ࡱ> p5@ 3bjbj22 Z'XXJFFF*p%^4^4^4 >hhh8D, >(D"ft[<$&&&&&&$RPJ^4?ɷ"??J33_?@$3R^4$?$ʅ | <4"^4  .]hdH4u083s 3 r4f:<*00333^4 $3&YE*JJ > >dZN > > MADD CANADA MODEL 2003  TABLE OF CONTENTS INTRODUCTION 2 SECTION I: LICENSING 9 (a) Conditional Licensing INNOVATION 9 (b) Minimum Age for Beginning to Drive 12 (c) Graduated Licensing Program 13 (d) A 0.00% BAC Limit for all Drivers Under 21 INNOVATION 20 (e) Enforcement of the Graduated Licensing Program and 0.00% BAC Limits 26 SECTION II: POLICE ENFORCEMENT POWERS 28 An Introduction to the Federal Enforcement Powers 28 Evidence that the Current Federal Law is Ineffective 33 (a) Authority to Stop Vehicles 35 (b) Authority to Demand Documentation 38 (c) Authority to Establish Systematic Sobriety Checkpoints 40 (d) Authority to Demand an Approved Screening Device (ASD) Test from Drivers at Sobriety Checkpoints INNOVATION 42 (e) Authority to Use Passive Alcohol Sensors 45 (f) Authority to Demand Standardized Field Sobriety Testing 51 (g) Authority to Demand A Drug Classification and Evaluation Test INNOVATION 56 (h) Authority to Demand Blood, Urine and Saliva Samples for Drug Testing INNOVATION 61 (i) Authority to Demand Samples from Drivers in Fatal or Personal Injury Crashes INNOVATION 63 SECTION III: LICENCE SUSPENSIONS AND REVOCATIONS 70 A Summary of Current Licence Suspensions in Canada 70 (a) 24-Hour Administrative Licence Suspensions 72 (b) 30-Day Administrative Licence Suspensions for Driving with a BAC of 0.05% or Higher INNOVATION 73 (c) 90-Day Administrative Licence Suspensions 76 (d) Automatic Provincial Licence Revocations for Criminal Code Offences 78 SECTION IV: VEHICLE SANCTIONS AND REMEDIAL PROGRAMS 81 (a) Alcohol Interlock Program 81 (b) Vehicle Impoundment and Immobilization 86 (c) Vehicle Forfeiture 89 (d) Licence Certification Program 90 (e) Remedial Programs 91 CONCLUSION 94 INTRODUCTION The MADD Canada Model 2003 updates and expands the model that was produced for Rating the Provinces 2000. That initial study affirmed the importance of provincial and territorial initiatives in reducing impaired driving. In the interim, MADD Canada undertook a comprehensive review of the federal impaired driving laws and outlined its priorities for reforming the Criminal Codes impaired driving provisions. Now, we are again focusing on the provinces, looking to them to introduce new measures. Although much work remains to be done at the federal level, the provinces have the constitutional power to enact legislation that would significantly reduce impaired driving and enhance traffic safety. In fact, almost all of the most promising initiatives in recent years have been introduced at the provincial level. MADD Canada is generally pleased with the progress the provinces have made since we released Rating the Provinces 2000. Many of our recommendations have been implemented, particularly with respect to graduated licensing, provincial suspensions for federal impaired driving offences, alcohol interlocks, vehicle impoundment, alcohol and drug assessments, and treatment. While some provinces have made major strides, others have done very little. More importantly, research from Canada and abroad indicates that all the provinces can do far more to reduce impaired driving within their jurisdictions. As MADD Canadas first comprehensive review of the provincial law, Rating the Provinces 2000 rightly focused on incremental changes, drawing largely from best practices in Canada. In this report, we have consciously been less conservative, concentrating on initiatives that hold the greatest promise for substantial improvements. This will allow the provinces to take greater advantage of the successful measures that have been introduced by the world leaders in traffic safety and other modern democracies. As with its predecessor, the MADD Canada Model 2003 has been drafted to reflect Canadas unique division of constitutional powers between the provincial and federal governments. What is essential to note for immediate purposes is that the provinces have very broad constitutional authority over licensing, enforcement, vehicles, and related issues affecting traffic safety. While the MADD Canada Model 2003 has been designed to accord with the values of the Canadian Charter of Rights and Freedoms, some Charter litigation is an inevitable effect of any legislative reform. MADD Canada intends to provide a detailed analysis of the potential Charter challenges in a subsequent study. Nevertheless, our preliminary analysis suggests that the Charter would not preclude the implementation of any of the recommended measures, if they were carefully drafted and included appropriate safeguards. Some of the recommendations in the 2003 model mirror those that we made to the federal government in 2001. Due to their overlapping spheres of legislative authority, both the federal and provincial governments can implement these provisions. MADD Canada is not primarily concerned about who enacts these measures, only that they are enacted. While federal legislation would provide uniform standards throughout Canada, it is inappropriate for the provinces to wait for federal leadership when they have the power to reduce impaired driving within their jurisdictions without delay. The MADD Canada Model 2003 focuses on a narrower set of issues than its predecessor. First, the 2003 model does not encompass victims issues and insurance matters that warrant a detailed and independent review, which is currently scheduled for 2004. Second, even within the traffic safety field, the 2003 model concentrates on three areas, in which Canada has demonstrably lagged behind other jurisdictions: the protection of young drivers; the enactment of effective and efficient police enforcement powers; and the development of policies to deal with persistent drinking drivers and repeat offenders. Canada has taken fewer steps to protect young drivers than many other countries, particularly the United States. For example, every American state has a minimum drinking age of 21, and a zero or low blood-alcohol concentration (BAC) limit for drivers under this age. Increasing the drinking age alone was reported to have saved 20,970 lives between 1975 and 2001. Similarly, the zero and low BAC laws have significantly reduced deaths and injuries among this age group. Further, Australia and most European countries have adopted a higher minimum driving age than the Canadian provinces. Nevertheless, given that there may be little political support for raising the minimum drinking and licensing ages in Canada, we have considered other measures to safeguard young drivers and their passengers. We have accordingly emphasized comprehensive graduated licensing programs and 0.00% BAC limits for all drivers under 21. These initiatives are essential to protect Canadian youth who, despite the progress that has been made, remain far more vulnerable than other drivers to alcohol-related and other crashes. In addition, there is a significant enforcement gap between Canada and most modern democracies. Relative to their counterparts in Australia, New Zealand, Europe and even the United States, Canadian police have far more limited powers to detect, apprehend and gather evidence of impaired driving. These other countries have given police statutory authority to, among other things, use passive alcohol sensors, conduct standardized field sobriety tests, demand breath samples at random from drivers, and collect breath or blood evidence from drivers involved in crashes. In contrast, Canadian police are left to struggle with complex technical requirements and time constraints, and their evidence is often discounted or ruled inadmissible by the courts. Faced with these difficulties, it is not surprising that Canadian police have a charge rate for impaired driving per licensed driver that is only 42% of that of their American colleagues. In the MADD Canada Model 2003, we have addressed these deficiencies by giving Canadian police the powers they need to detect impaired, suspended and unlicensed drivers, and to ensure that offenders are charged and, where appropriate, convicted. Finally, we have recommended a number of measures to deal with persistent drinking drivers and repeat offenders. MADD Canada remains a strong advocate of licence suspensions as a means of deterring and incapacitating impaired driving offenders. However, given that a large percentage of offenders continue to drive at least occasionally during their suspension periods, it is clear that licence suspensions alone are insufficient. Moreover, the over-representation of repeat offenders in alcohol-related crashes indicates that current programs are not meeting their objectives of deterring and rehabilitating first offenders. As a result, we propose a range of initiatives that will reduce impaired driving among these drivers and address their underlying alcohol problems, including: increased and more effective licence checks; impoundment and forfeiture legislation; comprehensive alcohol interlock programs; and greater use of mandatory assessment and treatment. As with the 2000 model, we have tried to focus on pragmatic, achievable changes. Such incremental changes are more likely to receive widespread public and political support, and be administratively and financially feasible. MADD Canada also recognizes that innovative changes are required if we are to make significant progress in reducing the incidence of impaired driving. Consequently, in an effort to balance the need for both incremental and innovative changes, the MADD Canada Model 2003 includes two categories of recommended policies standard practices and innovations. The first category encompasses standard practices that we believe are essential elements of all provincial highway traffic legislation. These recommendations are eminently achievable and can be easily integrated into existing provincial legislative schemes. The second category includes innovative programs that will require a greater change in attitudes and legislation, but are essential if the provinces are to make major gains. These programs are no less important simply because they are innovative, rather than standard. MADD Canada fully supports all of its recommendations. However, the distinction between standard and innovative provisions is important in the 2003 rating of the jurisdictions, in that the provinces were only assessed on the standard programs. This reflects the fact that the innovative programs have not yet been adopted by any jurisdiction in Canada. The MADD Canada Model 2003 is based on the same underlying principles as the 2000 model, although the emphasis has shifted in some areas. We still believe that traffic authorities should be empowered to take pro-active steps to prevent crashes before they occur, rather than simply reacting afterwards by sanctioning those responsible. To this end, we have recommended significant additions to police detection and enforcement powers, which will greatly increase the probability that unlicensed, suspended and impaired drivers will be apprehended, charged and convicted. Consequently, these changes should dramatically increase the deterrent impact of the law. Further, MADD Canada continues to advocate the use of administrative, rather than criminal, proceedings to govern licensing issues. Administrative procedures are typically more streamlined, cost effective and efficient. They also tend to focus on public safety, rather than on the offenders moral blameworthiness. In addition, administrative procedures allow provincial and territorial licensing authorities more flexibility to develop creative solutions to deter impaired driving and rehabilitate repeat offenders. The 2003 model reaffirms MADD Canadas ultimate focus on public safety. Impaired driving remains the leading criminal cause of death and a leading criminal cause of injury in Canada. The situation is particularly dire among those aged 16 to 24, for whom traffic crashes remain the leading cause of death. We believe that the provinces must improve the apprehension, conviction, and rehabilitation of impaired driving offenders, so that fewer Canadians are needlessly killed or injured in the future. The following report provides the provinces with legislative suggestions for achieving these important goals. SECTION I: LICENSING (a) Conditional Licensing INNOVATION As will be described in greater detail in the police enforcement section, the current breath and blood-testing provisions of the Criminal Code are not achieving their intended goals. The vast majority of impaired drivers go undetected, and a considerable number of those who are detected are not charged because of the time-consuming and often frustrating process of gathering the necessary evidence. Moreover, many of the impaired drivers who are charged succeed in negotiating to plead guilty to a provincial offence or in escaping penal liability altogether by challenging the breath or blood evidence. Perhaps even worse, only a very small fraction of impaired