Research

Questions & Answers

1.	a)	Do alcohol interlocks reduce drunk-driving recidivism among offenders?
1.	b)	Do alcohol interlocks reduce recidivism among first time offenders?
1.	c)	Do alcohol interlocks reduce recidivism among hardcore offenders?
1.	d)	Do alcohol interlocks reduce alcohol-related crashes?
2.	a)	How often do judges impose sentences of alcohol interlock?
2.	b)	What are some factors that contribute to judges' reluctance to impose alcohol interlocks as a sentence?
3.	a)	Do offenders comply with orders to install alcohol interlocks?
3.	b)	Once installed, do offenders comply with alcohol interlock devices?
4.		Can the information from the data recording device be used to predict future offenses?
5.	a)	What do participants/offenders think about alcohol interlocks?
5.	b)	What do key stakeholders think of alcohol interlocks?
6.		How does the alcohol interlock impact the participant's family?
7.		What is the relationship between alcohol interlocks and treatment?
8.		What happens when the alcohol interlock is removed from the vehicle?
9.		How do the costs for ordering alcohol interlocks compare to the benefits?
10.	a)	What are the strengths of the existing research?
10.	b)	What are the limitations of the existing research?
11.		What research is still needed?

Downloads: Student Handout - Word file   Student Slides - PowerPoint file

1a) Do alcohol interlocks reduce drunk-driving recidivism among offenders?

• Research shows that while alcohol interlocks are installed in the vehicle, they reduce recidivism among both first offenders and repeat offenders, including hardcore offenders (also known as persistent/chronic drinkers and repeat offenders who repeatedly drive after drinking with extremely high blood alcohol concentrations and are resistant to change this behavior). More than 10 evaluations of interlock applications have reported reductions in recidivism ranging from 35-90% (Voas and Marques 2003; Vezina 2002; Tippetts and Voas 1997; Coben and Larkin 1999) with an average reduction of 64% (Willis et al. 2005).


• A more recent study commissioned by the Centers for Disease Control and Prevention (CDC) that involved a systematic review of 15 scientific studies found that while interlocks were installed, the re-arrest rate of offenders decreased by a median of 67% compared to groups who did not have the device installed (Elder et al. 2011).


• A Swedish study (Bjerre and Torsson 2008) revealed that the frequency of annual DWI offenses decreased by approximately 60% among offenders who completed a two-year interlock program. Similar reductions were found two to four years after removal of the device.

1b) Do alcohol interlocks reduce recidivism among first time offenders?

• Four studies, each with a unique population, different measures of recidivism, and varying evaluation periods, have concluded that alcohol interlocks are effective in reducing recidivism among first-time offenders while the interlock is installed in the vehicle (EMT Group 1990; Morse and Elliot 1992; Tippets and Voas 1998; Voas et al. 1999).


• An evaluation study conducted in New Mexico (2004) compared data from a court ordered interlock group (n=862) with data from a non-interlock group (n=11,973). The findings confirmed that the recidivism rate was significantly lower among first offenders on the interlock. The re-arrest rate per year for interlock participants was 3.51%, while the non-interlocked group had a significantly higher re-arrest rate of 7.09% (Voas et al. 2005).


• An Alberta, Canada study compared first offenders with alcohol interlocks to a control group of reinstated and non-interlocked drivers and found an 89% reduction in recidivism when comparing first offenders with alcohol interlocks to reinstated drivers (Voas et al. 1999). The study found a 95% reduction in recidivism when comparing first offenders with alcohol interlock to ineligible drivers. Ineligible subjects included those individuals who committed a DWI while their license was suspended resulting in an extension of the suspension period.


• A more recent study of New Mexico’s interlock program (Marques et al. 2010) examined the recidivism rate of first offenders arrested for aggravated DWI. The results of this research supported previous findings as offenders who participated in the program had a 61% lower recidivism rate while the device was installed and a 39% lower recidivism rate following the removal of the interlock when compared to counterparts who never installed the device.


• In Washington, first offenders who installed an interlock had a 12% reduction in recidivism when compared to those who did not (McCartt et al. 2012).


