Predicting DUI recidivism: blood alcohol concentration and driver record factors.

This study examined the relationship between blood alcohol concentration (BAC) at arrest, driving history and other demographic factors, and the 1-year post-arrest probability of recidivism for drunk driving (DUI) convictees. Complex and simple prediction models were developed. All models found a statistically significant cubic relationship between BAC and recidivism, reflecting a relatively high rate of recidivism at a BAC of 0.00%, decreasing to a minimum at ca 0.09% BAC, then increasing to another relatively high rate at a BAC of ca 0.29%, followed by a decline in recidivism to BAC levels of 0.35% and beyond. High rates of recidivism at high BACs suggest alcohol dependency, while high rates at low BACs suggest the involvement of other impairing substances. The rate of DUI recidivism for offenders who refused alcohol testing was the same as for aggregated BAC-tested offenders who had prior DUIs at the time of the arrest. The probability of DUI recidivism predicted by a simple model using BAC, prior 2-year traffic convictions, and offender level (first or repeat offender) could be used along with other factors by presentence investigators, judges or in administrative settings to determine appropriate sanctions, treatment or other remedial measures. The findings support the notion that first offenders with high BAC levels and prior 2-year traffic convictions are at as high a risk of recidivating as many repeat offenders, and might therefore benefit from similar sanctions and/or remedial treatment. The findings also support viewing DUI arrestees with very low BACs as probable drug users with relatively high probabilities of recidivating.