drivers who kill or injure others are charged with, let alone convicted of, the more serious offences of impaired driving causing death or bodily harm. This is due, in large part, to the difficulties police encounter in gathering evidence of the drivers BAC following a crash. Nevertheless, proposals to strengthen the Criminal Codes breath and blood-testing provisions are often fraught with constitutional concerns. Section 8 of the Charter guarantees the right to be free from unreasonable search and seizure. The courts have interpreted this section as generally requiring authorization based on reasonable and probable grounds before a search or seizure can be conducted. As a result, Criminal Code amendments, such as those authorizing the police to demand breath samples from all drivers involved in crashes, may be seen as violating section 8. If the police have no specific reason to believe that a driver is impaired, other than the fact that a crash occurred, they will be unable to lawfully demand breath or blood samples. This is particularly problematic if the driver is injured and taken to hospital, because the police will have very little opportunity to detect and prove that there were signs of impairment. To alleviate these concerns and improve apprehension, detection and conviction rates, MADD Canada recommends that every province establish a system of conditional licensing. Pursuant to this system, it would be a mandatory and express condition of every licence that the driver consents to providing samples in two situations. First, a driver would be required to provide a breath sample for analysis on an approved screening device (ASD) when requested to do so by an officer at an established sobriety checkpoint. Thus, the police would be able to conduct what is known as random breath-testing (RBT). This provision would have a general deterrent effect, and greatly improve the apprehension and detection rates of impaired drivers. Drivers who fail to provide a breath sample without a reasonable excuse should be subject to a mandatory 90-day administrative licence suspension. Moreover, the failure to provide a sample should be made a provincial offence. Second, a driver would be required to provide a breath sample when requested to do so, if the police reasonably suspected that he or she was involved in a fatal or personal injury crash. Drivers who fail to provide a sample, without a reasonable excuse, should be subject to an administrative licence revocation. The failure should also constitute a serious provincial offence. If the driver is incapable of providing a breath sample, or breath testing cannot be conducted for any other reason, the police should be authorized to demand a blood or urine sample. Furthermore, if the suspect is unable to respond to the officers demand due to illness, impairment or injuries suffered in the crash, the officer should have express authority to have a blood, saliva or urine sample taken by medical personnel. These other testing provisions are necessary to ensure that drivers do not escape liability simply because they require or demand to be taken to the hospital. Since these provisions allow for the taking of bodily samples without reasonable grounds to believe that a particular driver is impaired, they will likely be found to infringe section 8 of the Charter. Given these Charter concerns, it is essential that the proposed powers be carefully defined, contain strong administrative safeguards and include a right to seek an internal review. MADD Canada believes that, in these circumstances, breath screening at sobriety checkpoints and mandatory testing of suspected crash-involved drivers would be upheld under section 1 of the Charter as reasonable limits that are demonstrably justified in a free and democratic society. The Canadian courts have repeatedly held that driving is a privilege. As a result, the provinces have considerable freedom to regulate driving, which includes imposing conditions on licences. Potential drivers should be informed that, as a condition of obtaining a licence, they will be required to provide a breath sample on an ASD if stopped at a sobriety checkpoint, and to provide a breath sample if suspected of involvement in a fatal or personal injury crash. They should also be informed of the consequences for failing to comply with the police demand. In order to ensure that these express conditions are brought to the drivers attention, they should be included on the licence and licence renewal forms. They may even be printed on the drivers licence. Thus, drivers would have clear advance notice of the requirement to provide these samples. The proposed conditions would make the licensing of drivers similar to other forms of statutory regulation, where individuals involved in certain activities agree in advance to submit to random searches, audits and other inspections. These individuals are aware that such intrusions are part of the regulated activity. They can choose to be subject to these intrusions or to refrain from that activity. For instance, airline passengers agree to submit to screening, and personal and baggage searches as a precondition to boarding a flight. Those who do not wish to be searched are unable to fly. Similarly, drivers currently agree to certain conditions as part of the licensing system, such as the duty to stop and provide evidence of insurance when involved in a collision. The proposed new forms of conditional licensing will simply add to the existing requirements of a drivers licence. By limiting the breath screening power to established sobriety checkpoints (rather than routine patrol activities), the legislation would address potential concerns about arbitrariness or the targeting of certain drivers. Arbitrariness would also be minimized by requiring all drivers involved in fatal or personal injury crashes to provide samples. This universal testing would have the additional benefit of improving the inclusiveness of alcohol-related crash statistics in Canada. (b) Minimum Age for Beginning to Drive The minimum age for entry into the graduated licensing program should be no lower than 16, regardless of whether the applicant participates in a driver education program. Research indicates that the minimum driving age affects the crash rate. For example, Victoria has the highest licensing age (18) of all Australian states and the lowest rate of fatal and personal injury crashes per licensed driver. Similarly, New Jersey has the highest licensing age (17) of American states and the lowest relative risk of crash for young drivers. Conversely, Delaware has the highest relative crash rate, and a minimum licensing age of 15 years and 10 months. Moreover, 58% of teenagers surveyed in Delaware admitted to driving on public roads before obtaining a drivers licence. Only 35% of teenagers in New Jersey reported such illegal driving. Thus, a higher minimum licensing age not only improves the safety of beginning drivers, but also appears to discourage unlicensed driving by younger teenagers. To accommodate young people who work on family farms, some provinces may wish to have a limited exception to the minimum licensing age for those operating farm equipment. However, the exception should be narrowly defined and should not allow youth under 16 to drive machinery on lengthy stretches of major roads. (c) Graduated Licensing Program Research from Ontario, Nova Scotia, Quebec, and abroad has established that graduated licensing programs reduce crash deaths and injuries among young and beginning drivers. These studies also suggest that both age and inexperience are responsible for the increased risk of collision among beginning drivers. For this reason, graduated licensing programs should apply to all new beginning drivers, not just young drivers. These programs provide beginning drivers with practical experience, while minimizing their exposure to risk. There should be three stages to the program. To enter stage one, applicants should be required to pass a traffic rules and road signs test, and establish that they are fit to drive. Drivers should remain at stage one for 12 months. They should be supervised at all times by a fully-licensed driver who is at least 21, has two years of experience with full driving privileges, and occupies the front passenger seat. The adult supervisor assists in the learning process by monitoring and correcting the drivers actions. Research indicates that young drivers are at increased risk when they carry passengers. The presence of passengers is a source of distraction and a potential source of peer pressure to engage in risky behaviour. A recent American study conducted by the Ford Motor Company found that the relative risk of crash for 16-year-old drivers increases with the number of passengers. For example, a 16-year-old carrying one passenger is 39% more likely to be killed in a crash than a 16-year-old driving alone. This increases to 86% for two passengers, and 182% for three or more passengers. In addition, several studies indicate that teenage passengers are at greater risk when riding with a young driver. Thus, the research supports limiting the number of passengers in the car with a beginning driver. However, the situations examined in the preceding research typically occurred outside the scope of a comprehensive graduated licensing program and without a mature supervising driver. Presumably, the risks associated with passengers are much lower for a beginner who is driving with his or her parents and younger siblings, than for an unsupervised beginner with a carload of teenagers. Thus, if jurisdictions are successful in enacting comprehensive graduated licensing programs with strict supervision of beginning drivers, the passenger restriction will be less critical. Nevertheless, MADD Canada supports a condition that sets the maximum number of passengers lower than the number of seatbelts in the vehicle. This would provide beginning drivers with an atmosphere conducive to concentrating on the task of driving. Drivers holding a stage-one licence should be prohibited from late-night driving. This helps ensure that beginning drivers do not have to cope with the added visibility and judgment problems posed by night driving, or with the presence of dangerous or impaired drivers, who tend to be more prevalent at night. Since it appears that a majority of nighttime crashes among young drivers occur before midnight, earlier curfews (such as 10 p.m.) are preferable to later ones (midnight). Stage-one drivers should also be prohibited from driving on high-speed, multi-lane highways, if such roads exist in the province. These roads typically carry heavy traffic, including large trucks moving at high speeds, which can be overwhelming for a new driver. Moreover, crashes on high-speed roads are generally more serious than those occurring on lower-speed roads. Although MADD Canada recognizes the potential benefits of driver education programs, we do not include mandatory driver education as a cornerstone of graduated licensing. Research has failed to establish that the current driver education programs have a long-term positive impact on the likelihood of crashes. Furthermore, we would advocate against shortening the minimum periods of the graduated licensing program for drivers who participate in such programs. While the goals of driver education are admirable, the programs do not provide an adequate substitute for extensive, supervised, on-the-road driving experience. Instead, MADD Canada recommends that provinces enact a required minimum number of total and nighttime supervised driving hours, which could be recorded in a driving log and verified by a parent or qualified driving instructor. This requirement would encourage more practice driving under safe conditions, and would help to prevent situations where a beginning driver simply holds the stage-one licence for a year without gaining the benefits of supervised driving. One of the key components of the graduated licensing program is the requirement of abstaining from alcohol. All stage-one and two drivers should be required to have a BAC of 0.00% when driving. Beginning drivers are already disadvantaged because of their in-experience, and should not have their judgment further impaired by alcohol. This limitation should apply to all stage-one and two drivers, regardless of age. While older beginning drivers may be more mature and experienced with alcohol than young beginning drivers, they still lack driving experience, and this is reflected in their crash rates. For example, 30-year-old beginners have a 41% higher collision rate than 30-year-olds with five years driving experience. Similarly, 20-year-old beginners have a 28% higher collision rate than 20-year-olds with five years driving experience. Zero and low BAC restrictions have been shown to have positive results. A study of the American states that introduced zero or low BAC requirements for young drivers between 1983 and 1992 found a 16% decrease in the