is a genuine problem — participants with identical needs can receive dramatically different plans depending on which planner they see and in which state. A structured assessment framework, in principle, offers more consistency.

The question being asked by researchers and advocates is whether this particular tool — in its current, unvalidated form, backed by an unreviewable algorithm and delivered by non-registered assessors — is the right mechanism to achieve that consistency.

---

What PWDA Is Advising Participants

People with Disability Australia (PWDA) held a community forum on 18 September 2026 to explain the changes and offer practical guidance. Their advice to participants navigating the new system is notable for what it emphasises.

Among their key recommendations:

PWDA's advice was specific: evidence should explain support needs, not only diagnosis — what support is needed, how much and how often, what happens without it, and whether needs fluctuate. As PWDA put it, "what you can do once, or on a good day, may not show what you can do safely and reliably over time."

This is not coincidental advice. It is a direct response to the reality that a structured assessment tool — whatever its limitations — will ask structured questions about functional capacity. The quality of the answers, and the evidence backing them up, is the most controllable variable in a system where the algorithm is not.

---

What This Means for Participants: A Practical Reading

The concerns raised about I-CAN v6 are genuine and deserve serious public debate. They have been formally put to parliament. Researchers, clinicians and advocates are right to scrutinise a tool that will determine the funding of 800,000 people.

But for participants facing this assessment — with or without these debates resolved — the practical implications point clearly in one direction:

Thorough preparation is more important, not less.

If the algorithm's conversion of assessment answers into funding is opaque and not independently reviewable, then the quality of the inputs — your documented evidence, the specificity of your answers, the completeness of your functional picture — is the most controllable variable you have.

An assessor who is entering data into a system, rather than making an independent clinical judgement, can only work with what you give them. If your answers to questions about daily functioning are vague, incomplete, or don't capture your needs on a bad day, the algorithm cannot compensate for that.

Specifically, this means:

Describe function, not diagnosis. The I-CAN assessment asks about what you can and cannot do, not what diagnosis you carry. "I have spinal muscular atrophy" tells the system less than "I cannot raise my arms above shoulder height and require assistance to dress, prepare food, and reposition during the night."

Document fluctuation explicitly. What you can do once, on a good day, in a familiar environment, is not the same as what you can do safely and reliably every day. Your evidence needs to capture bad days, high-demand periods, and what happens when supports are not available.

Gather professional evidence that goes beyond diagnosis. Occupational therapy reports, physiotherapy assessments, speech pathology reports, and psychiatric assessments should all describe functional limitations in daily activities — not just diagnose a condition.

Build a support diary. A contemporaneous record of daily support needs and their consequences is often the most persuasive evidence in any review process. It documents the lived reality that a structured interview may not capture.

Know your review rights. The ART's ability to override outcomes has been constrained, but internal plan reassessments remain available when circumstances change. If you believe your assessment was inaccurate, seek independent advocacy support promptly.

---

The Broader Context: Pricing and Workforce Changes

This week, NDIS Minister Mark Butler also made the first pricing determination under the new Act, setting 2026-27 price limits for NDIS supports. The government is also consulting on differentiated pricing — adjusting support prices based on whether they are delivered by registered or unregistered providers — with recommendations expected from 1 January 2027.

Separately, the Fair Work Commission's gender-based undervaluation decision for the SCHADS Award takes effect 1 December 2026, with consequential pricing updates to follow. These changes will affect the cost of disability support workers — one of the most commonly claimed NDIS supports.

---

The Standard of Evidence the Government Applies to Itself

Perhaps the sharpest observation in the Michael West investigation is this:

The NDIS Amendment Act 2026 allows the NDIA to refuse a support if no peer-reviewed research backs it — even when a participant's own doctor recommends it.

I-CAN version 6 — the tool used to assess that same participant — has no peer-reviewed independent validation.

The government is applying to people with disability a standard of evidence it cannot meet itself.

This tension will not be resolved before April 2027. The rollout will proceed. For the 800,000 Australians it will affect, the most powerful response available is the same one that has always mattered most in any assessment process: a thorough, honest, evidence-based account of what your life actually requires.

---

Sources: Michael West Media (September 2026, reporting on Senate Submission 1567); PWDA Community Forum (18 September 2026); NDIA ndis.gov.au (September 2026).

ICANReady is an independent tool designed to help NDIS participants articulate their functional support needs across the 12 I-CAN v6 domains. It does not provide legal, clinical or financial advice. The concerns described in this article are drawn from publicly available sources and Senate submissions; ICANReady does not independently verify the claims made in those sources. If your NDIS plan or assessment is being reviewed, seek support from an independent disability advocate or legal service.