Dr Maddison Riachi1,2,3, Associate Professor Benjamin Spivak1, Dr Nina Papalia1, Dr Caitlin Cohen1, Dr Harry Dent1, University Distinguished Professor James Ogloff1,3, Dr Abigail Sheed1,3
1Swinburne University of Technology, Centre for Forensic Behavioural Science, Melbourne, Australia, 2Swinburne University of Technology, School of Psychology, Melbourne, Australia, 3Victorian Children's Court Clinic, Melbourne, Australia
Biography:
Dr Maddison Riachi holds a joint appointment as a Lecturer in the Department of Psychology at Swinburne University of Technology and a Postdoctoral Research Fellow at the Centre for Forensic Behavioural Science. Maddison is a registered Clinical and Forensic Psychologist who has worked across youth and adult forensic mental health settings, and she continues to act as an expert witness in cases of child abuse and neglect. Her research centres on understanding, assessing, and managing the risk and impacts of child maltreatment. She is committed to translating research into practice to support children, young people, and families.
Abstract:
Structured professional judgement (SPJ) tools are being used in forensic child maltreatment assessments, yet their evidence base remains thin. Our recent evaluations of two SPJ instruments, the Child Abuse Risk and Resilience Assessment (CARRA) and the Child Protection Removal Assessment (Chi-PRA), found variable interrater reliability and modest predictive performance when validated against judicial outcomes (Cohen et al., 2026a; Cohen et al., 2026b). However, judicial outcomes such as order escalation, whilst meaningful, are imperfect proxies for maltreatment recurrence as they reflect later system responses to reports of harm.
The present study extends this program of research by evaluating the CARRA and Chi-PRA against linked child protection data, providing more granular and direct indicators of subsequent maltreatment concerns. Risk assessments were coded from court-ordered family assessments completed by the Victorian Children's Court Clinic between 2017 and 2022, comprising 433 parent–child dyads. Data were then linked with child protection records to examine post-assessment outcomes, including re-reports and substantiations of abuse. Analyses examine whether summary risk judgements and item-level ratings differentiate children with and without subsequent reported concerns. Findings will be discussed with a focus on implications for practitioners conducting forensic child maltreatment assessments.