proportion of single-vehicle nighttime fatal crashes among affected drivers, while the proportion in the control states increased by 1%. The authors estimated that, if the remaining 21 states introduced a zero or low BAC limit for drivers under 21, at least an additional 375 fatal single-vehicle nighttime crashes could be prevented each year among 15 to 20-year-old drivers. The largest traffic safety improvements occurred in states that lowered the BAC limit to 0.00%. For example, Maine had a 0.02% BAC restriction on drivers under 21 from 1983 to 1995. When a 0.00% restriction was introduced in 1995, there was an additional 36% reduction in the number of nighttime single-vehicle injury crashes among youth under 21. This is likely because a 0.00% limit prohibits drinking and driving altogether, whereas a 0.02% limit suggests to the driver that some drinking before driving is acceptable. Studies from Ontario confirm the traffic safety and other benefits of a 0.00% BAC restriction for beginning drivers. The supervising driver should also be subject to a 0.00% BAC restriction. Alcohol consumption would impair the supervising drivers ability to monitor the beginning driver and respond quickly to any urgent situation. Moreover, permitting the supervising driver to have consumed alcohol sets a poor example for the beginning driver, and increases the likelihood that beginning drivers will be used as convenient designated drivers for their older impaired friends. To progress to stage two, a driver should be required to submit the required signed logs and successfully complete a stage-one road test. The requirements for passing this road test should reflect the fact that successful candidates will be permitted to drive unsupervised, except on high-speed roads and late at night. In these two circumstances, they would have to be accompanied by a supervising driver with a 0.00% BAC. Drivers would be required to remain at stage two for a minimum of 12 months, giving them at least 24 months of some form of supervised driving in the graduated licensing program. These drivers should be required to complete a minimum number of hours of supervised driving on high-speed roads, which should be recorded in a driving log and verified by the beginners supervisor(s). Finally, the 0.00% BAC restriction would remain in effect for the beginning driver during stage two. In order to obtain a stage-three licence, an applicant would be required to successfully complete an exit road test, which would include driving on a high-speed road. The successful applicant would then be granted full driving privileges. Stage three would last for two years. During this period, drivers should be subject to demerit point thresholds that are 50% lower than those for experienced drivers. This would assist the Registrar of Motor Vehicles in identifying potentially at-risk new drivers and in taking remedial action before they have a serious crash. In addition, the threat of suspension generally has a considerable deterrent impact and discourages newly-licensed drivers from taking risks. While suspensions can, at times, be counter-productive because they reduce the drivers opportunities to gain practical experience, it should be remembered that stage-three drivers will already have a minimum of two years driving experience. Thus, a short-term suspension is justifiable for beginning drivers who develop poor driving records at this early stage. (d) A 0.00% BAC Limit for all Drivers Under 21 INNOVATION Given the preceding discussion on the effectiveness of zero BAC limits for beginning drivers, it is unnecessary to repeat that material he re. Rather, this subsection addresses the reasons for extending that measure to all drivers under the age of 21. While our proposal addresses 16 to 20-year-olds, the Canadian research does not report aggregate data on this exact age bracket. Consequently, some of the following data will be over or under-inclusive of the target age group. Despite the progress that has been made in recent years, crashes remain the leading cause of death and injury among 15 to 19-year-olds. For example, over one-third of all deaths and almost one-quarter of all serious injuries among this group result from traffic crashes. As the following graphs illustrate, young people have far higher per capita rates of crash death and injury than older Canadian drivers.  EMBED MSGraph.Chart.8 \s  Source: D. R. Mayhew & H.M. Simpson, Youth and Road Crashes: Reducing the Risks from Inexperience, Immaturity and Alcohol (Ottawa: Traffic Injury Research Foundation, 1999) at 6.  EMBED MSGraph.Chart.8 \s  Source: D. R. Mayhew & H.M. Simpson, Youth and Road Crashes: Reducing the Risks from Inexperience, Immaturity and Alcohol (Ottawa: Traffic Injury Research Foundation, 1999) at 7. Unfortunately, the per capita rate of crash deaths among 15 to 19-year-olds has been hovering around 20 per 100,000 since 1992, and the relative proportion of youth to total crash deaths has not changed significantly since 1990. Moreover, a large percentage of these deaths and injuries are alcohol-related. For example, 40% of all teenage drivers killed and 23% of those seriously injured in 1996-97 had been drinking. Within this group, 50% of the fatally injured and 28.8% of the seriously injured 19-year-old drivers had been drinking. Although no separate figures are provided for 20-year-olds, almost 55% of the fatally injured 20 to 25-year-old drivers had been drinking. These high rates of alcohol-related traffic fatalities among youth are not surprising, given their patterns of alcohol consumption. In a survey conducted in 1996-97, over 70% of 15 to 19-year-old Canadians reported consuming alcohol. A more recent survey found that, among Canadian drinkers of this age, almost two-thirds of males and almost half of females reported binge drinking (5 or more drinks in a sitting), with a significant minority doing so on a routine basis. Again, there are no separate data for 20-year-olds. However, the rate of alcohol consumption and the percentage of occasional and frequent binge drinking rise sharply in the 20 to 24-year-old age group. While graduated licensing programs have significantly reduced alcohol-related crashes among young beginning drivers, the reach of these programs is limited. Part of the problem is that the BAC restrictions are lifted upon completion of the graduated licensing program, which usually occurs around the age of 18. This corresponds to the legal drinking age in most provinces and territories, and to a period in which alcohol consumption and binge drinking increase. Moreover, this is precisely the age at which teenage drivers are most vulnerable to alcohol-related crash deaths and injuries. A 1999 Canadian study reported that 18 and 19-year-old drivers account for almost 74% of such teenage deaths and 71% of such teenage injuries. As explained earlier, 20-year-old drivers are likely to be at least as vulnerable to alcohol-related crash deaths and injuries. It is dangerous to expose 18 to 20-year-olds to their first experiences of unsupervised driving at the same time as their first legal use of alcohol. Consequently, the 0.00% BAC requirement should apply beyond the graduated licensing program until a driver is 21. Young beginning drivers usually lack experience with both driving and alcohol. Moreover, young people tend to be risk-takers and are less cautious than their older counterparts. Thus, as indicated, even sober young drivers are at greater relative risk of crash than older, more experienced drivers. Moreover, as illustrated in the following chart, which is based on American data, young drivers who drink are at a far greater relative risk of death than older drivers with comparable BACs. The Relative Risk of a Fatal Single-Vehicle Crash for Males at Various BACs .020%-.049%.050%-.079%.080%-.099%.100%-.149%Age 16-204.6417.3251.87240.89Age 21-342.756.5313.4336.89Age 35+2.575.7911.3829.30 Source: P.L. Zador, S.A. Krawchuk & R.B. Voas, Alcohol-Related Relative Risk of Driver Fatalities and Driver Involvement in Fatal Crashes in Relation to Driver Age and Gender: An Update Using 1996 Data (2000) 61 J. Stud. Alcohol 387 at 392. Given that the alcohol-related crash rates decrease at about the age of 21, it is justifiable to extend the 0.00% BAC restriction until a driver reaches this age. As indicated in the Graduated Licensing section, 0.00% BAC restrictions have been shown to reduce impaired driving deaths among youth. Studies also indicate that such BAC restrictions are more effective if they are extended until the driver reaches age 21. For example, Oregon experienced a 40% reduction in single-vehicle nighttime crashes among affected drivers when it extended its 0.00% BAC restriction from drivers under 18 to drivers under 21 in 1991. The 0.00% BAC proposal is particularly time-sensitive, because Canadas youth population is increasing. By 2011, it is estimated that the number of young licensed drivers will have increased by 10.3% over 1999 levels. Therefore, unless significant progress is made, the number of Canadian youth who are killed or injured in alcohol-related crashes will rise sharply in the next decade. (e) Enforcement of the Graduated Licensing Program and 0.00% BAC Limits In order to effectively enforce the graduated licensing program and the 0.00% BAC restriction on young drivers, provincial legislation must specifically authorize the police to demand that beginning drivers and their supervisors identify themselves and present their drivers licences. Moreover, the police must be authorized to demand breath samples on an ASD from any stage-one or two driver, any supervising driver, and any driver under the age of 21. The police would have great difficulty detecting those in breach of the 0.00% BAC limit through standard investigating techniques, because many of these individuals would exhibit no or few visible signs of alcohol consumption or impairment. Consequently, the ability to demand breath samples from young drivers and their supervisors should not be dependent on a suspicion that they have consumed alcohol. Although this may conflict with the general principle that searches should not be conducted in the absence of reasonable grounds, this provision should withstand Charter scrutiny because it would be used for the limited purpose of enforcing the 0.00% BAC provision, and would not be admissible in evidence at a criminal trial. If the supervising driver breaches the 0.00% BAC restriction, his or her licence should be suspended for 24 hours, and the beginning driver should not be allowed to continue driving unless someone else is able to take over as the qualified supervisor. Stage-one and two drivers and drivers under 21 who violate the 0.00% BAC restriction should have their licences immediately suspended. They should also be required to successfully complete a remedial alcohol education program at their own expense before being allowed to apply for reinstatement. If reinstated, stage-one and two drivers should face licensing sanctions or restrictions, such as extended periods of supervised driving. Drivers who are involved in at-fault crashes, commit serious provincial traffic offences, or breach conditions of the graduated licensing program should also be subject to extended periods of supervision and, depending on the drivers record, possible licence suspensions. While licence suspensions may appear to be a harsh penalty for those who violate the 0.00% BAC restriction, research indicates that they may be the most meaningful punishment for young people. For example, at least 36 American states have Use and Lose laws, whereby youth who drink, possess, or attempt to purchase alcohol while underage have their drivers licences suspended or, if not yet licensed, have their ability to apply for a licence delayed. The theory behind such laws is that young people greatly value the ability to drive, and will be deterred by a law that threatens to remove that ability. A recent study of the Use and Lose law in Pennsylvania found that youth who received licence sanctions as a result of alcohol-related charges had lower rates of subsequent crashes and traffic violations than youth who received no licence sanction. Thus, graduated licensing programs that include suspensions for violations of the 0.00% BAC limit or other serious breaches should have a considerable deterrent impact, thereby enhancing traffic safety. SECTION II: POLICE ENFORCEMENT POWERS An Introduction to the Federal Enforcement Powers There is a complex interplay between the provincial enforcement powers and the Criminal Code breath and blood-testing provisions. Consequently, this section begins with a brief summary of the relevant Criminal Code sections and then documents the major enforcement difficulties the police face under the current laws. This background information is essential in understanding the proposed provincial enforcement powers that are set out in the remainder of the section. It should be emphasized that the recommended