Implementation and Validation of "SER-DesVi": An Empirical System for Assessing the Police Risk of Injury, Suicide and Homicide Cases of Missing Persons

Abstract

This research presents the development and validation of police assessment forms for the risk of injury, suicide, and homicide in cases of missing persons that are part of the SER-DesVi system. To meet the study objectives, 1,111 cases were analysed, divided into some analytical case groups (injury, suicide and homicide, n=260) and a control case group (good health, n=851). A retrospective study was proposed to study the predictive capacity of the system, and a cross-validation model using the K-Fold approach was proposed to study its performance. The analysis strategies focused on studying the ROC curves and the discrimination and calibration parameters. The findings empirically show that SER-DesVi, constructed using a specific set of characteristic indicators, is a useful, valid and effective system for assessing the risk of the missing person suffering injuries or dying by suicide or homicide during the disappearance.