I was born and raised in the Pilbara. Before becoming a psychologist, I spent close to a decade as a child protection worker in some of our highest-risk remote Indigenous communities. In 25 years of practice, every Aboriginal client I have seen has forced removal in their family background. So when more than a dozen Aboriginal children were reportedly found living in squalor in a state-owned home in South Hedland, my first question was not how this happened, but how anyone could be surprised.

In 2019, my company completed a comprehensive cultural competency audit of the WA Department of Communities. It gave government a measurable, evidence-based road map to solutions. It was buried, and WA Government and Department leadership denied the data was accurate. More on that later.

The data has been telling us for years

• Almost 60% of children in out-of-home care in WA are Aboriginal, from just 7.3% of the state’s child population. When I undertook my audit in 2019, it was 54%.
• In the decade to June 2024, WA’s out-of-home care population grew by 1,089 children. 1,034 were Aboriginal. 55 were not.
• Across Australia, Indigenous child removals have blown out by 119% over a decade while non-Indigenous removals fell by 12%.
• Between 2009 and 2020, as Aboriginal removals in WA climbed, the Department’s Aboriginal staffing almost halved, from 9.6% to 5.8%, due to systemic racism. The correlation between the two was 0.865, a very strong relationship.

When leadership denies racism, it becomes systemic, and it is inextricably linked to rising Indigenous child removals.

Driver 1: The evidence-based programs that keep families together do not reach Aboriginal families

Intensive family therapy and complex attachment-based programs are our best evidence of what keeps at-risk families together. In Queensland, the NT and WA, up to 77% of these programs are used by non-Aboriginal families, because they are built on non-Indigenous models of parenting and attachment. Governments have never attempted to adapt them culturally, yet a workforce that is 94% non-Indigenous keeps using them. Aboriginal families do not engage, receive no intensive support, and are left with removal. It is a system designed to favour the privileged that delivers its services to the least privileged.

Driver 2: The skills on the ground are failing

My 2019 audit tested the actual skills, attitudes and beliefs of 295 WA child protection staff, including in the Pilbara. Not one agreed the organisation provided a culturally safe workplace. Only three met the benchmark for a minimum working standard of cultural competence.

Staff were clear they receive no evidence-based training in culturally determined trauma, attachment or mental health, and that they need it. At best they get a short cultural awareness induction. Kinship mapping is not required, so the more remote you go and the more complex the kinship, the more those biases play out in removals. The mandatory assessment tools also conflate cultural differences in parenting with maltreatment risk (Jenkins et al., 2019). Queensland Child Safety has dropped such tools as its core assessment. Every other jurisdiction still uses them.

That matters because 80% of substantiated harm for Aboriginal children is emotional abuse or neglect: the categories most vulnerable to cultural bias, at the low end of the triaged risk continuum, and the most amenable to prevention. A worker who cannot tell cultural difference from risk will remove the child who is safe in a kinship network and miss the child who is genuinely in danger. Too quick to remove. Too slow to protect.

In my time as a child protection worker, I did not remove one child. I also did not leave one child in an abusive situation. You can be about child removal and against it, if you are child-focused and culturally and clinically skilled enough to translate mainstream programs for Aboriginal families in desperate need of help. Most Aboriginal child protection workers also have forced removal in our own families, so our motivation to keep families together is far greater. We know its generational devastation first-hand.

Driver 3: The science has never been done

Ainsworth's Strange Situation Procedure, the foundation of parent-child attachment assessment, has been validated in more than 40 countries but never tested with Aboriginal Australians. So collective caregiving, kinship obligation, and skin and avoidance relationships are often misread as attachment deficits, and in some cases confused for abuse. A child with many mothers looks like a child with none. A child whose caregiver is living with trauma can respond in ways that look like disorganised attachment, the category most associated with abuse.

Add the trauma of forced removal. A parent who has lived with the ever-present threat of losing their child can learn that attaching is unsafe. That is not a failure to love. It is fear of loss. But the racial empathy gap reads it as Aboriginal parents caring less, and a lack of power in the child protection system makes removal impossible to fight.

No evidence-based cultural attachment program has ever been developed in this country, and no attachment program has shown improved attachment between Aboriginal caregivers and their children. Aboriginal families are simply denied the opportunity non-Indigenous families have to heal.

We can see risk before it emerges: what geomapping gives us

In the Kimberley, 100% of children in care are Aboriginal. In the Pilbara, 96%. The Goldfields, 87%. Murchison-Gascoyne, 86%. These are also the most densely Aboriginal-populated regions in WA. Take those five districts out of WA's 17 and the Aboriginal share of removals plummets from 57.18% to 39%, a fall of almost a third. Those regions account for roughly one in three children in care in WA, and more than half of all Aboriginal children in care.

The more remote you go, the higher the rates of removal. Aboriginal children are seven times more likely to be involved with child protection and almost 11 times more likely to be in out-of-home care. Children in very remote areas are four times more likely than city children to have abuse substantiated. Why? The less power and the greater the cultural difference, the higher the removals, the more bias feeds into assessments, and the fewer resources there are for the families in most need.