provincial powers are intended to operate in addition, and not as an alternative, to the existing Criminal Code provisions. (i) Preliminary Screening and Evidentiary Breath Tests The Criminal Code authorizes the police to demand breath samples from drivers in two situations. First, the police can require a breath sample for analysis on an ASD from any driver who they reasonably suspect has alcohol in his or her body. ASDs are small, hand-held, breath-testing machines that are typically carried in police patrol cars, and are generally used to administer roadside tests. The manner of driving, the odour of alcohol on the drivers breath, difficulties answering the officers questions, clumsiness in handing over requested documents, and the drivers admission that he or she was just at a bar could all create a reasonable suspicion that the driver had consumed alcohol. The police need not believe that the driver is drunk, impaired or committing an offence. Second, the police can demand breath samples for analysis on an approved instrument from any driver who they have reasonable grounds to believe has committed the offence of impaired driving or driving with a BAC above 0.08% within the previous three hours. These grounds for demanding breath samples are far more limited than those for ASD testing. Approved instruments are larger, more sophisticated machines than ASDs, and are typically kept at police stations or in specially-equipped vans. Tests taken on approved instruments are commonly referred to as evidentiary breath tests. A driver who fails or refuses, without a reasonable excuse, to provide breath samples for an ASD or evidentiary breath test is guilty of a federal criminal offence that carries the same penalties as impaired driving and driving with a BAC above 0.08%. The results of an ASD test are not admissible in criminal proceedings as evidence of a drivers BAC. Nevertheless, the results often provide the police with the legal grounds for demanding a test on an approved instrument. As indicated, an officer needs only a reasonable suspicion that a driver has alcohol in his or her body to demand an ASD test. Since ASDs are typically set to register a fail at a BAC of 0.10%, a drivers failure on the ASD provides the police with the required reasonable grounds to demand an evidentiary breath test. Thus, ASDs can be used to test large numbers of suspected drinking drivers at roadside, only some of whom will be required to take a test on an approved instrument. The results of the latter test are admissible in criminal proceedings and are deemed, in the absence of evidence to the contrary, to be proof of the drivers BAC at the time of the offence. (ii) Evidentiary Blood Tests The Criminal Code authorizes the police to demand blood samples from a driver if they have reasonable grounds to demand evidentiary breath samples and to believe that the driver is incapable of providing breath samples or it is impracticable to obtain them because of the drivers physical condition. A driver who fails or refuses to provide a blood sample, without a reasonable excuse, is guilty of a federal criminal offence that is subject to the same penalties as impaired driving and driving with a BAC above 0.08%. As outlined below, these provisions often prove to be unworkable. Some courts have indicated that the police should base their beliefs about the drivers inability to provide breath samples or the impracticability of obtaining them on medical opinion. However, medical staff has no obligation to provide such information and may face disciplinary sanctions for breaching patient confidentiality if they do so. In addition, police authority to demand blood samples and the evidentiary weight of the test results are dependent on rigid time constraints. Particularly in serious crashes, these time limits may expire before the police can secure the scene, oversee the victims transportation to hospital, get to the hospital, locate the suspect, make the blood demand, give the suspect a reasonable opportunity to contact and consult with a lawyer, and find a doctor who is willing to take or arrange for the taking of the sample. Moreover, the courts interpretation of the impracticability requirement has posed major problems in attempting to obtain blood samples. The fact that the suspect is in hospital is not viewed as necessarily making breath testing impracticable, even though this is a daunting task. An approved instrument and qualified technician must be brought to the hospital, an appropriate room for testing has to be found and made available, and the instrument must be set up, calibrated and tested. Many of these cases arise at times of peak demand on hospital and police resources, when facilitating evidentiary breath testing is not likely to be a priority for hospital staff. Further, other impaired driving suspects who are brought to the police station may escape criminal liability if an approved instrument and technician are at the hospital. Similarly, the case law has established that police concern over the running of the time limit does not, by itself, render the taking of breath samples impracticable. However, it is unclear what combination of additional factors is sufficient to make it so. Moreover, judges have interpreted the requirement that the impracticability stem from the suspects condition with varying degrees of strictness. Unfortunately, these conflicts in the lower court decisions have not yet been resolved by the appellate courts. The Criminal Code also permits the police to apply for special judicial warrants, which authorize the taking of blood samples from impaired driving suspects who are incapable of responding to a demand for a sample. However, the circumstances in which these warrants are available are very narrowly defined. First, the police must have reasonable grounds to believe that the suspect committed an impaired driving offence within the preceding four hours, and that the suspect was involved in a crash causing death or bodily injury. Second, the police must have reasonable grounds to believe that a medical practitioner is of the opinion that, by reason of either alcohol consumption or the crash, the suspect is unable to consent to providing a blood sample. Third, the police must have reasonable grounds to believe that a medical practitioner is of the opinion that the taking of the sample would not endanger the suspects health or life. As indicated, doctors are under no obligation to provide the police with any patient information. Nor are doctors required to participate in the taking of the sample, even when the police have obtained a special warrant. Perhaps the greatest obstacle for the police is establishing that they had reasonable grounds to believe that the suspect had committed an impaired driving offence. Since these suspects are typically unconscious, the police will have no ASD results and it will be difficult to detect any signs of intoxication. Furthermore, the signs that are apparent can often be attributed to a head injury, shock or some other injury or pre-existing illness. Thus, even if it is obvious that the driver has been drinking and the police reasonably suspect that he or she is impaired, they may still fail to meet the required grounds for obtaining a valid special warrant. Contrary to Parliaments stated goals in enacting these blood demand and warrant provisions in 1985, impaired drivers who need or demand to be taken to hospital often escape criminal liability. Since most of these cases involve fatal or personal injury crashes, the current law ironically protects drivers who have committed the most serious impaired driving offences. Evidence that the Current Federal Law is Ineffective Millions of Canadians continue to drink and drive. In a 2002 survey, 16.1% of licensed drivers said that they had driven within two hours of drinking in the past 30 days. In the authors words, when applied to the entire population of licensed drivers, it indicates that an estimated 3.4 million Canadians admit to driving after drinking. Moreover, an estimated 1.7 million drove in the past year when they thought they were over the legal limit, and 63% of these drivers admitted to doing so more than once. Thus, there are many thousands of impaired drivers on Canadian roads each night, and very few ever come to police attention. It has been estimated that only 1 in every 445 impaired driving trips in Canada results in a criminal charge, and some American studies have indicated that the figure may be as low as 1 in 2,000. Even if the police conclude that a driver is legally impaired, criminal charges may not be laid. In a recent national survey, 42% of Canadian police officers admitted that they sometimes or frequently release impaired driving suspects with a short-term provincial suspension rather than proceed with criminal charges. One-third of the officers indicated that they sometimes or frequently release the driver without any sanction, and merely arrange for safe transportation home. Among the reasons cited for not laying criminal charges was that the process was too time-consuming. Moreover, three-quarters of the officers specifically stated that they were discouraged because impaired drivers sometimes or frequently escape conviction on legal technicalities. In addition, a survey of British Columbia police found that nearly half of the officers refuse to lay Criminal Code impaired driving charges, even if they believe that the driver is impaired. The prosecution of impaired driving cases has become equally challenging. While holding police and prosecutors to exacting standards of proof, the courts have interpreted the law to unduly narrow the offences, and have recognized defences that appear to lack credibility. Figures from the Canadian Centre for Justice Statistics indicate that, depending on the offence, only 23% to 61% of impaired driving charges between 1994 and 1998 resulted in a guilty disposition in the provincial courts. To make matters worse, those charged with the most serious offences were the least likely to be convicted. While the conviction rate for driving with a BAC above 0.08% was 61%, the rate dropped to only 33% for impaired driving causing bodily harm and 23% for impaired driving causing death. The federal government introduced amendments to the Criminal Code impaired driving provisions in 1999 and 2000, but these focused almost exclusively on increasing the penalties. The government ignored proposals to enhance police enforcement powers, redefine some of the offences and more carefully frame some of the defences. In short, Parliament failed to consider the most pressing issues, namely, the diminishing deterrent impact of the federal law and the obstacles to the efficient apprehension, processing and prosecution of impaired drivers. The federal governments continued inaction has increased the need for comprehensive provincial legislation. Authority to Stop Vehicles In R. v. Dedman, the Supreme Court of Canada held that the police have a common law power to stop motor vehicles to conduct spot checks or sobriety checkpoint programs. The Court held that this power is necessary if the police are to carry out their duty to prevent impaired driving. However, the decision was limited to organized spot checks, and the Court did not comment on whether there is a broader common law power to stop any vehicle at random. Thus, provincial legislation is still required to give the police authority to stop any vehicle at random during routine patrol activities. Failing to stop when directed to do so and wilfully attempting to evade police pursuit should be provincial offences. Given the risks involved, the conduct should also result in a mandatory administrative licence revocation and a lengthy driving disqualification. Although it is an offence in most provinces to fail to stop when lawfully directed to do so by an officer, some jurisdictions do not authorize police to stop vehicles at random. Rather, the police can only demand that a driver stop if they have reasonable and probable grounds to believe that the vehicle is being driven in contravention of the highway traffic legislation. Having clear legislative authority to stop vehicles at random would be beneficial in several regards. First, it would assist the police in detecting impaired drivers during routine patrol, where no erratic driving or offence is witnessed. By stopping the vehicle and speaking to the driver, the police will be able to observe him or her for signs of drinking and determine if there are grounds to demand an ASD test. While organized checkpoint programs provide a strong deterrent effect, the majority of impaired driving arrests in Canada are apparently made during routine patrol. For example, an 18-month study of all impaired driving charges in the Sudbury area found that 5% arose from organized checkpoints, 32% arose from collisions and 62% arose from routine patrol activities. Second, the ability to stop vehicles at random greatly increases the likelihood