• First-time DWI offenders may not be “first” offenders in the true sense of the word. Rather, these are often individuals who have driven drunk repeatedly but were never apprehended. As such, leading researchers agree that the interlock is a sufficient measure to address impaired driving among first offenders, as over 50% of first-time offenders are repeat drunk drivers who have only been arrested once (Beirness and Robertson 2003; Dewey-Kollen and Ellinger 2008)

1c) Do alcohol interlocks reduce recidivism among hardcore offenders?

• Five studies involving chronic or repeat DWI offenders found that alcohol interlocks were one of the most promising strategies to prevent a subsequent occurrence of DWI behavior (EMT Group 1990; Popkin et al. 1992; Morse and Elliot 1992; Jones 1993; Weinrath 1997).


• An evaluation conducted in Maryland used offenders with multiple convictions who were eligible for license reinstatement (Beck et al. 1999). The offenders (N=1,387) were randomly assigned to participate in an alcohol interlock program (experimental group: N=689), or in the post-licensing treatment program (control group: N=689). The alcohol traffic violation rate during the first year was significantly less for participants in the interlock program (2.4%) than for those in the control group (6.7%). Participation in the alcohol interlock program reduced offenders’ risk of committing an alcohol traffic violation within the first year by approximately 65%.


• A study conducted in Illinois examined the recidivism rates of two groups of drivers who had their drivers licenses revoked for alcohol-impaired driving and who received restricted driving permits (Raub et al. 2003). Participants in both groups were chronic offenders and had to meet the rigorous criterion for inclusion in the study. The participants were divided into those who were required to install the alcohol interlock (N=1,408) and those who were not (N=1,384). It was found that 20 (1.3%) of the 1,560 drivers originally in the experimental group were rearrested compared to 94 (8.7%) of the 1,384 drivers in the control group who had been rearrested. In other words, offenders who used the alcohol interlock were one-fifth as likely to be arrested for a DWI during the first year compared to those who did not have the device. The authors also noted that once the alcohol interlock was removed, there was a rapid return to pre-device recidivism.


• A 2007 New Mexico study (Roth) examined the recidivism of 437 repeat offenders who installed interlocks compared to a control group. Only 2.5% of the interlock group was re-arrested for DWI while the device was installed whereas 8.1% of the control group was re-arrested during an equivalent time period. The overall recidivism rate for those who installed the interlock was 22% less than the recidivism rate of those who did not install the device (Marques et al. 2010).


• Rauch et al. (2011) found that offenders who participated in a two-year administrative interlock program in Maryland has a 36% reduction in DWI recidivism during the period the device was installed; there was also a 26% reduction in recidivism two years post-intervention.

1d) Do alcohol interlocks reduce alcohol-related crashes?

• No conclusive research is available that establishes a correlation between the use of alcohol interlocks and the prevention of alcohol-related crashes. When compared to DWI arrests, alcohol-related crashes are an infrequent occurrence and the low rate of interlock installation (approximately 20%) makes it difficult to create a pool of data for analysis purposes (Fieldler et al. 2012).


• Preliminary research by Marques et al. (2010) found that as interlock installation rates increased in New Mexico, the frequency of alcohol-related crashes decreased.  


• A systemic review of literature that was conducted for the CDC’s Guide to Community Preventive Services revealed limited evidence that alcohol-related crashes decrease while the interlock device is installed (Elder et al. 2011).


• Overall, the crash rates for those with interlocks were similar to that of the general driving population but greater than those offenders who drove with suspended licenses (Guide to Community Preventive Services 2011; Bjerre 2005; DeYoung et al. 2005).

2a) How often do judges impose sentences which require alcohol interlock?

• The rate of imposition of alcohol interlocks has been historically low and inconsistent. A California study reported that following the passage of legislation requiring judges to order interlock ignition devices for all drivers convicted of driving on a DWI-suspended driver license, an installation order was part of sentencing for less than one-quarter of repeat DWI offenders (DeYoung 2002).


