Ms Panayiota Zingirlis1, Associate Professor Caleb Lloyd1, Dr Janet Ruffles1, Professor Michael Daffern1
1Swinburne University, Australia, 2Justice Health and Forensic Mental Health Network, SYDNEY, Australia
Biography:
Ms Yiota Zingirlis is a Forensic and Clinical Psychologist, Director of Zingirlis Psychology, and a part-time Senior Psychologist with Justice Health. She has over 20 years of experience delivering psychological assessment, treatment, and intervention across forensic, clinical, and correctional settings. She provides education and training to mental health professionals through universities, government agencies, and private organisations. Ms Zingirlis is an accredited trainer in the HCR-20 Version 3, Violence Risk Scale Version 2, and Structured Assessment of Protective Factors. She has co-developed violence reduction programs and co-authored research in the areas of custodial and forensic mental health.
Abstract:
Background:
The Historical Clinical Risk Management-20 (HCR-20) is a structured professional judgement tool widely regarded as the gold standard for violence risk assessment in forensic mental health. While its predictive validity is well established, limited research has examined whether repeated HCR-20 assessments improve the prediction of outcomes in routine community practice, where risk is dynamic and subject to change over time.
Method:
This longitudinal study examined 171 forensic patients granted conditional release in NSW between 2008 -2020. A total of 852 HCR-20 assessments were analysed. HCR-20 total and subscale scores served as the primary risk measures. Justice outcomes, and other adverse clinical outcomes, were extracted through systematic review of clinical reports and case files. Cox regression models compared the predictive performance of the initial HCR-20 assessment conducted proximal to conditional release with subsequent repeated assessments to determine whether reassessment provided incremental predictive validity over time.
Results:
Repeated HCR-20 assessments demonstrated improved prediction of justice and adverse outcomes compared with baseline assessments alone, supporting the value of ongoing violence risk reassessment in community forensic mental health. Repeated assessments showed incremental predictive validity across multiple outcomes, highlighting the contribution of dynamic risk factors to prospective risk estimation.
Conclusions:
This study represents one of the largest longitudinal evaluations of repeated HCR-20 assessments in an Australian forensic mental health cohort. Findings support routine reassessment as a means of enhancing risk prediction and have important implications for clinical decision-making, risk management, and evidence-informed policy within community forensic mental health services.