of apprehending unlicensed, disqualified, suspended, prohibited, and uninsured drivers. Similarly, it assists the police in detecting drivers who may be unfit to drive due to fatigue, illness, drug use, or some other factor. Thus, it gives officers an opportunity to take pro-active steps to remove these potentially at-risk drivers from the road before they are involved in a crash. Third, if the police power to stop vehicles is limited to stated grounds, it provides defence counsel with a basis for challenging the officers conduct. If the challenge is successful, it will likely result in the dismissal of any provincial traffic or Criminal Code charge that arose from that stop. Fourth, clear legislative authority to stop vehicles at random is required if such a stop is challenged as an arbitrary detention under section 9 of the Charter. In order for the Crown to justify the random stop under section 1, it must be prescribed by law. That is, it must be expressly provided for by statute or regulation, or result by necessary implication from the terms of the statute or regulation or from its operating requirements or from the application of a common law rule. Clear statutory authority would assist the Crown in meeting this burden. Authority to Demand Documentation As a corollary to the power to stop vehicles at random, police should also have statutory authority to demand that drivers identify themselves and present their licences, insurance and ownership documents. Drivers should have a legal obligation to comply with an officers request, and failure to do so should be a serious provincial offence. In addition, police should be authorized to arrest, without a warrant, any driver who refuses to identify him or herself, or is believed, on reasonable grounds, to have provided false information. The deterrent impact of the impaired driving legislation is significantly undermined if suspended and prohibited drivers can continue to drive with little risk of being caught. A recent study in the Moncton area found that the number of suspended drivers stopped at four checkpoints was 57% of the expected total, if all suspended drivers had continued to drive as much as they did prior to the suspension. In other words, the suspensions only reduced the incidence of driving among suspended drivers by 43%. The suspended drivers on the roads, particularly those suspended for an impaired driving offence, was highest after midnight. This roadside data suggests that most suspended drivers continue to drive, and that many do so on a routine basis. American studies estimated that 65% to 75% of suspended drivers continue to drive. A 1997 study, using data from the Fatal Accident Reporting System (FARS) for the period 1987-1992, estimated the number of suspended/revoked and unlicensed drivers on California roads. The authors indicated that 8.8% of the drivers, during the times examined, had a suspended or revoked licence, and that a further 3.3% had no record of a licence. Thus, roughly 12% of California drivers were driving illegally. Moreover, relative to the amount of driving they did, suspended and revoked drivers were 3.7 times more likely to be involved in a fatal crash than licensed drivers. Unlicensed drivers were over-represented in fatal crashes by a factor of 4.9 to 1. Finally, a 2003 study found that 20% of all fatal crashes in the United States between 1993 and 1999 involved at least one driver who was suspended or did not otherwise have a valid licence. Most suspended and unlicensed drivers are never apprehended or charged. The only effective way to apprehend prohibited, suspended, unlicensed, or uninsured drivers is to stop vehicles and demand documentation, and the police should be given clear statutory authority to do so. Authority to Establish Systematic Sobriety Checkpoints Provincial legislation should authorize the police to establish systematic sobriety checkpoints. Such checkpoints are already used in many Canadian jurisdictions, and generally operate in the following manner. All drivers approaching the checkpoint are stopped and engaged in a brief conversation with the police, who may ask to see documentation. During the conversation, the police attempt to determine whether the driver has been drinking by asking the driver directly, looking for visible signs of impairment and attempting to detect the odour of alcohol. As described, police can demand a breath sample on an ASD if they reasonably suspect that the driver has consumed any alcohol. Systematic sobriety checkpoints have two major objectives. First, they increase the deterrent impact of the federal and provincial impaired driving laws through publicity and by increasing the perceived risk of apprehension. Second, the checkpoints greatly increase the detection and apprehension rates of impaired, unfit, suspended, uninsured, and unlicensed drivers. By allowing the police to assess the drivers condition and inspect his or her documents, systematic sobriety checkpoints can greatly enhance the enforcement of federal and provincial impaired driving laws and countermeasure programs. Studies show that well-publicized sobriety checkpoints have a significant general deterrent effect. A recent literature review reported that selective breath-testing (SBT) at sobriety checkpoints resulted in a median decrease of 20% in fatal and personal injury crashes, and 24% in property damage crashes. The review also concluded that sobriety checkpoints can maintain their effectiveness over time. The National Highway Traffic Safety Administration (NHTSA) funded a demonstration project of sobriety checkpoints in Tennessee. The program, known as Checkpoint Tennessee, lasted for one year and consisted of at least 576 individual checkpoints in different regions of the state. In addition, at five times during the year, weekend blitzes were held in each of the states 95 counties. The program was accompanied by a widespread public information campaign, including 720 minutes of televised public service announcements, as well as print, radio and billboard advertising. Police had brief conversations with drivers passing through the checkpoints, but only demanded breath tests when they had the requisite probable grounds. The researchers found that the checkpoint program reduced fatal impaired driving crashes (drivers BAC 0.10% or above) by 20.4%, or approximately nine fatal crashes per month. They concluded that it is incumbent on policy makers and administrators to find ways to implement similar programs in their states. A study of a five-month intensive checkpoint campaign in British Columbia found a 45% decrease in the number of drivers who had been drinking. There was also a 65% reduction in the proportion of drivers with a BAC above 0.08%. The effectiveness of the campaign may have resulted from its high visibility nine of ten drivers surveyed at a checkpoint reported that they were aware of the campaign, and 62% stated that they had already been through a checkpoint during the five-month campaign. Conversely, if a checkpoint program is not well-publicized, its deterrent impact may quickly fade. This is likely because the actual risk of apprehension remains too low to be an effective deterrent. MADD Canada would ultimately like to see sobriety checkpoint programs operate in conjunction with RBT, as in Australia. However, we recognize that checkpoints can, on their own, increase the perceived probability of arrest, thereby increasing general deterrence. This deterrent effect will only occur if the checkpoints are frequent, highly visible and well-publicized. Thus, provinces must not only introduce explicit authorizing legislation, but also invest resources in publicizing and operating the checkpoints. Studies indicate that this investment will be cost effective. For example, an American cost-benefit study concluded that every dollar spent on sobriety checkpoints, including enforcement, travel delay and criminal justice costs, will save the community six dollars in medical care, public emergency services, property damage, future earnings, and insurance. Thus, in addition to the reductions in deaths and injuries, there is an economic justification for introducing systematic sobriety checkpoints. Authority to Demand an Approved Screening Device (ASD) Test from Drivers at Sobriety Checkpoints INNOVATION To maximize the effectiveness and general deterrence of sobriety checkpoints, police should have express statutory authority to demand a breath sample on an ASD from any driver stopped at such a checkpoint. Accordingly, the breath-screening power would operate similarly to stationary RBT, as currently practised in Australia, New Zealand and some European jurisdictions, which has been shown to have very positive effects on traffic safety. A rigorous time-series analysis of RBT in four Australian states found that it had significant initial and ongoing effects on serious, fatal and single-vehicle nighttime crashes. For example, in Queensland, RBT was estimated to have reduced fatal crashes by 35% between 1988 and 1992, preventing an estimated 789 fatal crashes during that period. Another study, which evaluated RBT in New South Wales, found that the number of fatally-injured drivers with illegal BACs (over 0.05%) dropped by 36% in the four years after RBT was introduced. That study also reported substantial declines in the number of individuals who reported driving when they believed themselves to be at an unsafe BAC. Finally, RBT in New South Wales significantly increased the perceived probability of arrest, with 90% of drivers surveyed believing they might be caught in a breath-testing checkpoint. RBT is effective as a general deterrent because it significantly increases the perceived certainty of punishment. However, to maintain this perception, researchers suggest that a very high testing rate (i.e., each licensed driver is tested at least once a year) and substantial publicity are required. For example, a study of two Swedish counties with different rates of RBT found that an individuals personal encounters with RBT were the most effective way to increase the perceived probability of arrest. Currently, it is estimated that only one-third of Canadian drivers encounter a sobriety checkpoint annually, and these do not even entail RBT. Thus, for breath screening to be effective in Canada, provincial governments will have to publicize the program, increase the number of officers in impaired driving enforcement, and probably purchase more ASDs. Much of the hesitation to introduce RBT in North America stems from concerns about civil liberties. In Canada, RBT will invariably be challenged as an unreasonable search and seizure under section 8 of the Charter, because drivers would be required to provide a breath sample without any suspicion that they were impaired. However, MADD Canada believes that properly designed RBT programs should withstand Charter scrutiny. While we will address this issue at greater length in a separate article, some of the major lines of argument are noted below. First, as indicated, consenting to an ASD test when stopped at a sobriety checkpoint should be made an express condition for the issuance of every drivers licence or licence renewal. Thus, drivers would have advance notice of the requirement to provide a breath sample similar to those involved in other licensed and regulated activities who agree in advance to submit to spot audits and inspections. Second, if all drivers passing through checkpoints are asked to provide a sample, the police conduct is less likely to be challenged on the basis of arbitrariness or the improper exercise of their discretion. Moreover, the breath screening power proposed here would be limited to tests on ASDs. Drivers would not be required to provide evidentiary breath samples unless the relevant Criminal Code requirements were satisfied. They would also be entitled to contact counsel before providing an evidentiary breath sample. Finally, similar to the Administrative Licence Suspension (ALS) legislation that exists in many provinces, drivers who face revocation for having a prohibited BAC or failing to provide a breath sample at a sobriety checkpoint should have the right to apply for a review by the Registrar. These safeguards, taken together, should establish that the breath screening power is a justifiable limit on the right to be free from unreasonable search and seizure. Authority to Use Passive Alcohol Sensors In order to demand that a driver submit to a breath test on an ASD, the police must reasonably suspect that the driver has alcohol in his or her body. Currently, the police make this determination by observing the manner of driving, questioning the driver, trying to detect the odour of alcohol on the drivers breath, and relying on their other unaided senses. Two problems have become evident with ASD testing. First, some Canadian judges have applied a very rigorous and, in our view, inappropriate standard for breath and blood-testing, and too readily reject police