• Despite these findings, increases in installation rates suggest that progress is being made. One of the contributing factors is likely judicial education and training. If judges are aware of the benefits of the interlock device and the advancements in interlock technology, they are more inclined to enforce interlock orders. Nationally, the number of interlocks installed in the United States increased from 212,300 in 2010 to 279,000 in 2011 (Roth 2012) but this only represents 20% of all offenders arrested for impaired driving annually.

2b) What are some factors that contribute to judicial reluctance to impose alcohol interlocks as a sentence?

• Historical and personal experiences in sentencing offenders may be based on inconsistent findings with respect to whether and to what extent the alcohol interlock is effective and can influence decision-making.


• The interlock device appears too harsh for first offenders and too lenient for repeat offenders (Vanlaar 2005).


• There is concern that the cost to offenders is prohibitive, contrary to existing research which reveals that the cost of this sanction can range from $1.90-$4.00 per day or $60-$100 per month (Robertson et al. 2006; Kanable 2010).


• There are concerns regarding the technological effectiveness of the device, complications surrounding its use, and the amount of effort required to monitor offender compliance.


• Lack of current and comprehensive information and educational materials has resulted in a pattern of decision-making based on out-dated information. This creates challenges for practitioners who lack time to gather and review relevant information (Beirness 2001).


• Enhanced and continuing educational efforts to eliminate misperceptions, myths and misinformation, and other legal concerns can provide a solid basis to overcome judicial reluctance to impose the alcohol interlock as a sanction.

3a) Do offenders comply with orders to install alcohol interlocks?

• Many studies conducted over the past two decades estimate that between 25-75% of suspended or revoked drivers continue to drive (Waller 1985; Hagen et al.1980; Sadler and Perrine 1984; Peck et al. 1985; Ross and Gonzales 1988; Griffin III and De La Zerda 2000).


• The difficulty associated with the detection of unlicensed drivers makes the decision not to install an interlock and instead drive without a license seem like a low-risk option for offenders (Voas et al. 2010).


• Recent research suggests that judges do not consistently impose alcohol interlock sentences and that offender non-compliance is common. A California study revealed that of 775 DWI offenders sentenced to install an alcohol interlock as a condition of probation, 191 offenders (approximately 25%) did not have the device installed (DeYoung 2002). Similar results were found in Florida as only 25.6% of arrested offenders installed a mandatory interlock (Marques et al. 2010).


• Research also indicates that repeat offenders are less likely to install an alcohol interlock compared to first offenders (Voas and Tippetts 1997).


• Offenders fail to install alcohol interlocks for various reasons: to avoid compliance, inconvenience, embarrassment, and cost.


• In order to remedy low participation and compliance rates, offenders can either be offered incentives (e.g., reduced fines, reductions in hard suspension periods) or face a more unpleasant alternative (e.g., house arrest, vehicle impoundment). Compliance can also be increased through the use of close supervision and monitoring of offenders (Zador et al. 2011).

3b) Once installed, do offenders comply with alcohol interlock devices?

• During the first several weeks, offenders often try to circumvent the alcohol interlock by tampering with breath samples and attempting to electronically disengage the device.


• Research indicates that tampering and circumvention rates decrease over time, perhaps as the offender becomes more used to the device and gains an understanding of the sophisticated technology involved and the futility of trying to beat the device (Vanlaar et al. 2010).


• As laws have been strengthened and the level of program monitoring has increased, non-compliance has generally occurred less frequently.


• Some offenders do bypass the alcohol interlock by driving a vehicle that is not equipped with one. However, strong reductions in recidivism found in research suggest this is infrequent.


• Zador et al. (2011) found that non-compliance could be reduced through close supervision, installation of a data logger, the provision of thorough interlock training at the time of device installation, and the provision of information to offenders regarding the consequences of circumvention, refusals, and failed breath tests.

4) Can the information from the interlock data recording device be used to predict future offences?

• Research suggests that a high rate of blood alcohol concentration (BAC) fail readings from the alcohol interlock data recording device, particularly in excess of .02, is predictive of the likelihood of recidivism (Marques et al. 2003; Beirness and Marques 2004).