These families have also lived under the most identity-based human rights restrictions, cashless welfare cards and alcohol restrictions, instead of real prevention. When parents are controlled and punished for their cultural identity rather than their behaviour, it deepens helplessness and race-mediated trauma, which flows straight into caregiving. No one thrives under subjugation.

This is not just child protection. It is every measure. Close the gap in remote communities and you close the gap everywhere. And we are not talking about a lot of people.

In the NT, a staggering 71% of the annual 25,500 child abuse notifications involved the same children, and 85% were already in the system. In every high-risk region there are usually ten to fifteen families, at best, who make up the bulk of service provision. Work in this industry long enough and you will have a client; that client will have kids, and one of those kids will become your client. These are the generationally removed, passing that risk on through the same families again and again.

Bowlby's seminal attachment research tells us that removing a child from their primary attachment figures is close to unrecoverable. Removal is the pipeline to every poor outcome: the justice system, suicide, violence, and the removal of the next generation. Trauma feeds trauma.

Geomapping child protection data would identify those families precisely, debunk the false narrative of endemic abuse across whole Aboriginal communities, and let us target risk before it emerges in the next generation. That is prevention, and it is not even close to what is being done.

The opportunity WA buried

My 2019 audit did something that existed nowhere else: measured the cultural competency of the child protection workforce so it could be tracked against Indigenous child removal rates. Measurable, trackable, improvable. The Department buried it for three years. It became public only when a whistleblower leaked it, and a follow-up review by PwC Indigenous found internal concerns that the recommendations had not been acted on.

Then came the real strategy: ignore it until everyone forgets. Seven years on, the report the WA Government commissioned sits unimplemented, my expertise is never called on, and the Aboriginal share of children in care has climbed from 54% to almost 60%.

What has to happen

• Geomap and target. Map child protection data to identify generationally at-risk families and treat risk before it emerges, starting in the Kimberley, Pilbara, Goldfields and Murchison-Gascoyne.
• Measure the workforce. Mandatory, measurable cultural competency learning outcomes for every child protection worker, tracked annually against removal rates by district.
• Fix the tools. Replace culturally biased assessment tools with culturally valid, co-designed ones, as Queensland did.
• Map kin. Make kinship mapping a mandatory skill, and train workers to do it.
• Build the missing science. Test attachment with Aboriginal families, and fund culturally informed attachment programs commensurate with need.
• Put Aboriginal people in the decisions. Aboriginal staffing as a Department KPI, including in decision-making roles on removal.
• Analyse the causal drivers. Report what is causally driving Aboriginal child removals, not just that removals are rising.

The expertise is here

The cultural attachment work has not waited for government. I secured an MRFF grant for the first study to test attachment with Aboriginal families, and to design and deliver Australia's first culturally informed attachment program for generationally removed caregivers and their children. It does exactly what the system does not: works with the families, builds the evidence, and treats the caregiver trauma of forced removal and racism that compromises attachment, before that risk reaches the next generation. It is being built through the Westerman Jilya Institute with my donated time and intellectual property.

Working for free to deliver Australian-first programs is not new to me. I have done it for over 25 years and racked up 20,000 volunteer hours in my name charity. When the WA Government failed to act on Indigenous child suicide, I funded the Dr Tracy Westerman Indigenous Psychology Scholarship Program to build the Aboriginal psychology workforce our children need. The results speak for themselves. It has supported 79 Indigenous psychology students in a country with fewer than 400 Indigenous psychologists. Close to 70% the first in their family to go to university, and over half from our highest-risk remote communities. Scholars have completed 66 degrees, and 82% go on to postgraduate study, against 27% for all universities nationally.

I hold multiple Australian firsts, much of it peer-reviewed and data-driven: the first culturally validated psychological tests for Aboriginal people at risk; the first national clinical norms for at-risk Aboriginal youth; the first whole-of-community mental health interventions to show risk reduction; the first measure of child protection cultural competency, tested on WA's own workforce; and now the first study to test attachment theory with Aboriginal families and build a culturally informed attachment program from it. The WA Government has chosen to ignore all of it.

I am a Nyamal woman and psychologist from the Pilbara who started my career in child protection. Yet since the 2019 audit and fallout from its release, not one person from the WA Government has reached out for even a conversation. The minute you call out the system failures based on data and science, you are no longer invited to the table.

When children's lives are at stake, you cannot afford to exclude any expertise. WA has chosen to anyway.

South Hedland was not a failure of information. The data existed. The audit existed. The map of where the risk sits existed. What was missing was the will to use any of it.

So here is the test. Pick up the report you buried, geomap the families who carry the burden of risk, measure your workforce's cultural competencies against your removal rates, and fund actual prevention, including the culturally informed attachment programs that can show they work.

Or go through the motions of more public roundtables, maybe another inquiry, because nothing buys a government time like an inquiry while the public forgets. Keep doing what you are doing, and wait for the next South Hedland. The numbers have already told us it is coming.

Nyamal woman Dr Tracy Westerman AM is an award-winning psychologist and the Founder of Westerman Jilya Institute for Indigenous Mental Health