|Home | Contents|
This section presents the major findings of the study and describes the strengths and weaknesses of the research.
The study in Section II on the effect of license suspension in the 3 years before Ohio implemented ALS and VA laws demonstrated that suspended DUI offenders had 38% to 43% fewer DUI convictions and 24% to 35% fewer crashes than did similar DUI offenders who were fully licensed. Before implementing new laws in Ohio, as in many States, a license suspension occurred only after conviction by the court and notification by the BMV. Consequently, initiating a suspension following an arrest was significantly delayed, and a large number of offenders received no suspension. The evidence presented in Section II, points particularly to the significance of the delay in application of the license sanction because the period immediately following arrest, up to the third or fourth month, was a time when additional DUI offenses were most likely to occur.
After implementing the ALS law, 95% of both first and multiple offenders were suspended within a few days of their arrest date, thereby reducing the 4-month delay in imposing this sanction, which had been typical before the ALS law. Further, whereas many as 40% of the first offenders never received a license suspension, 98% of the arrested drivers were suspended after the ALS law. Of particular significance was the fact that the suspension rate following ALS was essentially identical for first and multiple offenders.
Comparison of reoffense and crash rates of DUI offenders before and after implementing the ALS law demonstrated that DUI offense rates were reduced by at least two-thirds for both first and multiple offenders, and moving offense rates and crashes were reduced by half. These large reductions can be attributed only in part to ALS, because of other legislation implemented at the same time as ALS. The other legislation doubled the license suspension period from 3 to 6 months for first offenders and subjected multiple offenders to vehicle immobilization in addition to driver license suspension. Nonetheless, the ALS law clearly had a significant impact in reducing overall offense and crash rates by moving the suspension date forward into the high-risk recidivism period immediately following the arrest and by increasing the comprehensiveness with which license suspension was applied to offenders.
Simultaneously implementing the ALS and the VA laws ensured that multiple offenders were suspended for at least 1 year since the vehicle sanctions were usually applied on the date of arrest or, at the latest, on the date of court conviction and extended for 3 to 6 months. All DUI offenders were suspended during the impoundment period. A comparison of these fully suspended offenders who had their vehicles immobilized or impounded with similar fully suspended offenders who escaped a vehicle action demonstrated that DUI recidivism was reduced while the vehicle was impounded. Further, evidence that reduced recidivism continued to a lesser extent during the period following the release of the vehicle. This is an important finding because it normally might be assumed that, since the offender was fully suspended, impounding or immobilizing the vehicle would have little impact on the offenders driving record. However, there is strong evidence from the research literature that up to 75% of the suspended drivers continue to drive to some extent. It appears from the present results that impounding or immobilizing the vehicle reduces illicit driving, which reduced DUI offenses during both the impoundment and the post-impoundment periods.
This study has a number of significant limitations that should be considered when interpreting the results. First, this research does not attempt to determine whether the simultaneous implementation of the ALS and VA laws had a general deterrent effect on the driving public. This would best be measured by a time series analysis of alcohol-related crashes before and after the implementation of the law. In the present instance, data for the entire State of Ohio was not available. Only the driving records of DUI offenders were obtained from the BMV because the objective of this study was to determine the effect of incapacitation on offenders receiving sanctions because of ALS and VA laws. This study was not designed to study the effect on the public.
Second, vehicle forfeiture, a potentially important sanction, which applied only to fourth DUI and third DWS offenders was not evaluated in this study. With the sample of offenders available, there was insufficient data to determine whether this sanction was effective in reducing offenses and crash involvements of drivers whose vehicles were forfeited. Third, this study is limited to the effect of the immobilization law on DWS offenders whose suspensions resulted from a DUI conviction. DWS offenders whose suspensions were for failure to maintain financial responsibility (insurance payments) were also subject to vehicle immobilization, but these individuals were not included in this study.
A particularly important limitation in the present study is disentangling the effects of the ALS law from the effects of the VA law. Both were implemented on the same date; therefore, a time series does not provide a method of separating their effects. The two laws, however, varied in their application to different groups of offenders in three ways:
1. Members of the general driving population receiving a DUI for the first time were not subject to any of the incapacitating effects of either law unless arrested although they could have been deterred by the fear of being suspended.
2. First DUI offenders following their arrest were affected only by the ALS law, which insured that their incapacitation through license suspension occurred earlier and more comprehensively than had been the case before the law.
3. Multiple DUI offenders and first DUI offenders who were also apprehended for driving while suspended were affected by both the ALS and the VA laws.
Thus, the general deterrent effect of the ALS law could be studied by determining whether the number of first offenders (Group 1) was reduced. There was no evidence of this. The incapacitating effect of the ALS law could be studied using the first DUI offenders (Group 2) by comparing their recidivism rates before and after the law. However, a portion of the reduction in recidivism observed could be the result of the extension from 3 to 6 months of the license suspension period for first offenders. Finally, it was possible to study the effect of the VA law by comparing the multiple offenders (Group 3) whose vehicles were impounded with those who were similarly eligible but received no sanction. Essentially all of the multiple offenders who were subject to vehicle action were suspended for at least 1 year so that both those who received the vehicle sanction and the comparison group were fully suspended. Therefore, the ALS license action was the same for both groups.
A major limitation in the studies reported was the inability to assign the vehicle action penalties at random to eligible offenders. It is clear that the combination of administrative and personal factors that determine sanction versus nonsanction group membership resulted in some differences between groups that may have effected this studys outcomes. For example, the DUI offenders who were in the nonsanction group in Hamilton County were more likely to have a prior DUI offense than the sanction group whose vehicles were impounded (Table IV-10). Though the Cox Regression analysis used prior offenses as a covariate, this procedure may not fully correct for the relationship between priors and recidivism or other factors not measured produced the differences observed.