• Marques and Voas (2008) found that the number of failed BAC tests logged is predictive of repeat DWI offenders. The higher the rate of failed tests, the more likely offenders will recidivate once the interlock is removed. Also, those offenders who are in the top 20-30% of elevated interlock BAC tests have significantly higher levels of alcohol biomarkers associated with problem drinking.


• The presence of elevated BAC tests during early morning hours can also assist in predicting future DWI offenses. The presence of two or more elevated BAC test results during the early morning hours further bolsters the predictive model regarding the likelihood of future DWI offenses (Beirness and Marques 2004).


• Several variables, but primarily more prior DWI’s, more interlock warnings and failures logged during the first five months of interlock usage predict more than 60% of repeat DWI violations, with a false positive rate of less than 10% (Marques et al. 2001).

5) What do participants/offenders think about alcohol interlocks?

• A study conducted in the United Kingdom found that many offenders credited the alcohol interlock for stopping or reducing impaired driving and/or reduced their drinking, invoked serious thought about their drinking habits, and assisted in changing their drinking patterns (Beirness et al. 2007).


• There are two types of concerns that are usually raised by offenders. First, offenders express some concerns regarding social aspects such as the embarrassment, stigma, and frustration of providing a breath sample, particularly when it is inconvenient for the offender to do so (Beirness 2001). Second, some offenders report technical challenges such as the perceived long warm-up time, invalid samples, and the frequency of re-tests (Beirness et al. 2007).


• Findings from offender focus groups in New Mexico revealed similar concerns. However, while offenders had a range of mixed responses to the interlock (from anger to acceptance), there was general recognition that having a driver’s license was worth both the cost and inconvenience associated with the device (Marques et al. 2010).

5b) What do stakeholders think about alcohol interlocks?

• As part of their evaluation of New Mexico’s interlock program, Marques et al. (2010) interviewed key stakeholders in order to learn about their perceptions of the device and its use. Stakeholders noted that the interlock works best as a sanction for ‘honest’ or compliant offenders. They also recognize that the device allows offenders to continue a relatively normal life and separates drinking from driving.


• Stakeholders also recognize that the interlock in and of itself does not have the ability to promote and maintain offender abstinence. As such, treatment and supervision are important.


• Stakeholders pointed out that in the absence of timely reporting practices and graduated sanctions, there is limited accountability for violations and non-compliance (Marques et al. 2010). This leads to frustration on the part of practitioners and results in a reduced level of confidence in the interlock as an effective intervention.


• These concerns can be addressed through education and training as well as the use of specific features such as real-time reporting and the early recall.

 

6) How does the alcohol interlock impact the participant's family?

• Family members were particularly favorable to the use of the alcohol interlock as a measure seen as providing reassurance that the offender is not driving while impaired (Beirness et al. 2007). Families were generally positive about the interlock experience and the impact it had on their loved ones. Many reported a positive change in drinking behavior and the peace of mind provided by having the alcohol interlock installed.


• As part of an evaluation of New Mexico’s ignition interlock program (Marques et al. 2010), focus groups were conducted with offenders to determine the impact that the interlock had on their lives and the lives of their loved ones. Focus group participants reported that both their parents and their children expressed relief when the interlock was installed. Common complaints reported by family members included the financial burden of the device and the inconvenience associated with its use.


• A limited number of family members participated in interviews as part of the Nova Scotia interlock program process evaluation (Robertson et al. 2010). They noted that their loved ones participation in the interlock program had a positive effect. One finding was that continued communication with and training of family members about the interlock device and program may be useful.

7) What is the relationship between alcohol interlocks and treatment?

• Alcohol interlocks serve as a nexus between criminal justice sanctions and substance abuse treatment, by effectively restricting offenders’ driving privileges while giving them the opportunity to learn how alcohol consumption affects behavior (Beirness 2001).


• While treatment for alcohol abuse can be a lengthy process with setbacks and relapses, the alcohol interlock provides a safety net to greatly reduce the likelihood that such relapses do not result in impaired driving (Beirness et al. 1998).


• Alcohol interlocks were never intended as a treatment for alcohol abuse, thus the installation and use of an alcohol interlock device alone cannot be expected to result in a long-term change in the amount and extent of alcohol consumption (Beirness 2001).