testimony as to their observations. A recent survey of Canadian police found that one-third believed that judges gave greater credibility to the expert witnesses for the defence than to their evidence. In addition, there is confusion in the law as to whether the smell of alcohol on the drivers breath is alone sufficient to constitute a reasonable suspicion that the driver has alcohol in his or her body. Typically, these problems result in the exclusion of critical evidence, including any evidentiary breath or blood tests. This, in turn, results in the charges against the accused being dropped. The second major concern with the current approach is that it fails to detect significant numbers of drinking drivers. American studies suggest that the police miss about 50% of the drivers with BACs of 0.10% or higher, and about 75% of drivers with BACs between 0.05% and 0.099%, that they stop and question at sobriety checkpoints. For example, one study reported that over 60% of drivers with BACs of 0.08% or higher were passed through checkpoints, as were almost 90% of drivers with BACs of between 0.05% and 0.079%. This low detection rate is not surprising, given the need to make checkpoints fast and convenient for the large majority of sober drivers. However, it indicates a need to develop more effective enforcement techniques, while maintaining the convenience of checkpoint programs. Currently, police rely heavily on behavioural and other observations in detecting suspected impaired drivers, such as the manner of driving, the odour of alcohol on a drivers breath, a flushed face, a lack of coordination, bloodshot eyes, slurred or indistinct speech, and inappropriate responses to questions. These signs may be difficult to detect in the brief time that motorists are currently stopped at checkpoints. Alcoholic beverages vary in type and intensity of aroma, and police differ in their abilities to detect alcohol. Moreover, experienced drinkers may be able to conceal the signs of intoxication or avoid raising police suspicions, and those who do not fit the impaired driver stereotype are less likely to be identified as being impaired. The deterrent effect of checkpoint programs is severely compromised if drivers believe that there is a low risk of detection and apprehension. In response to these problems, a majority of American states have introduced the use of passive alcohol sensors and related devices. Similarly, MADD Canada recommends that each province enact legislation giving police explicit authority to use passive alcohol sensors. Of course, if provinces introduce RBT at systematic sobriety checkpoints, the need for passive alcohol sensors will be greatly reduced. There are various types of passive alcohol sensors. In some American jurisdictions, the devices are built into police flashlights or clipboards. In Canada, there are several light-weight baton or wand models on the market. All of the devices work in essentially the same way. The device is held close to the drivers mouth and a fan draws in a sample of the ambient air. The sample is then analysed. The models can be designed to provide a range of readings, such as pass (BAC below 0.05%), warn (BAC 0.05% to 0.099%), or fail (BAC of 0.10% or higher). Alternatively, the devices can be designed to provide simply a positive (BAC of 0.02% or higher) or negative reading (BAC of 0.00% to 0.019%). The Canadian baton model provides a simple positive or negative result. In the context of current Canadian law, a positive reading on the latter type of sensor can quickly and easily provide an officer with a reasonable suspicion that the driver has alcohol in his or her body. In turn, this finding authorizes the officer to demand a breath sample for analysis on an ASD under section 254(2) of the Criminal Code. The results of a test on a passive alcohol sensor could not be used to establish a drivers BAC, and would not be admissible in evidence for this purpose. Rather, the device would be used as a preliminary screening tool, providing the officer with a better nose to determine if a driver has been drinking. Both laboratory and field tests have shown that passive alcohol sensors can reliably and accurately detect the presence of alcohol. For example, an early study of two passive alcohol sensors conducted by NHTSA in 1988 found that the sensors were able to detect BACs above 0.05% between 77% and 100% of the time. A more recent study found that the sensors were particularly reliable in detecting moderate alcohol levels, accurately identifying 93% of drivers with BACs above 0.02%. This means that passive alcohol sensors are able to detect drivers who show few or no visible signs of impairment. Thus, they are particularly helpful in preliminary screening of drivers for the presence of alcohol, and in enforcing zero and low BAC laws among new and other special categories of drivers. We are primarily interested in passive alcohol sensors as an initial screen for the mere presence of alcohol, a task at which they are very accurate. However, they are also widely used to identify drivers with higher BACs. An early study found that, compared to police using conventional methods, officers using passive alcohol sensors identified 80% more drivers with BACs above 0.05%, and detained 50% fewer drivers with BACs below this threshold. Thus, passive alcohol sensors not only improved detection rates, but also decreased false positives, thereby reducing the delay and inconvenience experienced by lawful drivers at checkpoints. Moreover, studies have shown that tests on passive alcohol sensors take as little as ten seconds to administer. A 1998 study concluded that officers using the PAS III sensor could identify 75% of drivers with BACs at or above 0.10% and 70% with BACs at or above 0.08%, which was a vast improvement over the 40-50% detection rates of police at checkpoints without sensors. Theoretically, there is nothing prohibiting the use of passive alcohol sensors in the provinces and territories. However, clear legislative authority could assist enforcement in several ways. First, the legislation would serve a public education function, alerting drivers to the use of the devices and the heightened risk of apprehension. If the public were aware that such accurate devices were being used at checkpoints, it would likely increase the laws deterrent impact. Second, legislation would remove all doubts about police authority to use passive alcohol sensors. Third, and perhaps most importantly, statutory authority would address potential Charter challenges to the use of the devices. The use of passive alcohol sensors would most likely be challenged as constituting an unreasonable search and seizure, contrary to section 8 of the Charter. The Supreme Court of Canada has held that, for a search to be reasonable under section 8, it must at a minimum be authorized by law. In the absence of authorizing legislation, police use of passive alcohol sensors would be held to violate section 8 at this threshold stage. Even if authorizing legislation were enacted, the courts might still conclude that the use of sensors infringes section 8. However, an infringement of a Charter right may be permitted under section 1 of the Charter, if the infringement is a reasonable limit prescribed by law that can be demonstrably justified in a free and democratic society. Legislation authorizing the use of passive alcohol sensors is essential in meeting the first requirement of section 1, namely, that the infringement is prescribed by law. The Supreme Court has held that an infringement is prescribed by law if it is expressly provided for by statute or regulation. Thus, authorizing legislation is critical in addressing both the section 8 and section 1 elements of a Charter challenge. Although the constitutional regimes are not directly comparable, it is noteworthy that several American lower courts have upheld the constitutionality of passive alcohol sensors. In particular, these courts have held that there is no reasonable expectation of privacy in an individuals naturally-exhaled breath, and no search is conducted simply because an officer uses his or her sense of smell when observing a suspect. Moreover, the American courts have upheld the use of various sense-enhancing devices and objects, such as binoculars, fluorescent powders and sniffing dogs. Accordingly, a passive alcohol sensor, which essentially provides officers with an enhanced sense of smell, will not likely offend the American Constitution. Indeed, it could be argued that passive alcohol sensors better protect drivers constitutional rights, because they provide objective results that are independent of the personal biases that may affect the officers unaided sense of smell. Many of these same considerations would be involved in determining whether the use of passive alcohol sensors violates section 8 of the Canadian Charter. Authority to Demand Standardized Field Sobriety Testing Standardized Field Sobriety Testing (SFST) was first developed in 1975 with funding from NHTSA. Early studies examined various roadside tests to determine which were most predictive of BACs above 0.10%. Based on the results of these studies, a set of roadside tests were developed which are now used in all 50 states to detect impaired drivers. This set of tests, commonly referred to as SFST, has three components: the Horizontal Gaze Nystagmus (HGN), Walk-and-Turn, and One-Leg Stand tests. Nystagmus, which is the rapid involuntary movements of the eyes, occurs naturally as the eyes gaze to the side. This eye movement is visibly affected by central nervous system depressants, such as alcohol. When a person is impaired by alcohol and attempts to track a slowly moving object, his or her eye will jerk perceptibly. The officer evaluates this reaction by observing a suspects eyes as he or she focuses on a slowly moving object, such as a pen or a penlight. Of the three SFST components, HGN is the most predictive of impairment, accurately detecting 77% of those with a BAC above 0.10%. Moreover, it is the test most trusted to reflect the suspects level of impairment, because the eye movement is an involuntary or autonomic reaction. Consequently, even seasoned drinkers cannot control their eye movements to the same degree that they can control their physical co-ordination. The Walk-and-Turn test is a divided attention task that most unimpaired people can easily perform. Because impaired persons have difficulty performing tasks requiring their attention to be divided between simple mental and physical exercises, this test identifies impairment on a very basic level. During the test, the suspect takes nine steps, heel-to-toe, along a straight line, following the oral instructions and physical demonstration provided by the officer. After taking the steps, the suspect turns on one foot and returns in the same manner in the opposite direction. The One-Leg Stand test is also a divided attention task easily performed by most unimpaired people. After listening to the instructions and watching the officers demonstration, the suspect stands with his or her heels together and arms down. The suspect is then instructed to stand on either leg, raising the other foot six inches off the ground and holding it out in front, while counting out loud for 30 seconds. Taken together, the three components of SFST can accurately predict BACs above 0.10% between roughly 85% and 90% of the time. Provincial legislation should authorize the police to demand participation in SFST from any driver they reasonably suspect has any alcohol or drugs in his or her body. A drivers refusal or failure to comply, without a reasonable excuse, should constitute a serious provincial offence, and result in a mandatory 90-day administrative licence suspension. The police should be specifically authorized to videotape the test. The successful implementation of a SFST program is dependent on police officers being adequately trained and certified, and having a record-keeping protocol in place. The ability to demand that a suspect participate in SFST would assist in impaired driving enforcement in several ways. First, SFST can be used as an alternative to an ASD test to establish the necessary grounds for demanding an evidentiary breath test on an approved instrument under section 254(3) of the Criminal Code. For example, the need for an alternative would arise if the officer did not have an ASD readily available. Moreover, some courts have indicated that police only have authority to demand an ASD test from a person who is operating or has care and control of a motor vehicle. Therefore, the police may not be able to demand an ASD test from a driver who has left or been removed from the immediate scene for safety, medical or other reasons. In both situations, the results of SFST, coupled with the officers observations, would provide a far stronger basis for demanding an evidentiary breath test than the officers observations alone. Second, unequivocal legislation would clarify the