It was not possible to study the effect of the VA law statewide because of the lack of data on vehicle immobility in the BMV drivers record files. This is a general problem with most State record systemsone that must be remedied if studies of State VA laws are to be fully evaluated. In this case, the observed statewide reductions in DUI convictions for multiple DUI drivers following the implementation of the ALS and VA laws must be interpreted with caution because information on actual immobilization was not available. Another reason for caution is that, based on the Franklin and Hamilton Counties results, only about a quarter to a half of the DUI offenders received a vehicle sanction. Finally, other law and court administrative changes occurred at the same time as implementation of the VA law. Although there was no evidence that these factors effected the number of DUI cases, these or other unrecognized changes occurring at the time the VA law was implemented may have also contributed to the observed reductions in DUI offenses.
An important strength of the current study is the large differences found in the recidivism rates for all three dependent variables: DUI offenses, moving violations, and crashes between offenders before and after the implementation of the ALS and VA laws. The relatively large number of subjects followed over significant periods (up to 2 years) adds to the credibility of the overall results. The Kaplan-Meier Survival Analysis procedure provides a powerful method for making maximum use of the variable exposure periods available for offenders in the before-law and after-law groups. Further, the credibility of the finding of reduced recidivism rates following the implementation of the new laws is strengthened by the evidence that arrests and prosecutions of the general driving public committing first DUI offenses were relatively unaffected by the ALS and VA laws. Finally, the fact that all three dependent measuresDUI offenses, moving violations, and crashesdecreased strengthens the conclusion that the ALS law and, for multiple offenders, the VA law reduced overall driving exposure. The reduction in crashes is particularly impressive given that this measure involved all crashes. Though alcohol is involved in approximately 40% of fatal crashes, its role in less severe crashes is significantly smaller and well under 10% for the routine fender-bender crashes that make up 90% of all crash events recorded on State motor vehicle files (NHTSA, 1997, p. 29).
An important feature of the two county-level impoundment/immobilization studies is that they represent independent replications of the implementation of the same law in two different contexts. In Franklin County, vehicle immobilization served as the major vehicle action; in Hamilton County, only vehicle impoundment was used. Each county had to develop procedures that fit its own application of the State law. In this process, there were differences in the proportion of offenders who received the vehicle sanction penalty. Despite these differences, the effect of the vehicle action on offender recidivism was essentially the same. It reduced offenses while the vehicle was not available to the offender and, to a lessor extent, following the vehicles return. This independent replication of the study in the two localities strengthens the confidence in these results. Moreover, the fact that each community implemented the law in a different manner suggests that the results can be extrapolated to other locales, which may apply still other procedures in implementing an immobilization or impoundment law.
The results of the two-county study are further strengthened by the outcome of the ALS study that demonstrated that, after the ALS law was in place, 99% of second and third DUI offenders were suspended. Thus, both the offenders who had their vehicles impounded and the comparison group of offenders who escaped this sanction were fully suspended during the period of the vehicle action study. Therefore, license status appeared not to be a factor in the observed differences between the offenders who suffered a vehicle penalty and the comparison group of similar offenders who escaped the vehicle action. It also provides an indication that the effect of this VA sanction is in addition to the effect of the ALS sanction.
As expected from the results in the two county-level study, there appears to be a statewide reduction in DUI offenses involving suspended multiple offenders who were at-risk or actually experienced a vehicle sanction following implementation of the VA law in Ohio on September 1, 1993 (Figure III-3). The reduction was greater for the multiple offenders with more priors who faced more serious vehicle sanctions, but the difference was not significant.
In this study, a vehicle sanction was coupled with license suspension. Thus, the offense reductions demonstrated generally occurred to offenders who were suspended and should not have been driving. Thus, this study demonstrates that vehicle impoundment or immobilization reduces the risk to the public over and above that achieved by license suspension alone. Driver license suspension was shown to reduce offenses and crash involvement of convicted drinking drivers by 25% despite the fact that many continue to drive. The use of a VA law seems to provide an additional margin of public safety beyond license suspension. The combined effect of the two sanctions appeared to reduce the reoffense rate of DUI offenders compared to the traditional court-ordered suspension system by more than 50%.
The evidence that the effect of impoundment and immobilization on DWS and DUI offenses may persist beyond the length of the penalty period is particularly significant. It is not clear whether this is a deterrent effect (the cost and inconvenience of having their vehicles impounded or immobilized is so painful that offenders are motivated to avoid being caught again) or whether it is an incapacitation effect. Incapacitation could result from the lack of a vehicle because some offenders whose vehicles were sanctioned did not reclaim them, possibly because the vehicle was worth less than the towing and storage costs. Those drivers whose apprehension resulted in the impoundment or immobilization of an employers or spouses vehicle may have also been denied access to it after it was released (9). In contrast, there is no evidence that the incapacitation effect of license suspension extends beyond the time the drivers license is reinstated (McKnight & Voas, 1991). The finding that the effect of the vehicle sanction persists beyond the penalty period is unexpected. If future studies confirm this finding, this effect would significantly add to the value of vehicle sanction programs. Overall, preventing the use of the vehicle by the offender for 1 to 6 months appears to be a promising sanction for DWS and DUI offenses. It assists in enforcing driver license suspension, reduces drunk-driving offenses, and protects the public from these high-risk drivers.