Background. Child abuse and neglect seriously affect a child’s healthy development and should therefore be identified in time. On the other hand, it is also important not to falsely identify unsafe situations, because false... [ view full abstract ]
Background. Child abuse and neglect seriously affect a child’s healthy development and should therefore be identified in time. On the other hand, it is also important not to falsely identify unsafe situations, because false accusations may be traumatic and harmful to families. Assessing and deciding on child maltreatment cases is difficult, due to both the characteristics of the decision situation and professionals’ limited cognitive resources to process big amounts of information. Risk assessment instruments have been developed for professionals in child protection, to improve their assessments of children’s safety and risks related to abuse and neglect. In The Netherlands, the LIRIK (Light Instrument Risk Assessment Child Safety) was constructed as a tool for professionals in youth care to reach explicit conclusions about the actual safety of children and possible risks in the family situation in a structured way.
Purpose. We conducted two studies that present insights into the reliability and validity of the LIRIK in comparison to unstructured professional judgment (Bartelink, De Kwaadsteniet, Ten Berge, Witteman, & Van Gastel, 2015). In a reliability study, professionals made safety and risk assessments for 12 vignettes, with (n=36) or without (n=43) the LIRIK. In a validity study, we compared professionals’ safety and risk assessments for 370 children, made with or without the LIRIK, with outcomes on child maltreatment investigations, child protective orders and out-of-home placement in case files six months later. In addition, we conducted a literature review to find promising ways to further improve decision-making in child maltreatment cases (Bartelink, Van Yperen, & Ten Berge, 2015).
Key findings. Agreement on safety and risk assessments were poor to moderate in the LIRIK and control group (Krippendorff’s alphas of .19 to .48 with the LIRIK; 25 to .46 without). Differences between groups were small and inconsistent. Agreement on the presence of signs and risk factors of maltreatment was also poor. The predictive validity for safety and risk assessments was poor, both with the LIRIK (rs between -.01 and .37) and without (rs between .04 and .44).
Implications and recommendations. Other risk assessment instruments similarly have been found to show limited reliability and validity. These findings raise important questions about the value of risk assessment instruments. Other ways to improve professional safety and risk assessment should be studied. We assume that a combination of strategies is needed to improve decision-making on child maltreatment cases. In our literature review, indications were found that shared decision-making with parents and children, team decision-making and feedback to professionals about the accuracy of their assessment may lead to better assessments. Some studies suggest that shared decision-making with parents and children could have a positive impact on treatment outcomes. Methods such as motivational interviewing, solution-focused therapy and family group conferencing can help practitioners to make the preferences of clients explicit and stimulate them to cooperate. These methods have in common that they focus on the clients’ perspective on the problem and the possible solutions, they empower clients by seeing them as experts of their own life and by asking them for solutions, and they support self-efficacy. Although team decision making can be biased too, it may be helpful for increasing transparency and discussing normative aspects and professional experiences and boundaries. In line with this, the systematic provision of feedback about the accuracy of assessments can create better learning opportunities and lead to better assessments.