current uncertainty concerning both the police authority to demand that suspects participate in SFST and the consequences of a suspects refusal. Qubec is the only Canadian jurisdiction that specifically authorizes the police to demand field sobriety tests and imposes a penalty on drivers who refuse to participate. In other provinces, drivers are under no obligation to perform these tests, and cannot be penalized for refusing to do so. In these provinces, the police have common law authority to request participation in SFST, but a driver can refuse with impunity. General provisions in some of the provincial traffic acts have been interpreted as authorizing an officer to request that suspects submit to SFST, without first informing them of their right to counsel. However, in other provinces, the courts have held that the police must inform drivers of their right to counsel before they perform the test. Needless to say, counsel would likely advise the suspect not to consent to testing, unless provincial law required participation. Third, and probably of greatest importance, SFST can provide evidence establishing that the suspects ability to drive was impaired by alcohol and/or drugs. The police commonly complain that judges refuse to accept their evidence that the accuseds ability to drive was impaired. SFST, conducted by nationally certified officers using standard record-keeping procedures and supported by a videotape record, would greatly enhance the weight given to an officers testimony. Indeed, two-thirds of officers who participated in a recent Canadian survey felt that a videotape record of traffic stops would increase the likelihood of guilty pleas a great deal. The benefits of a video record were confirmed in a recent experiment conducted by the Oregon Department of Transportation, which provided video cameras to law enforcement agencies to record traffic stops. The study found that impaired driving cases involving video evidence were resolved sooner than those without such evidence (64 versus 80 days to disposition). Cases involving video evidence were also somewhat more likely to result in resolutions favourable to the state than cases without video evidence (86.7% versus 81%). Finally, the results and videotape of SFST may be essential if an approved instrument is unavailable, the evidentiary breath tests cannot be conducted within the prescribed time, or there is some other successful challenge to the evidentiary breath-test results. Similarly, the SFST evidence may be pivotal if a driver is impaired solely by drugs or by a combination of drugs and alcohol in circumstances in which his or her BAC is below 0.08%. Authority to Demand A Drug Evaluation and Classification Test INNOVATION Each province should enact legislation authorizing police to demand a drug evaluation and classification test from a driver they have reasonable grounds to believe is impaired by one or more drugs, or a combination of alcohol and drugs. A drivers failure or refusal to comply, without a reasonable excuse, should be a provincial offence. This recommendation is intended to address a gap in the current law regarding the apprehension, prosecution and conviction of drivers who are impaired by substances other than alcohol. Although the Criminal Code makes it an offence to operate a motor vehicle while ones ability to do so is impaired by drugs, enforcement of this provision is difficult, given the lack of drug detection procedures. While ASDs and approved instruments enable officers to identify and quantify the amount of alcohol in a suspects breath and blood, the Criminal Code provides for drug testing only in very limited circumstances. Moreover, the fact that a suspect tests positive for drugs does not necessarily mean that he or she is impaired. Presently, there is no standard or accepted procedure to identify and detain drug-impaired drivers. Therefore, as noted by the R.C.M.P., drug-impaired drivers go unreported and the courts dismiss drug-impaired driving charges because there are no means to test and provide the necessary proof. Accordingly, this recommendation seeks to provide police with a reliable means of detecting and establishing drug impairment. Studies from various jurisdictions indicate that drug use is not uncommon among drivers, particularly among those subsequently involved in fatal crashes. For instance, an Australian study found that drugs were detected in 22% of fatally injured drivers, 13% of whom had only drugs in their systems, and 9% of whom had a combination of drugs and alcohol. These findings are similar to a British Columbia study of 227 fatally injured drivers. It found that 9% of fatally injured drivers tested positive for drugs only, and 11% tested positive for a combination of alcohol and drugs. A more recent study of fatally injured Qubec drivers reported that 17.8% tested positive for drugs, and that 12.4% tested positive for both alcohol and drugs. In both Canadian studies, cannabis was by far the most prevalent drug, followed by cocaine and benzodiazepines. A significant percentage of young Canadians drive after smoking cannabis. In a 2001 Ontario high school survey, almost 20% of licensed grade 10 to OAC students reported driving within an hour of smoking cannabis. A recent Qubec study found that 24.3% of 16 to 19-year-old drivers and 22.4% of 20 to 24-year-old drivers who provided a sample in a nighttime roadside survey tested positive for cannabis. Given that Canadas youth population is increasing, and that young people are among the most likely to drive after using drugs, it is likely that drug-impairment among drivers will increase in the near future. Thus, it is increasingly important that the police be given broader legal authority to gather evidence of drug impairment. This recommendation includes three major components. First, the police should be authorized to demand that a suspected drug-impaired driver submit to standardized procedures to detect and classify drugs in his or her system. The tests performed under this protocol have been used for several decades in American jurisdictions and have been proven to assist in detecting drug impairment in motorists. Second, to improve the value of the officers evidence, police should be explicitly authorized to videotape the procedure. Finally, this recommendation encourages driver compliance by making it an offence to refuse to participate in a test without a reasonable excuse. Drivers who refuse to participate should also be subject to a 90-day administrative licence suspension. The drug evaluation and classification test has several distinct steps, beginning with preliminary questioning and a physical examination. This stage is meant to ensure that the suspect is not suffering from a medical condition that has symptoms similar to drug impairment. Next, the officer conducts various tests of the suspects eye movements, including both horizontal and vertical gaze nystagmus. This is done because certain drugs are known to cause nystagmus or to impede an individuals ability to converge (cross) his or her eyes. The suspect is then asked to perform divided attention tasks, such as the Walk-and-Turn and One-Leg Stand tests, which require him or her to balance, co-ordinate body movements, remember instructions, and perform more than one task at once. During the next phase, the officer examines the suspects blood pressure, body temperature and pulse, and tests the suspects pupils for reaction to varying light levels. The officer then examines the suspects muscle tone, since certain drugs can cause an individuals muscles to become either rigid or flaccid. The officer will also visibly inspect the suspects arms and ankles for injection sites. Finally, the officer conducts a structured interview with the suspect about the use of drugs. Based on the above tests, the officer forms an opinion regarding the suspects possible drug impairment, and attempts to determine the category of drugs causing the impairment. The test is designed to distinguish between seven categories of drugs: (i) Central Nervous System Depressants, such as alcohol and barbiturates; (ii) Inhalants, such as gasoline, glue and paint; (iii) Phencyclidines; such as PCP and ketamine (Special K); (iv) Cannabis; (v) Central Nervous System Stimulants, such as cocaine and amphetamines; (vi) Hallucinogens, such as LSD, ecstasy and mescaline; and (vii) Narcotic Analgesics, such as morphine, codeine and heroin. Based on his or her conclusion, the officer may demand that the suspect provide a blood, urine or saliva sample to confirm or refute that a specific drug is in the suspects system. This last test forms a separate recommendation, described below. The drug evaluation and classification test would be used in conjuction with the Criminal Codes existing impaired driving provisions. A police officer who suspects or has reasonable grounds to believe that a driver is impaired will likely demand that the driver provide a breath sample on an ASD or approved instrument, as the case may be. If the drivers BAC exceeds 0.08%, he or she will be charged under section 253(b) of the Criminal Code. However, if the drivers BAC is less than 0.08%, but he or she nevertheless appears to be impaired, the officer will then have grounds to believe that the driver is impaired by one or more drugs or a combination of alcohol and drugs. In such cases, the officer will demand that the driver participate in the drug evaluation and classification test. If the officer concludes, based on this test, that there are reasonable and probable grounds to believe that the suspect is impaired by a drug, he or she will demand a saliva, blood or urine sample to provide chemical confirmation of the presence of the drug. Since its inception by the Los Angeles Police Department in the 1970s, the drug evaluation and classification test has spread to almost all American states. This is due, in large part, to the reported accuracy rates of its practitioners. A double-blind study conducted by Johns Hopkins University found that drug recognition officers were over 90% accurate in determining impairment and the category of drug causing it. A later NHTSA double-blind study reported a 94% accuracy rate in identifying impairment by a drug other than alcohol. The participating officers correctly identified at least one drug other than alcohol in 87% of the cases. Like various other enforcement provisions, the drug evaluation and classification test will likely be challenged under the Charter. It will most likely be argued that requiring suspects to provide incriminating evidence under threat of punishment violates their right to silence and the privilege against self-incrimination contained in sections 11(c) and 13 of the Charter. However, as long as the suspects right to counsel is respected, compelled participation in the drug evaluation and classification test should not be found to violate these rights. In our view, the obligation to participate in this test is no more intrusive or incriminating than the requirement to provide breath or blood samples in impaired driving cases, which has already been upheld by the courts. Authority to Demand Blood, Urine and Saliva Samples for Drug Testing INNOVATION Section 253(a) of the Criminal Code makes it an offence to have care or control of a motor vehicle when ones ability to do so is impaired by alcohol or drugs. However, there are no federal provisions allowing police officers to effectively gather evidence of drug impairment. While blood samples lawfully obtained under the provisions of the Criminal Code may be tested for drugs, these provisions apply only in very limited circumstances and are extremely cumbersome to use. Therefore, MADD Canada recommends that each province enact provisions authorizing drug testing that are parallel to the Criminal Code breath and blood-testing provisions for alcohol. The drug tests would work in conjunction with SFST, drug evaluation and classification testing, and the existing Criminal Code provisions for testing on ASDs and approved instruments. As described earlier, a police officer who suspects or has reasonable and probable grounds to believe that a driver is impaired will first demand an ASD or evidentiary breath test in accordance with the Criminal Codes provisions. If the results indicate a BAC that is inconsistent with the drivers degree of impairment, the officer will demand that the driver participate in a drug evaluation and classification test. Finally, if the results of this test provide reasonable and probable grounds to believe that the driver is impaired by a drug, the officer will demand a saliva, blood or urine sample to provide chemical confirmation of the presence of the drug. This entire process is important because, in many cases, the chemical drug test will only indicate that a drug is present not when the drug was taken or whether the driver was impaired. Thus, the chemical test is best used to confirm or negate the officers finding in the evaluation and classification