The objective of this study was to examine the factors used by juries to make their decisions on compensatory damages awarded for physical injuries in automobile crash cases involving driving under the influence of alcohol and to compute a value that juries might award for hypothetical total impairment. This study used data obtained from Jury Verdict Research on drunk driving, non-fatal injury cases occurring in the time period from 1980 through 1990, before tort reform started in several states. To achieve the objective of this study, we used two log-linear regression models, with the logs of award and net award amounts as dependent variables. The award variable is a comprehensive figure that includes punitive damages and medical expenses. To calculate the net award, the latter were subtracted from the award amount. The regression explained 57% of the variance in 323 injury awards for impaired driving.The present value of impairment was the dominant predictor. Alone, it explained 21% of the variance. The type of verdict was also very important. Awards achieved through jury trials were approximately 75 percent higher than those achieved through mediation, arbitration or settlement. Using the model where the log of net award was the dependent variable, the value juries placed on total impairment compared to no impairment was approximately $3.1 million at 4% discount rate and $4.3 million at 2.5% discount rate (in 2005 dollars).