Impact of Ontario's remedial program for drivers convicted of drinking and driving on substance use and problems.

• Author: Stoduto, Gina

Introduction

Remedial programs have been widely implemented to prevent impaired driving and the resulting collisions, injuries, and deaths (Health Canada 2004). Early research suggested they had little beneficial effect or could even increase adverse outcomes (Preusser, Ulmer, and Adams 1978). However, recent research provides more consistent evidence of beneficial effects when remedial programs are used to supplement licensing actions (Mann, Vingilis, and Stewart 1988; Mann, Anglin, Wilkins, Vingilis, MacDonald, and Sheu 1994; Wells-Parker, Bangert-Drowns, McMillen, and Williams 1995; Wells-Parker, Mann, Dill, Stoduto, Shuggi, and Cross 2009; Wickens, Butters, Flam-Zalcman, Stoduto, and Mann 2013). Nevertheless, the evidence for the beneficial impact of these programs is not always consistent, perhaps due to differences in program types and evaluation methods (e.g., Brown and Ouimet 2013; Mann et al. 1988).

In randomized trials, remedial program participation has been shown to influence knowledge, attitudes, recidivism, collisions, and health-related measures (Health Canada 2004; Mann, Leigh, Vingilis, and De Genova 1983; Mann et al. 1988,1994; Wells-Parker and Williams 2002; Wells-Parker et al. 1995). Wells-Parker et al. (1995) conducted a meta-analysis of remedial program evaluations and found that they reduced recidivism by 7-9% when compared to punitive models with no treatment (e.g., license suspension), and 24% for programs that combined education, treatment, and supervision. However, recidivism data lack sensitivity, since many convicted drinking drivers may continue to drink heavily and to drink and drive but not be rearrested or reconvicted (e.g., DeYoung, Peck, and Helander 1997). Furthermore, individuals rearrested but ultimately not convicted are not defined as recidivists.

Problematic alcohol use is a widely noted characteristic of drinking drivers (Health Canada 2004; Lapham, Skipper, and Simpson 1997; Macdonald and Mann 1996; Mann, Anglin, Wilkins, Vingilis, and Macdonald 1993; Brown and Ouimet 2013). A substantial proportion of convicted drinking drivers can be considered problem drinkers or alcohol abusers (Smart and Schmidt 1961; Stasiewicz, Nochajski, and Homish 2007; Vingilis 1983). Measures of alcohol use and problems may identify those who continue to drink excessively and continue to pose a risk for impaired driving, and thus changes in alcohol use and problems over the course of participation in a remedial program may be sensitive and useful measures of remedial program impact.

There is also evidence implicating use of drugs other than alcohol as contributors to collisions (Drummer, Gerostamoulos, Batziris, Chu, Caplehom, Robertson, and Swann 2004; Mann, Stoduto, Macdonald, and Brands, 2008; Stoduto, Mann, Ialomiteanu, Wickens, and Brands 2012; Asbridge, Hayden, and Cartwright 2012). While there is little information on other drug use among convicted drinking drivers, use of drugs other than alcohol among those with alcohol-related problems appears common (Chipman, Macdonald, and Mann 2003; Macdonald, Anglin-Bodrug, Mann, and Chipman 2005). Programs that successfully reduce alcohol and other drug consumption would offer not only health benefits but also reduced risk of alcohol- and/or drug-impaired driving. Several studies suggest that treatment of alcohol and other drug problems reduces post-treatment impaired driving convictions and collisions (Macdonald, DeSouza, Mann, and Chipman 2004; Mann et al. 1995). Thus, evaluating the effects of remedial programs on use of drugs other than alcohol may offer additional important measures of outcome.

To date, few studies have examined remedial program impact on amount of alcohol consumed, and no studies have examined amount of other drug use. Early studies suggested little effect of these programs on alcohol use measures (Mann et al. 1983). However, more recent studies based on evidence-based alcohol treatment interventions, including brief interventions such as motivational interviewing, have reported beneficial effects on alcohol consumption (Wells-Parker et al. 1995; Mann et al. 1988; Nochajski and Stasiewicz 2002; Brown, Dongier, Ouimet, Tremblay, Chanut, Legault, and Kin 2010) and do suggest that these programs can reduce drinking behaviour.

In 1998, Ontario introduced a requirement that all convicted drinking drivers must complete a remedial program based on principles of brief intervention for alcohol problems, called Back on Track (BOT), before their drivers' licenses could be returned following a period of mandatory license suspension. A goal of the BOT assessment is to detect individuals showing elevated levels of alcohol and drug problems and assign them to a longer and more intensive program to reduce hazardous levels of alcohol and drug use (Shuggi, Mann, Flam-Zalcman, Chipperfield, and Nochajski 2006). We report here an evaluation of the impact of program participation on alcohol and drug use and problems measured at assessment and at 6 month follow-up. Because all participants complete an assessment before participation and also complete a 6 month follow-up interview following program participation, BOT provides a unique and valuable opportunity to examine the province level impact of this remedial program.

Methods

Back on Track (BOT) includes an assessment, a brief intervention (an education program or a treatment program), and a follow-up interview 6 months following completion of the intervention. The assessment and follow-ups are conducted at sites across the province. Three scales were used to base the decision for assignment to the education program or the treatment program--the Research Institute on Addictions self-inventory (RIASI) (Nochajski, Miller, Augustino, and Kramer 1995); alcohol dependence scale (ADS) (Skinner and Horn 1984); and drug abuse screening test (DAST) (Skinner 1982). Assignment to the treatment program is based on a threshold score being reached on any of these measures (Mann, Rootman, Shuggi, and Adlaf 2006; Shuggi et al. 2006).

Data were based on 22,277 consecutive participants who completed their assessment on or after 1 November 1999 and completed the follow-up interview by 30 April 2005. Sample demographics were 88% male, mean age 44, 77% employed, 47% married, 72% high school graduates, and median income $20,000-$34,999.

The assessment interview includes self-reported number of drinks per drinking day, number of drinking days, and number of days using cocaine, amphetamines, cannabis, benzodiazepines, barbiturates, prescription opioids, codeine, or tobacco in the 90 days before assessment. The adverse consequences of substance use scale (ACSUS) (Mann et al. 2006) measures negative events related to substance use experienced in the preceding 90 days in the following domains: physical health, cognitive abilities, mood, relationships, aggressive behaviour, school or work, legal and financial (recoded as simple counts for each). These measures were also collected at the 6 month follow-up interview.

Comparisons of number of days of substance use (and for alcohol, the average number of drinks per drinking day), adverse consequences of substance use experienced, and number of users of either alcohol or drugs in the 90 days before assessment and the 90 days before follow-up were conducted using Bonferroni-adjusted (for family-wise error rate) Student's t tests and tests for proportions. Comparisons of the number of participants who changed status from substance users at assessment to non-users at follow-up with the number who changed from non-users at assessment to users at follow-up were conducted using Bonferroni-adjusted (for family-wise error rate) McNemar chi-square...