Child removal in child protection practice: comparing structured professional judgement and actuarial risk assessment instruments

Investigator: Lillian de Bortoli

Supervisors: James Ogloff, Jan Coles, Mairead Dolan

Rationale and aims:

  • The decision to remove a child from the family home is affected by a multitude of factors. Decision-making in child protection practice occurs in uncertain contexts leading to errors in judgement.
  • Risk assessment instruments have emerged to assist practitioners with decision-making. In Australia, instruments adopting consensus and actuarial approaches dominate child protection practice. Consensus instruments comprise broad themes to guide risk assessment and actuarial instruments which are evidence-based, tend to be inflexible and focus upon historical items.
  • The PhD study involved the development and a pilot evaluation of an alternative instrument adopting a structured professional judgement (SPJ) approach to assessing risk. The instrument was developed to assist practitioners decide whether sufficiently serious protective concerns exist to warrant child removal from the family home.


  • A selective literature review of factors associated with child removal and child abuse-related fatalities formed the basis for developing the items in the newly devised instrument, the ChiPRA (Child Protection Removal Assessment).
  • A cohort of 298 cases from the Victorian Children’s Court was coded using (1) ChiPRA and (2) FRE-SDM™ (Family Risk Evaluation-Structured Decision Making), a comparison actuarial instrument.
  • A Custody to Secretary Order was granted in 49% of cases which led to child removal and a Supervision Order was granted in the remaining cases which led to the child continuing to reside within the family home.


  • The predictive validity for ChiPRA was significantly better than the FRE-SDM™. ChiPRA scores correctly classified 86.3% of cases compared with the comparison instrument, 60.9%.
  • ChiPRA items most associated with removal were unmet child needs, concerning housing/physical safety and parental non-compliance.
  • Factors most predictive of removal tended to be focussed upon the child’s needs, agency requirements and parental attitudes and behaviours, whereas measures of child vulnerability determined by age or disability were not associated with child removal.
  • Other items not associated with child removal included parental mental illness, economic hardship and child-related stresses.


  • Overall, the findings indicate that court-ordered child removal decisions rely mostly upon dynamic factors and place limited emphasis upon static factors and child vulnerability.
  • Given the emphasis placed upon dynamic factors, further examination of SPJ is warranted in child protection practice
  • Given the promising preliminary results, prospective evaluative research is warranted to further explore ChiPRA. Further research in this field will mature risk assessment practice in child protection and contribute to an innovative field for study and research.

Clinical implications and research translation:

  • Child protection practitioners are required to work collaboratively with many organisations within the broader family welfare system. A structured framework that improves consistency, transparency and accountability of risk assessments will increase the defensibility of risk and risk management strategies to external organisations including courts.
  • High staff turnover rates result in frequent changes to practitioners allocated to families. SPJ standardises decision-making thresholds which maintains consistency for clients engaging with child protection irrespective of practitioners or location of residence.
  • At entry level, practitioners work at the forefront of child protection practice. SPJ establishes minimum standards that guide entry level child protection practitioners conduct an assessment that, at a minimum, is based upon evidence-based research.


De Bortoli, L., Coles, J., & Dolan, M. (2013). Maternal infanticide in Australia: Mental disturbance during the postpartum period. Psychiatry, Psychology and Law, 20(2), 301-311.

De Bortoli, L., Coles, J., & Dolan, M. (2013). Parental substance misuse and compliance as factors determining child removal: A sample from the Victorian Children’s Court in Australia. Children and Youth Services Review, 35(9), 1319-1326.

De Bortoli, L., Coles, J., & Dolan, M. (2013). A review of maternal neonaticide: A need for further research supporting evidence-based prevention in Australia. Child Abuse Review, 22(5), 327-339. doi: 10.1002/car.2250

De Bortoli, L., Coles, J., & Dolan, M. (2014). Linking illicit substance misuse during pregnancy and child abuse: what is the quality of the evidence? Child & Family Social Work, 19(2), 136-148. doi: 10.1111/cfs.12002

De Bortoli, L., Coles, J., & Dolan, M. (2015). Aboriginal and Torres Strait Islander children in child protection: A sample from the Victorian Children’s Court. Journal of Social Work, 15(2), 186-206. doi: 10.1177/1468017314529511

De Bortoli, L., & Dolan, M. (2015). Decision Making in Social Work with Families and Children: Developing Decision-Aids Compatible with Cognition. British Journal of Social Work. doi: 10.1093/bjsw/bcu087

De Bortoli, L., Ogloff, J., Coles, J., & Dolan, M. (2015). Preliminary Evaluation of Structured Professional Judgment to Assess Removal in Child Protection Practice. Journal of Forensic Social Work, 5(1/3), 29-48.

De Bortoli, L., Ogloff, J., Coles, J., & Dolan, M. (2016). Towards best practice: combining evidence-based research, structured assessment and professional judgement. Child & Family Social Work.  doi: 10.1111/cfs.12280