Methods: This study adopts a mixed-methods approach combining quantitative and qualitative research methods. Firstly, data from six cities and counties where TIPVDA2.0 was piloted were collected, including 1,931 TIPVDA 2.0 assessment records. Then, among these, 164 cases involving the same victim and the same incident, seeking help from both police and healthcare units within 2 days, were selected for analysis to examine the differences in assessment results between police and healthcare systems. Subsequently, police officers, medical social workers, and nursing staff were invited to participate in one focus group discussion (n=10) and in-depth interviews (n=3) to understand the factors influencing frontline workers' assessments.
Results: There were significant differences in the danger assessment results between police officers and healthcare personnel for the same victim in the same IPV incident. Regardless of which unit the victim sought help from first, the results showed that overall, healthcare personnel gave higher risk assessment scores than police officers, with the differences being most pronounced in items regarding individual sensitivity. The related influencing factors can be categorized into personal and operational factors. The personal factors involve the personal motivations and backgrounds of the assessors and the survivors. The operational factors involve the professional attributes and the approaches assessors adopted.
Conclusion and Implications: This study found that the quality of danger assessment depends on the attitudes of both the assessor and the survivor. Only when the danger assessment is seen by frontline workers as an important process rather than just a workflow, the accuracy and efficiency be improved. On the operational procedure, it is necessary to appropriately integrate risk assessment into the case processing steps and maintain a thorough understanding of the use of danger assessment tools. This study suggests adopting a structural professional assessment approach in practice.