By Jayde Fuller
Every decision in the chain was correct. The harm continued anyway. That isn’t a coordination problem — it’s what happens when nobody’s obligation is the outcome.
The sentence
"That sits with the services, not with us."
I have heard versions of this sentence from almost every part of a health system — regulators, departments, services, funders, and colleges. And here is what makes it difficult.
It's usually true.
A regulator genuinely cannot direct how a health service treats patients. A health service genuinely cannot touch a practitioner's registration. A tribunal can only consider what is put in front of it. A funder holds more leverage than any of them and has no complaints function at all. Each person is describing their own remit accurately.
That is what makes it so effective, and so hard to argue with. Every actor in the chain is downstream of someone else, and every one of them is right.
Which leaves an uncomfortable conclusion. If everybody in the chain is telling the truth, the problem isn't any of them.
Follow one complaint
Take a single complaint and watch where it goes.
An Aboriginal woman is treated dismissively in an emergency department. Her pain isn't believed. She leaves before she is properly seen, and she does not come back.
Suppose she complains. Most don't, and that is its own finding.
It goes first to the health service, which investigates it as a conduct matter against its own policies. Those policies are about communication and professional behaviour. None of them addresses racism specifically. The service finds no breach and closes the file.
Nothing is referred onward, because a regulator can only act on what reaches it.
Had it reached them, the threshold would have been whether one practitioner's conduct fell below an acceptable standard — not whether a ward has a pattern. A pattern is not a regulatory object. No single practitioner owns it.
The tribunal considers only what the regulator refers.
The health complaints body can examine the service, but holds nothing over anyone's registration.
Count the handoffs. Now look for the failure.
There isn't one. Nobody declassified that complaint improperly. Every decision in the chain was correct within the remit of the person making it.
That is the part worth sitting with. This is not a chain of people doing their jobs badly. It is a chain of people doing their jobs properly, producing an outcome none of them would defend.
Which is why "take it more seriously" has never worked on this, and won't.
Why nobody failed
Look at what each actor in that chain is actually obliged to do. Handle the complaint properly. Within time. With reasons. Apply the threshold correctly. Refer what meets it.
Each of those is a process obligation.
Not one of them is an obligation about outcome — that Aboriginal and Torres Strait Islander people experience less racism in their care this year than they did last year.
So the system can run at full compliance, indefinitely, while the outcome does not move. And no one in it has failed, because nobody was ever given the outcome to hold.
Compliance and continuing harm are not in tension here. They sit together comfortably. That is the whole problem, and it is a design problem rather than a moral one.
Who loses something
So here is the question I would put to any health organisation.
Not who is responsible? Everyone answers that one, sincerely, and claims their share. Shared responsibility is what accountability diffusion sounds like when it is being earnest.
The question is this: Who in your organisation would lose something if Indigenous-specific racism didn't reduce?
Lose something specific. A budget line. An accreditation. A registration. A pay increment. A position.
Not would be disappointed. Not would raise it at the next meeting. Would lose something.
If the honest answer is nobody, you have your explanation — and it is a better one than anything about culture, or commitment, or how hard this work is.
I wrote last month about the levers a health regulator actually holds, and how few of them get pulled. This is the same finding approached from the other side. Of all those levers, only one points inward — the one attached to someone's own progression. It is almost never attached to this.
An outcome nobody can fail at is an outcome nobody is managing.
The community-KPI trap
There is an answer to this that sounds right, and I want to take it seriously, because I have argued versions of it myself.
Let community set the measures.
It is the instinct of everyone who has watched institutions mark their own homework. It is also the first thing accountability diffusion eats.
Hand an institution a community-set target and the sentence comes straight back. Not within our operational control. We don't deliver that service. We can influence but we can't require. The measure becomes the thing that gets explained away — and now community has spent its time, its knowledge and its authority on a process that still goes nowhere.
In the rapid literature review and environmental scan I prepared for the Australian Commission on Safety and Quality in Health Care, this was the dominant pattern across every Australian jurisdiction examined. Cultural safety frameworks designed with community input, implemented by institutions, reported on by governments. Progress reports produced by the bodies being assessed, against frameworks those same bodies designed, published at their discretion. Accountability running through institutions rather than to communities. Community oversight of progress, as I put it then, incidental rather than foundational.
The resolution is two things, and it does not work with only one.
Community authority over the evidence. Who defines what counts as cultural safety, who holds the data, who analyses it, and who decides that it is published. Not consultation on the framework. Authority over it.
A named institutional owner who fails if it doesn't move. A role, not a committee. At a level that can actually be held.
Community measures without an institutional owner is diffusion with better manners. An institutional owner without community authority over the evidence is marking your own homework with extra steps. It is the pair of them that does the work.
What this actually buys
None of this guarantees the outcome. Racism in healthcare will not be solved by a sole accountability instrument, and I would distrust anyone who told you otherwise.
What it does is narrower and more useful.
It guarantees that when the outcome does not move, somebody has failed. Right now, in most of the systems I have worked in, nobody has. The harm continues, and the record stays clean, and those two facts have never had to meet.
Accountability that can be handed on is not accountability. It is a referral.
So: who in your organisation would lose something if Indigenous-specific racism didn't reduce?
Not who is responsible. Everyone is responsible.
Who fails.
Sources
Australian Commission on Safety and Quality in Health Care, Evidence to inform options for a national cultural safety standard, prepared by Indigenous Regulatory Practice, September 2026
Jayde Fuller, Advisory or Authority, Indigenous Regulatory Practice, September 2026
Jayde Fuller, The Levers You Already Hold, Indigenous Regulatory Practice, October 2026
