This study examined police officer decisions to make an arrest in intimate partner violence (IPV) cases in a jurisdiction with a discretionary arrest policy that uses risk assessment at the scene of a domestic violence incident.
Police-involved victims of IPV were interviewed at two time-points. Of the 266 cases, the offender was arrested in 105 (39.5 percent) cases. In a logistic regression model, physical violence was the strongest predictor of arrest while risk for homicide was not predictive. Upon examination of the factors that predicted further violence, there was a linear relationship between risk of homicide and the severity and frequency of future violence. Arrest and physical violence at the index offense were not associated with violence at follow-up. The use of risk assessment by law enforcement officers may assist with criminal justice decision-making and referral to appropriate interventions. (publisher abstract modified)