21 recorded incidents as a family violence perpetrator. Prior imprisonment for violent offending. Breached court orders. Managed under high-risk police monitoring for violence against other women. By any reasonable measure, Toby Loughnane was not a man who should have slipped through the cracks. Yet on April 11, 2021, he killed his partner Maryam Hamka at his Brighton apartment in Melbourne's southeast, then buried her body in a shallow grave at Cape Schanck on the Mornington Peninsula.
Now, Victoria's State Coroner has delivered a direct finding: the system should have seen this coming sooner.
In findings released this week, State Coroner Liberty Sanger found that a March 6, 2021 incident, in which Loughnane chased Hamka armed with a knife to her mother's home, was triaged as medium risk by Victoria Police's Family Violence Investigation Unit. That classification was in line with the unit's formal guidelines. But Judge Sanger concluded it was the wrong call, given what police already knew about Loughnane, as reported by 9News.
"This incident should have been classified as high risk," the coroner wrote in her findings.
The systemic gap that made this possible is telling. A police report from the March incident noted that the parties had no previous family violence history together, despite Loughnane's documented record spanning 21 separate incidents with Victoria Police. His sister-in-law told officers at the time that Loughnane had "bashed her really bad previously", yet that information was not enough to shift the risk classification.
Victoria Police's risk assessment tool, known as the Family Violence Report (FVR), is a scored, question-based instrument developed in partnership with Swinburne University and Forensicare using data from more than 44,000 Victorian family violence cases. It is designed to guide officers toward appropriate responses, including referral to specialist Family Violence Investigation Units. The system also formally allows officers to use their professional judgement to override a score when high-risk factors are otherwise apparent.
Judge Sanger was careful not to frame her findings as a straightforward policy failure. She did not suggest that a high-risk classification would necessarily have prevented Hamka's death, nor that the triage decision was wrong under the relevant policy at the time. Her concern was subtler but no less serious: the tool, applied mechanically, may have obscured the full weight of Loughnane's history. "Even if the risk assessment tool did not assess this as high risk," she wrote, "this does not obviate the need for members to use professional judgment."
That is a distinction worth sitting with. Standardised risk tools exist precisely to reduce the subjectivity that can lead to bad outcomes, as research consistently shows that unstructured officer judgement performs poorly. But a tool calibrated around couple-level history can, as this case shows, underweight a perpetrator's broader pattern of violence. Loughnane had been subject to high-risk monitoring in relation to other women. That history, by any professional reading, should have coloured the March assessment.
Loughnane was convicted of murder by a jury and, in February 2025, sentenced to 28 years in prison with a 20-year non-parole period. Sentencing judge Justice Christopher Beale described the murder as occurring in the context of "protracted family violence" and noted a pattern of extreme threats and physical abuse during the couple's 15-month relationship.
Coroner Sanger has recommended the state government fund an independent evaluation of Victoria Police's risk models and its Family Violence Investigation Units. She also called on the Department of Families, Fairness and Housing to provide ongoing funding for long-term perpetrator intervention programmes. Those two recommendations reflect a clear-eyed understanding that no single reform will close every gap: better tools need to be matched with sustained, adequately resourced intervention upstream.
The question governments will face is a familiar one in public policy: how much independent scrutiny of institutional frameworks is enough, and how regularly should these models be stress-tested against real cases? Risk assessment tools require ongoing validation. The coroner's call for independent evaluation is not a criticism of the system's design so much as a reasonable expectation that any system managing life-and-death decisions must be subject to rigorous, arms-length review.
Maryam Hamka was 36 years old. Her family has said the justice system delivered a verdict, but justice cannot return her. The harder task, the one that remains unfinished, is making sure the institutional machinery built to protect women like her actually functions as designed, and that officers on the ground are both equipped and empowered to act on what the numbers alone cannot fully capture.
Support is available by calling 1800RESPECT on 1800 737 732, Lifeline on 13 11 14, or the Men's Referral Service on 1300 766 491.