Butler and National Disability Insurance Agency (NDIS)
[2025] ARTA 1579
“Recording symptoms as provided by the Applicant should not be confused with making expert recommendations about appropriate disability supports.”
What the record established
Ms Butler is an NDIS participant on the basis of Ehlers-Danlos syndrome, ME/CFS and postural orthostatic tachycardia syndrome [3]. She also lives with other conditions, including long COVID, lipo-lymphoedema and mental health conditions, that the NDIS hasn't accepted for access [6], [8]. Her current plan was funded at about $200,000 [30].
She asked for eleven categories of additional support: 415 hours a year of physiotherapy and exercise physiology, 47 hours a week of support work, a higher level of support coordination, about 55 consumables and low-cost items, a power wheelchair, air-conditioning, a dishwasher and more [28]. Seven practitioners gave evidence, one of them called by the NDIA, and all of them supported the requests [43], [70].
The NDIA's case was about how those recommendations were made. She had given her practitioners a written list of the supports she wanted, and they repeated it in their reports [18].
Where it stopped
The Tribunal agreed. Her GP “only repeated what she had been given” [104(e)], and even the NDIA's own occupational therapist “merely reflects what she had been told” [124(a)]. A list of supports passed through a practitioner's report is still the participant's list.
Practitioners who would deliver the supports were weighed accordingly: “some of the witnesses who recommend the increased support are likely to benefit from any such increase” [112(b)], [104(a)]. Two reports were written with AI assistance after a single telehealth contact, and one author couldn't say who had added the words “minimum necessary” [104(d)].
No one separated the conditions the NDIS accepted from the ones it didn't, so no one could say which hours served which [118(f)], [129(a)]. And the requests didn't fit the person the evidence described. A demanding 415-hour program was “disproportionate” for someone described as mostly in bed since a COVID infection [118(b)], and evidence that she couldn't stand safely sat beside evidence that she drives a manual car [129(d)].
Every item was refused, including two extra hours a week of support work the NDIA itself had offered [129].
Butlerat [112(b)]
some of the witnesses who recommend the increased support are likely to benefit from any such increase
The same reasoning the Guidance and Appeals Panel later applied in KDKJ.
Butlerat [104(d)]
The practitioner must, however, be responsible and accountable for the content of a report.
The Tribunal had no objection to software being used to write up a practitioner's own observations. Every word still belongs to the practitioner.
Butlerat [118(f)]
there is inadequate evidence about what proportion of the additional hours sought relates to the s 24 disabilities
Since October 2024, a support has to address the impairments the NDIS accepted. The evidence has to say which part of a request does.
Read the decision
Butler and National Disability Insurance Agency (NDIS) [2025] ARTA 1579
Published in full on AustLII, and citable by that reference.
Read with Brown [2025] ARTA 2029. The same city and the same advocacy service six weeks later: two specific items, each tied to an accepted impairment, both funded.
These are published Tribunal decisions, not my clients. They're summarised to show what the records in each matter did and didn't establish. Nothing here is a prediction about any other matter, and no outcome is implied. Every decision turns on its own evidence.