Using behavioral triage in court-supervised treatment of DUI offenders
Effectiveness bank home page. Opens new windowResearch analysis

This entry is our analysis of a study added to the Effectiveness Bank. The original study was not published by Findings; click Title to order a copy. Free reprints may be available from the authors – click prepared e-mail. The summary conveys the findings and views expressed in the study. Below is a commentary from Drug and Alcohol Findings.

Links to other documents. Hover over for notes. Click to highlight passage referred to. Unfold extra text Unfold supplementary text
Copy title and link | Comment/query |

Using behavioral triage in court-supervised treatment of DUI offenders.

Carey S.M., Allen Theresa H., Einspruch E.L. et al.
Alcoholism Treatment Quarterly: 2015, 33, p. 44–63.
Unable to obtain a copy by clicking title? Try asking the author for a reprint by adapting this prepared e-mail or by writing to Dr Carey at

From California, the first evaluation of a system which escalated drink/drug drivers to treatment if they failed a less intensive sentence found significantly reduced recidivism and accidents, and evidence that injuries related to accidents also fell.

Summary The featured study evaluated the impact of the introduction of a drug court model for dealing with drink/drug-driving offenders in a county in California in the USA who were not complying well with more usual enforcement-based procedures.

Key points icon

Key points
From summary and commentary

From California, the first evaluation of a system which escalated drink/drug drivers to treatment if they failed a less intensive sentence found significantly reduced recidivism and accidents, and evidence that injuries related to accidents also fell.

It is unclear whether the findings were due to the substance use treatment which the minority of offenders who did not respond to monitoring and sanctions were required to attend, or the greater accountability enforced on the majority of offenders not sent for treatment.

Strengths of the study include complete sampling of offenders; the main weakness is the lack of a comparison area.

In a typical drug court programme participants are closely supervised by a judge supported by a team of professionals who operate outside their traditional adversarial roles, including addiction treatment providers, parole and probation officers, and court officials and lawyers. Generally, there is a standardised treatment programme and often regular and frequent testing to identify substance use. This model has been extended to other populations and purposes including driving-under-the-influence offenders, and has been shown to reduce recidivism and taxpayer costs.

Before 2008, in San Joaquin County in California drug- or alcohol-dependent driving-under-the-influence offenders were seldom referred to treatment, and offenders were not held accountable for completing their probation conditions. Instead, many returned to illegal activity and were eventually reconvicted. Statewide sanctions for repeat offenders include license suspension, installation of a breath-testing device on the offender’s vehicle to prevent driving under the influence of alcohol, and a requirement to attend an 18- or 30-month driving-under-the-influence education programme.

In 2008 San Joaquin county additionally implemented a system requiring all repeat offenders in the largest judicial district to participate in a driving-under-the-influence court monitoring program. Many are not dependent on alcohol or other drugs and do not need the high level of supervision and treatment that would be appropriate for high-risk/high-need offenders in the typical drug court model. For this reason, there were two tracks:
• Track 1 was the “accountability track”. Participants are required to come to court at one month, six months, and a year to report on progress in completing the terms of their probation and must be compliant with court and motor vehicle department requirements (for a second offence within 10 years, generally including license suspension for at least a year). No treatment is required. If participants are successful, they graduate, are not required to attend court again, and qualify for their license to be returned.
• Track 2 was the “treatment track” for offenders who demonstrated through their behaviour that they are unable to comply with track 1 requirements and have been assessed as needing drug/alcohol treatment. This track more closely follows a drug court model. A driving-under-the-influence court team (judge, lawyers, a treatment representative, and probation officers) meets before court sessions and decides on incentives and sanctions. Track 2 participants attend court weekly during the time they are assigned to treatment and must be compliant with court, motor vehicle department, and treatment requirements throughout the remainder of their sentence (a minimum of a year).

The system used to allocate between the tracks is a version of ‘behavioural triage’, also known as ‘stepped care’. This reserves scarce treatment and supervision resources for offenders who demonstrate a clear need for those services and avoids the cost of conducting risk and need assessments; rather than assessment results, the offenders’ behaviour dictates the intensity of their treatment and supervision.

Under the new system, nearly all driving-under-the-influence offenders started in track 1. Abstinence was required by their probation orders, so repeated positive alcohol or drug tests and other substance-related violations was viewed as evidence that participants have a serious substance use problem requiring more intensive treatment and court monitoring.

