South Australia will be the first Australian state to introduce legislation to remove children born to parents convicted of manslaughter or murder and place them under the custody of the relevant minister, the SA government announced recently.
SA Coroner Mark Johns made the initial recommendation after investigating the many years of abuse and eventual death of four-year-old Chloe Valentine. Chloe was injured and died in 2012 after her mother and mother’s partner repeatedly put the girl on a 50-kilogram bike and filmed her crashing into objects.
Across Australia, 34 children were the victims of filicide – the killing of a child by a parent – in the two years leading up to 2012.
In a well-coordinated and well-resourced child protection system, health staff, child protection workers and forensic specialists should already be collaborating to share information about children and families at risk of harm.
These teams should be responding to the warnings some parents give prior to committing these crimes, then deciding whether parents convicted of murdering their children should have subsequent offspring removed.
However, responses to high-risk pregnancy and high-risk infants across Australia could be described as haphazard, and legislation alone will not correct the problem.
South Australia’s move is not without precedent. New Zealand gave the court responsibility for the safety of subsequent children born to parents convicted of murder or manslaughter in 2013. However there is no publicly accessible data on the rate of removal and no evaluations to date.
Importantly, the New Zealand legislation has some flexibility to work with parents convicted of murder or manslaughter, but the onus is on the parent to demonstrate they are safe to parent subsequent children.
We have strong research detailing the demographics of parents who kill children and other contributing factors, such as psychotic depression, parental separation, retribution for leaving a relationship, domestic violence, fatal maltreatment, substance use and so on.
Researchers have also made a number of suggestions about what can be done to prevent filicide aside from removing children, such as treating mental illness and introducing safe havens for anonymous birthing.
But there has been no evaluation of what works and what doesn’t. The ethics of designing a research project that compares treatments, interventions or approaches to protect subsequent children of those who commit filicide over time is an understandable barrier.
While the intent of the South Australian legislation to prevent harm is noble, it doesn’t address the barriers that currently exist in the child protection system.
First, the plan assumes we know who has been convicted of the manslaughter or murder of their child.
The judicial and law enforcement sector knows who these people are. However, other agencies with crucial involvement in child protection only know if they are told by law enforcement. The law enforcement sector cannot be responsible for actively tracking the reproductive activity of those convicted of the manslaughter or murder of children for the term of convicted person’s life.
Removal “at birth” also implies that the health department has been informed of the previous conviction, that the child protection agency knows the woman is pregnant and that they know where the woman will give birth.
By linking the removal “at birth”, the focus is on the woman and does not account for the paternity of the child. It’s unclear how this approach will include males previously convicted who may or may not be identified as the biological father of the unborn/newborn or current partner of the pregnant woman.
Responding to high-risk pregnancies
Child protection authorities were first notified about Chloe Valentine during her mother’s pregnancy. According to the Coroner, a report to the Child Abuse Report Line described Chloe’s mother as “dirty”, “smelly”, “unhygienic”, “taking drugs and drinking”, and “not obtaining proper antenatal medical treatment”.
It is not uncommon for child welfare risks to emerge during pregnancy. Indicators of possible future harm to the child include suicide risk, serious and persistent substance use, being a victim of domestic violence, homelessness and inadequate preparations for birth.
Child protection legislation, risk studies and structured decision-making tools are commonly employed in child protection situations to assess and respond to risk. One of the key planks in the platform to monitor pregnant women at risk of domestic violence, unmanaged mental illness and substance use, is the unborn child high risk birth alert operating in some jurisdictions.
Some states have legislation to support the prenatal reporting of children at risk. In states or territories where there is no law specifically relating to unborn child information exchange, inter-agency collaboration protocols allow agencies to share information about unborn children.
These policy responses are aimed at redressing the risks giving rise to the prenatal report, prior to the child’s birth; in other words, getting to the cause of maltreatment and working on those risks.
But the patchwork of responses states and territories have to high-risk pregnancy and high-risk infants is problematic. There is also a risk that pregnant women under surveillance will avoid contact with services, such as avoiding antenatal care, deny they are pregnant, move to other jurisdictions to give birth, not register the child’s birth, and so on.
Systems that seem to work best are integrated, supported by cross-agency policy and strive for active engagement with service users. Wrap-around intersecting interagency responses must be coordinated, backed by the legal sharing of child protection information.
Systems that don’t work well are fragmented, typified by lack of collaboration and communication. Structural issues such as conflicting service mandates, under-resourcing and difference in practice frameworks can result in services being provided in silos to at-risk families, rather than in cooperation.
This complex area requires a sophisticated response. There may be pockets where programs are operating well, but there is no systematic approach across the country, or even state-wide. Any risk-management system in child protection needs multilateral agency participation to ensure such a system delivers on what it is trying to do: protect subsequent children.