• An example of an alcohol interlock program that incorporates treatment within the program framework is the Interlock Enhancement Counselling program in Colorado. All participants in this program must install the interlock and attend cognitive behavioural treatment and motivational interviewing sessions (Timkin and Flavia 2011). The overall goals of this program are to reduce circumvention attempts, increase successful completion rates, and to address alcohol dependency.


• Nova Scotia’s interlock program also includes a rehabilitative component. Participation is voluntary for first-time offenders deemed to be a 'low' or 'medium' risk, as determined by Addiction Services of Nova Scotia through the Alcohol Rehabilitation Program, and mandatory for those who are deemed to be a 'high' risk or are repeat offenders. During their time in the interlock program, offenders must participate in on-going rehabilitation counseling sessions. Counselors also make recommendations to the Registry of Motor Vehicles (RMV) in relation to offenders’ exit from the program.


• Ultimately, the relationship between alcohol interlocks and treatment providers should be strengthened as part of an alcohol interlock program. More research examining the beneficial effect of the integration of rehabilitation and alcohol interlock programs is needed (Beirness et al. 2003).

8) What happens when the alcohol interlock is removed from the vehicle?

• Numerous studies demonstrate that alcohol interlocks have a beneficial impact on recidivism as long as the device is installed in the vehicle. Specifically, existing studies converge at the finding that once the device is removed, recidivism rates return to levels comparable to rates of those who did not have an interlock installed (Beirness 2001; Beirness et al. 1998; Jones 1993; Popkin et al. 1993; Coben and Larkin 1999; Marques et al. 2001; DeYoung 2002; Raub et al. 2003).


• As a consequence of increased recidivism following the removal of the alcohol interlock, several studies have reported that employing alcohol interlocks may be necessary as a long-term or permanent condition of driving for repeat offenders (DeYoung 2002; Rauch and Ahlin 2003; Raub et al. 2003; Beirness et al. 2003).


• More recent studies have begun to note declines in recidivism that are maintained following the removal of the interlock. For example, a Swedish study (Bjerre and Torsson 2008) found that the frequency of annual DWI offenses decreased by approximately 60% among offenders who completed a two-year interlock program. Similar reductions were found two to four years after the removal of the device.


• A New Mexico study found that offenders who participated in the interlock program had a 39% lower recidivism rate following the removal of the device than those offenders who never had the device installed (Marques et al. 2010).


• Rauch et al. (2011) found similar results in Maryland. Offenders who participated in a two-year administrative interlock program had a 36% reduction in DWI recidivism during the two-year intervention and a 26% reduction in DWI recidivism during the two-year post-intervention period.

9) How do the costs for ordering alcohol interlocks compare to the benefits?

• The findings of an overview of cost-benefit analyses demonstrate that new motor vehicle safety features, including the alcohol interlock, are some of the most cost-effective measures and contribute to the largest reductions in the number of fatalities (European Road Safety Observatory 2006).


• The report cites a Norwegian study (Elvik 1999) which notes that installing alcohol interlocks in the vehicles of all impaired drivers would have an estimated benefit-cost ratio of 8.75. This means that for every dollar spent on alcohol interlocks, there will be a saving of nine dollars.


• Roth et al. (2007) estimated that requiring alcohol interlocks for all first offenders would result in the accruement of three dollars for every one dollar spent in program costs.


• A 2011, cost-benefit analysis (Feustel-Seidl) of participation in a rehabilitation program and alcohol interlock program in Germany approximately 30% of subjects indicated that they were able to achieve a higher level of income by EUR 545 (on average) as a result of participation and 40% had lower travel costs (by EUR 165). Moreover, 72% of subjects noted time savings due to participation in the interlock program of almost two hours per day.  


• As alcohol interlocks become more widely used as a sentence, the installation and manufacturing costs will decrease, thus maximizing the benefits associated with the device.

10a) What are the strengths of the existing research?

• The strengths of the existing research include: highly comparable findings from several studies and a convergence of evidence produced by the majority of these studies; sufficient overall sample sizes to allow for broad application of findings; and a wide range of offender status and populations examined as control groups.