This system was introduced in 2008 for all individuals with a new driving-under-the-influence conviction in San Joaquin’s Stockton Judicial District who had at least one previous such conviction within the last 10 years. For the 18 months after their current offence, data on their progress was obtained from the court and from the motor vehicle department. This was compared with the same data over the same time period from a comparison group of similar offenders convicted in the city of Stockton of driving under the influence during 2006, before implementation of the new system could have affected their progress. Most of these earlier offenders only met with a probation officer if they violated probation; to have their licenses returned, they were required to pay for and participate in an education programme organised by the motor vehicle department.

To help ensure comparability, results were adjusted for all the demographic data available including the number of driving-under-the-influence convictions in the past 10 years, age, and gender.

Main findings

Over the study period, under the new system track 2 was used for about 1 in 8 offenders.

Over the 18 months after the driving-under-the-influence conviction which led them to be included in the study, offenders processed in 2008 under the new system consistently showed significantly less evidence of recidivism and greater compliance with court and motor vehicle department requirements. Differences were often small because recidivism was rare, but might still cumulate to significant benefits.

Compared to their predecessors, offenders processed in 2008 recorded fewer new driving-under-the-influence convictions (average 0.097 versus 0.128) and non-significantly fewer were reconvicted (9% v. 12%). They also averaged significantly fewer alcohol- or drug-involved accidents (0.012 v. 0.025 – 14 accidents over 18 months versus 32), and fewer than half as many (1.1% v. 2.3%) were involved in such accidents. Offenders processed under the new system were also significantly less likely to be involved any type of motor vehicle accident – 3.9% versus 7.0% of their predecessors.

The new system was also associated with half as many offenders losing those driving licences through suspensions or revocation – 1.7% versus 3.8% – and over twice as many had their licenses reinstated – 19% versus 8%. Compliance with court attendance requirements also improved.

Statistically non-significant but promising findings included a reduction in the average number of accidents which resulted in injuries, down from 0.028 to 0.017, and just 0.2% of offenders processed under the new system later had an accident associated with a drink or drug driving conviction compared to 0.8% of their predecessors. Though few, these incidents impose a high cost on citizens and pose a severe threat to public safety.

A remaining question was whether offenders escalated to track 2 really were higher risk than those left in track 1. Track 2 participants were four times more likely to have been convicted of a new drink- or drug-driving offence and 3.5 times more likely to have been in an accident involving drug or alcohol use. Track 2 participants were also three times as likely to have been in an accident which resulted in an injury and half as likely to have had their licenses reinstated.

The authors’ conclusions

The San Joaquin model showed substantial promise for increasing public safety by helping prevent impaired driving and reducing traffic accidents. Compare to their predecessors, in the 18 months following their conviction and subsequent entry into the two-track programme, offenders had significantly fewer new drink- or drug- driving convictions and accidents, and were significantly more likely to comply with court, probation, and motor vehicle department requirements and to regain their licenses. In addition, their records showed that track 2 participants do indeed have poorer outcomes than those left in track 1, suggesting that the triage system appropriately divided the offender caseload. However, it remains an open question whether outcomes improved due to more intensive court supervision, more intensive treatment, or better service matching via the two-track system.

Several of the practices implemented in San Joaquin from 2008 can be identified as ‘best’ practices likely to have contributed to the results:
Collaboration The team dealing with drink- and drug-driving offenders was drawn from a range of agencies in the judicial, treatment, and legal sectors, as well an agency which aims to raise awareness of drink- and drug-driving among Spanish speakers. Time spent communicating across agencies and sharing information about participant progress probably helped the programme respond quickly to participant behaviour and develop creative, effective strategies.
Rapid access to treatment Time from arrest to drug court entry was about 30 days, within the optimal time-frame associated with positive outcomes. Prompt programme placement and treatment has been shown to help reduce recidivism.
Stability and ‘clean’ time The longer clients are required to be abstinent before graduation, the longer they stay abstinent and the more positive their outcomes. In San Joaquin participants had to test free of substance use for at least 120 days before they could successfully exit the programme.
Availability, but little use, of jail as a sanction Although an available sanction, imprisonment was rarely used, and then minimally. Jail is an expensive resource that often removes the participant from positive influences (support systems, treatment, employment) and gives them access to negative influences (eg, other offenders and substance users). Using other sanctions whenever possible helps build skills and reduce recidivism.
Consistency of judicial leadership The current judge in the driving-under-the-influence court had been presiding for almost five years at the time of the study. When judges have served for at least two years, experience and longevity have been found to correlate with better participant outcomes and greater cost-savings.

