Nicholas J Bishop1. 1. Center for Population Dynamics, Arizona State University, PO Box 873701, Tempe, AZ 85287-3701, USA. Nicholas.Bishop@asu.edu
Abstract
BACKGROUND: This project examined the ability of the popular DUI/DWI offender assessment instrument, the Driver Risk Inventory (DRI; Behavior Data Systems, Ltd., 1985), to identify short-term DUI recidivists in a sample of Floridian DUI offenders who were charged with DUI between January 1st, 2008 and December 31st, 2009. The DRI provides a number of behavioral risk scales, DSM-IV substance abuse and dependence classifications, as well as measurement of demographic and criminal history characteristics. METHODS: Cox proportional hazards models were used to identify the demographic, criminal history, and behavioral characteristics most closely associated with the risk of rapid DUI recidivism. Follow-up analyses including ROC curves were used to further examine the ability of the DRI to identify short-term DUI recidivists. RESULTS: In the final model controlling for all variables, the DRI driver risk scale was the single strongest predictor of rapid DUI recidivism. The DSM-IV substance abuse and dependence classifications were also significant predictors of DUI recidivism. A number of the DRI risk scales and the DSM-IV classifications exhibited significant predictive validity and exhibited sensitivity in identifying recidivists similar to other popular DUI offender assessment instruments. CONCLUSIONS: The DRI provides useful identification of DUI recidivists in a sample able to capture only the most rapid DUI recidivists. The results of this research warrant further examination of the DRI's ability to identify DUI recidivists using longer intervals of time between DUI arrests.
BACKGROUND: This project examined the ability of the popular DUI/DWI offender assessment instrument, the Driver Risk Inventory (DRI; Behavior Data Systems, Ltd., 1985), to identify short-term DUI recidivists in a sample of Floridian DUI offenders who were charged with DUI between January 1st, 2008 and December 31st, 2009. The DRI provides a number of behavioral risk scales, DSM-IV substance abuse and dependence classifications, as well as measurement of demographic and criminal history characteristics. METHODS: Cox proportional hazards models were used to identify the demographic, criminal history, and behavioral characteristics most closely associated with the risk of rapid DUI recidivism. Follow-up analyses including ROC curves were used to further examine the ability of the DRI to identify short-term DUI recidivists. RESULTS: In the final model controlling for all variables, the DRI driver risk scale was the single strongest predictor of rapid DUI recidivism. The DSM-IV substance abuse and dependence classifications were also significant predictors of DUI recidivism. A number of the DRI risk scales and the DSM-IV classifications exhibited significant predictive validity and exhibited sensitivity in identifying recidivists similar to other popular DUI offender assessment instruments. CONCLUSIONS: The DRI provides useful identification of DUI recidivists in a sample able to capture only the most rapid DUI recidivists. The results of this research warrant further examination of the DRI's ability to identify DUI recidivists using longer intervals of time between DUI arrests.
Authors: Thomas G Brown; Marie Claude Ouimet; Louise Nadeau; Jacques Tremblay; Christina Gianoulakis; Sophie Couture; Nathaniel Moxley-Kelly Journal: Psychopharmacology (Berl) Date: 2019-09-11 Impact factor: 4.530