Can statistical data qualify assessments of children at risk?

Every day municipalities across Europe (and beyond) receive notifications about children at risk. The notifications come from teachers, health professionals, social workers, neighbors, or anyone else who sees a child, which appears not to thrive. The assessment and validation of whether the child is actually at risk is complicated and difficult for the individual caseworker for several reasons. First, within a short span of time, the caseworker must decide whether a notification should lead to further investigations or if the case should be closed. Second, the amount of accessible information differs significantly from case to case. Third, the relative importance of the different risk and protection factors is complex and difficult to assess – especially when the social worker has to assess both immediate danger as well as risk of long term failure-to-thrive.
Internationally, different risk assessment tools have been developed to support caseworkers’ decisions based on either ‘caseworker driven’ models (actuarial risk assessments,) or statistical models based on register-based information (predictive risk modelling). In municipalities in the United States, a statistical tool has been used to qualify the assessment done by social workers when they consider how to respond to a notification about a child at risk. Based on theories of risk assessment, the aim of the tool is to inform the assessment made by social workers. The information included in the tool are existing data, meaning data about the child and parents that are already registered in the municipality such as home address and school records.
A similar tool is being developed in a social work research project in Denmark. The idea is to include risk and protection factors such as
information about health, school absenteeism and family circumstances and analyze their correlation to assess the likelihood that a child needs help from the social services. The statistical tool is intended as a supplement – an information-processing tool – to the professional caseworker’s assessment of a notification, and not as a replacement of the professional judgement.