Findings logo commentary The study’s findings strongly suggest that improved monitoring, and/or treatment for offenders who do not respond well to such monitoring, can reduce the danger posed by drivers affected by substance use and possibly also reduce overall social costs due to recidivism and accidents. San Joaquin’s programme can be seen as another instance of the observation that many problem substance users respond well to consistently applied but minor sanctions and do not require treatment, a proposition explored in an Effectiveness Bank hot topic. The essential elements seem to be sufficient leverage in the form of something the subjects of the programme do not want to lose or wish to gain (in this case, restoration of license to drive and avoidance of more severe sanctions), and procedures which enable non-compliance to rapidly be identified and followed by sanctions which though aversive, are not initially so great as to prevent the subject getting back on track.

An important strength of the study was its inclusion of all offenders in the relevant category processed during the time periods being compared. This was possible due to reliance on routinely collected data rather than recontacting offenders for follow-up interviews. However, it also meant that few variables were available with which to ensure the offenders from 2006 and 2008 were similar, or to adjust the outcomes if they were not.

It should also be emphasised that the entire study was about repeat offenders, the ones considered suitable for the option of a drug court-type programme. Before journal publication, findings of the featured study were followed by further state funding for the San Joaquin driving-under-the-influence court. In the news report, the presiding judge said the decision to target repeat offenders was seen as a way reduce alcohol-related fatalities.

Perhaps the main question over the study’s findings is the lack of a comparison area which had not implemented the two-track system against which to benchmark the improvements in San Joaquin. Without this, the possibility cannot be eliminated that something happened more broadly in areas like San Joaquin which reduced recidivism among repeat drink- or drug-driving offenders, such as greater vigilance by traffic police or more intense, widespread or effective public or alcohol trade discouragement of driving under the influence.

Offenders were moved between tracks 1 and 2 in a way which made it difficult to establish how much of each track an offender or offenders in general experienced, and therefore difficult to say which track contributed most to the findings. Reinforcing the authors’ caution about what caused the recidivism reductions is a review of evaluations of drug courts, the courts used as a model for San Joaquin’s track 2. In respect of the courts which specialise in drink-driving offences, across all relevant trials these courts were associated with reduced recidivism, equating to 50% of offenders reoffending after normal procedures, but only 38% if they had been tried at a drink-driving court. However, much of this presumed effect was found in methodologically weak trials; findings from the most rigorous, randomised trials were ambiguous. If drink-driving courts are not an advance on usual procedures, then the finger points more at monitoring and sanctions as the elements which led to better outcomes in San Joaquin.

Findings such as those in San Joaquin suggest a possible way forward for the UK, which has relied largely on drink-driving education programmes to help prevent recidivism among drink-driving offenders on probation. A prominent example of these courses could not be shown to have reduced drink-driving reconvictions, even among participants who had completed the course. To regain their licences, high-risk offenders including those in the same category as in the featured study – convicted of at least two drink-driving offences within 10 years – must additionally convince a doctor appointed by the driving licence authority that they are no longer risky drinkers, a decision based largely on blood test results indicative of regular heavy drinking. This procedure exemplifies an all-or-nothing sanction after repeated heavy drinking which studies such as the featured study have led some US authorities to replace with incremental and minor sanctions imposed along the way to rectify behaviour before an all-or-nothing sanction becomes necessary.

Last revised 27 July 2017. First uploaded 17 July 2017

Open Effectiveness Bank home page

Top 10 most closely related documents on this site. For more try a subject or free text search

HOT TOPIC 2016 Don’t treat, just test and sanction

REVIEW 2011 A new paradigm for long-term recovery

DOCUMENT 2012 Will intensive testing and sanctions displace treatment?

STUDY 2010 South Dakota 24/7 Sobriety program evaluation findings report

REVIEW 2012 Assessing the effectiveness of drug courts on recidivism: a meta-analytic review of traditional and non-traditional drug courts

STUDY 2009 The 24/7 Sobriety Project

STUDY 2010 Is 24/7 Sobriety a good goal for repeat driving under the influence (DUI) offenders?

STUDY 2007 Analysis of South Dakota 24-7 Sobriety program data

STUDY 2012 The effect on reconviction of an intervention for drink-driving offenders in the community

REVIEW 2006 Motivational arm twisting: contradiction in terms?