• Alcohol interlock research covers a broad spectrum of studies conducted in worldwide jurisdictions in Scandinavia (Norway, Sweden), Europe (Belgium, the United Kingdom), the United States (California, Illinois, Maryland, New Mexico, Ohio, Oregon, West Virginia) and Canada (Alberta, Quebec).


• In addition to investigating alcohol interlocks in private vehicles, the available body of research also includes research on alcohol interlocks installed in commercial vehicles and public transportation.

• The current body of research includes evaluations of court-based and administrative programs, providing comparative material that may be useful for jurisdictions considering which approach to adopt.

10b) What are the limitations of the existing research?

• Control groups: Most of the difficulties surrounding control groups involve the inability and lack of resources to select a control group equivalent to the experimental – interlock – group (e.g., comparable groups of license-suspended impaired drivers). In order to create an appropriate control group a population of previously convicted impaired drivers would be randomly assigned to two groups: one receiving the alcohol interlock and the other one not receiving it (Raub et al. 2003). This can be challenging in offender studies for ethical and legal reasons.


• Selection bias: Studies involving voluntary participants may have a selection bias effect, which means that there are some differences in these participants compared to other eligible offenders which can affect the result. DWI offenders assigned to an interlock program are often those who present the highest risk of reconviction which may result in findings much different than the conclusions drawn from a group of volunteers.


• A particular type of selection bias which needs to be accounted for is the impact of judicial discretion (Coben and Larkin 1999). Experimental design or randomized studies are difficult to execute because judges, for important reasons, hesitate to impose this sanction on a random basis. The sanction severely restricts the activities of half of the offenders in this manner, while not imposing the same or similar restrictions on the other half on the basis of nothing more than the luck of when their case was sentenced. While this limitation is a difficult obstacle to overcome, the differences in experimental and control groups could be addressed in future studies by assigning 100 consecutive subjects to a control group (Fulkerson 2003).

11) What research is still needed?

• Effectiveness of alcohol interlocks in reducing alcohol-related crashes: additional research is needed to determine to what extent, if any, alcohol interlocks have in reducing the occurrence of alcohol-related crashes. Limited evidence supports the finding that these crashes occur while the device is installed. It is difficult to determine the effect that the interlock has on crashes due to the small number of offenders who participate in interlock programs.


• Optimal structure for alcohol interlock programs: such programs may be effective if they operate under either an administrative agency (e.g., motor vehicle administration) or the judicial system.


• Monitoring function for alcohol interlock programs: research must be conducted to determine who should be responsible for monitoring offenders, their compliance, and the performance of the device. This may involve a probation officer based on a judicial program or service providers based on an administrative program.


• Process evaluations of alcohol interlock programs: future research is needed to investigate the details of how an alcohol interlock program currently operates compared to how the program should operate. Evaluations also must be conducted on compliance with sentencing legislation requiring judges to order alcohol interlocks and compliance with court orders for installations by offenders. A 2010 process evaluation of Nova Scotia’s interlock program identified successes and challenges associated with program implementation, program fidelity in relation to the planning process, and the experiences of agencies involved in program implementation. Additional evaluations are needed. 


• Outcome evaluations of alcohol interlock programs: all prior studies of alcohol interlocks are limited in that they relied upon re-arrest rate, an incomplete measure of DWI recidivism, as the sole and primary outcome measure (Coben and Larkin 1999). Future studies should routinely and more carefully examine crash rates, alcohol-related crash rates, death, and injuries as additional outcomes. An ongoing outcome evaluation of Nova Scotia’s interlock program will be completed in 2014.


• Interaction between alcohol interlocks and treatment: research is needed to demonstrate the beneficial effect of the integration of rehabilitation and interlock programs. Evidence suggests that the alcohol interlock effectively incapacitates drivers from operating a motor vehicle while impaired, but the device is not intended to assume a treatment role which would include modifying an offender’s behavior. Further research could help solidify and strengthen the relationship with treatment providers (Beirness et al. 2003).

Updated November 28, 2012