test, that the driver was impaired by a certain type of drug. The fact that the drug identified by the chemical test matches the officers conclusion and is consistent with the officers recorded observations of impairment strengthens the evidence that the driver was impaired by that drug at the time of driving. Nevertheless, police authority to demand bodily samples should not be limited to situations in which a drug evaluation and classification test has been performed. Rather, the legislation should authorize the police to demand a saliva, blood or urine sample from any driver who they have reasonable and probable grounds to believe is impaired by one or more drugs, or a combination of alcohol and drugs. While this belief may result from the drug evaluation and classification test, it may also arise from less formal observations of impairment, the presence of drugs and related paraphernalia, and witness statements. This would make the drug testing provisions comparable to the evidentiary breath testing provisions in section 254(3) of the Criminal Code. Failing to comply with this demand, without a reasonable excuse, should give rise to a mandatory 90-day administrative licence suspension and constitute a provincial offence. Consistent with section 10(b) of the Charter, the driver should be informed of, and allowed to exercise, his or her right to counsel prior to providing the sample. This would likely be unnecessary if the driver had already been so informed and given an opportunity to exercise his or her section 10(b) rights prior to the drug evaluation and classification test. Authority to Demand Samples from Drivers in Fatal or Personal Injury Crashes INNOVATION The provincial legislation should authorize the police to demand a breath sample on an ASD from any driver involved in a crash that results in death or bodily injury. Drivers who are incapable of providing a breath sample should be required to provide a blood sample for analysis. As noted above, these provisions should be made a mandatory and express condition of every drivers licence. Drivers who fail to provide requested breath or blood samples, without a reasonable excuse, should be subject to a mandatory administrative licence revocation and lengthy driving disqualification. In 1985, Parliament enacted the new offences of impaired driving causing death and impaired driving causing bodily harm, because impaired drivers who killed or injured others were only being convicted of impaired driving or driving with a BAC above 0.08%. These new offences carried lengthy maximum sentences and federal driving prohibitions that were believed to appropriately reflect the seriousness of the crimes. Unfortunately, Parliaments goals in enacting these offences have not been achieved. The available data establish that only a small fraction of impaired drivers who kill or injure are charged with the more serious offences. Moreover, the majority of those charged either escape criminal liability altogether or are only convicted of the lesser offences of impaired driving or driving with a BAC over 0.08%. This situation undermines the seriousness of impaired driving, angers and frustrates victims, defeats the purpose of the 1985 Criminal Code amendments, and protects those impaired drivers who commit the most serious crimes and represent the greatest risks to the public. A 1998 Alberta case illustrates some of the problems with the current provisions. The accused impaired driver, a 20-year-old, crossed the centre line and crashed head-on with an oncoming vehicle, killing four of its occupants and injuring the fifth. The crash occurred at 3:45 a.m. on a Sunday morning in early summer, and there were no apparent road, vehicle or weather problems. The attending officer suspected that the accused was impaired. When informed by witnesses that she had been drinking, the officer obtained a warrant for a blood test. However, even though the accused admitted in court to having consumed at least four drinks before driving, the judge ruled that the blood test ordered at the hospital was inadmissible because the police did not have reasonable and probable grounds to believe that she had committed an impaired driving offence. Since the blood test results were inadmissible, there was no evidence available to indicate the accuseds BAC at the time of the crash. Consequently, the four charges of impaired driving causing death and the one charge of impaired driving causing bodily harm were dropped. Sadly, this result is not unusual in crash situations. It has been estimated that alcohol is involved in about 40% of all motor vehicle fatalities and almost 20% of motor vehicle hospitalizations. These percentages would be far higher in the case of a late-night weekend crash in the summer involving a young driver. In many cases, the mere occurrence of a crash involving death or bodily injury should create a reasonable suspicion that at least one of the drivers had been drinking. Since police can already demand an ASD test when they have reasonable suspicion that a driver has consumed any alcohol, provincial legislation making ASD tests automatic in these crash situations would represent a relatively modest change in the current law, but one that would greatly simplify the law and improve enforcement. If the suspect failed the ASD test (BAC of 0.10% or higher), the police would then have the right to demand that the suspect submit to evidentiary breath testing under section 254(3) of the Criminal Code. This recommendation recognizes that the current roadside and evidentiary breath test provisions of the Criminal Code are generally unworkable in crash situations. For example, ASD tests must normally be performed at roadside and very soon after the suspect was driving. Evidentiary breath tests are also subject to strict time constraints and legal presumptions. The already restrictive three-hour time limit under section 254(3) is further constrained by the courts strict interpretation of an accuseds right to counsel under section 10(b) of the Charter. While the current breath-testing provisions of the Criminal Code make it inordinately difficult for police to gather BAC evidence in crash situations, the legislation in other countries facilitates the gathering of such evidence. Having reviewed this international legislation, MADD Canada recommends that the provinces grant police explicit statutory authority to demand a breath sample for analysis on an ASD from any driver involved in a fatal or personal injury crash, whether or not there is a reasonable suspicion that the driver has consumed alcohol. The authority to demand breath samples should encompass any person who was driving, or who the police reasonably suspect was driving or in control of a motor vehicle, at the time of a crash. To ensure that this proposal is effective, the police need express authority to demand an ASD test away from the roadside. If a driver is injured or taken to hospital as a result of the crash, or is otherwise incapable of providing or responding to a demand for a breath sample, the police should be authorized to arrange for a blood or urine sample to be taken from the suspect. As explained above, the Criminal Codes breath and blood-testing provisions are particularly ineffective in these situations. Similarly, as noted earlier, two American studies found that less than one-third of drivers with BACs of 0.10% or higher when admitted to hospital were charged with impaired driving, and that only half of those charged were convicted. Moreover, the more serious the drivers injuries, the less likely it was that he or she would be charged. In order to address these concerns, several Australian jurisdictions place medical personnel under a duty to take blood samples from patients who they suspect were driving at the time of their injuries. This ensures that BAC evidence will be gathered against all impaired drivers, even if police do not initially attend at the crash scene or hospital. In order to minimize the intrusiveness of our recommendations, we have not proposed such a duty in this report. Nevertheless, at a minimum, the police should have authority to demand ASD tests from drivers suspected of being involved in fatal or personal injury crashes. SECTION III: LICENCE SUSPENSIONS AND REVOCATIONS A Summary of Current Licence Suspensions in Canada It is important to distinguish at the outset between provincial licence suspensions and federal driving prohibitions. Under the Constitution Act, 1867, only the provinces and territories have legal authority to issue, suspend, revoke, or cancel drivers licences. Nevertheless, Parliament has authority under its criminal law power to impose driving prohibitions on those convicted of federal driving offences. The federal driving prohibitions for the impaired driving offences are set out in section 259(1) and (2) of the Criminal Code. It should also be noted that, while the provinces and territories impose suspensions in a broad range of circumstances, our discussion is limited to suspensions related to drinking and driving. Before proceeding to the specific licence suspension recommendations, it is necessary to outline the three types of licence suspensions related to drinking and driving that apply in most provinces and territories: short-term roadside suspensions; 90-day administrative suspensions; and suspensions for impaired and other Criminal Code driving offences. While the general characteristics of the suspensions are similar, there are numerous differences among the 13 jurisdictions. We have summarized below the most common features of each type of suspension. First, if a driver is believed to be impaired or has a BAC above 0.05% (0.04% to 0.06%, depending on the jurisdiction), the police are authorized or required to seize the drivers licence for typically 12 to 24 hours. In most cases, police will arrange a ride home for the driver, or the driver will be required to relinquish the wheel to a sober licensed passenger, if one is available. The driver is able to retrieve his or her licence from the police the following day. Most provinces do not keep records of these roadside suspensions and there are few, if any, consequences for repeat occurrences. These roadside suspensions are not dependent on the laying of federal criminal charges. Second, most provinces have created 90-day administrative licence suspensions (ALS) for drivers who, based on an evidentiary breath or blood test, have BACs of 0.08% or higher. Similarly, 90-day suspensions are imposed on drivers who fail, without a reasonable excuse, to provide a breath or blood sample upon a lawful demand. Typically, the police may issue a 90-day ALS whether or not they lay a federal criminal charge. The police will often impose a 90-day ALS in combination with a short-term roadside suspension. In these cases, the driver is issued an immediate roadside suspension, which prevents him or her from driving home in an impaired condition. In most jurisdictions, the driver is then given a temporary licence (usually 7 days), which only comes into effect once the short-term roadside suspension is over. The temporary licence allows the driver to make arrangements for alternate transportation before the 90-day suspension comes into effect. Most jurisdictions give the driver a right to challenge the officers decision to issue the ALS, but the grounds for such challenges are very limited. The ALS takes effect when the temporary licence expires. Third, the provinces and territories typically impose licence suspensions on drivers convicted of federal impaired or other Criminal Code driving offences. These provincial and territorial suspensions are in addition to any mandatory or discretionary driving prohibitions that have been imposed under the Criminal Code. It should be noted that the provincial and territorial licence suspensions for second and subsequent federal convictions are typically considerably longer than the federal driving prohibitions. Moreover, in most provinces and territories, the offender must satisfy various requirements before his or her licence is reinstated. The conditions may include: installing an alcohol interlock; attending an education program; undergoing an alcohol assessment; or successfully completing an alcohol treatment program. A single impaired driving occurrence may trigger all three suspensions, as well as a federal driving prohibition. The short-term roadside suspension applies immediately to remove the suspected impaired driver from the road. The 90-day ALS provides a more significant sanction and prohibits those charged with an offence from driving for at least part of the time prior to trial. If the driver is convicted, he or she is then subject to both the mandatory federal driving prohibition and the provincial licence suspension. 24-Hour Administrative Licence Suspensions The police should be required to suspend a drivers licence for 24 hours if: they reason-ably believe that his or her ability to drive is impaired by alcohol or drugs; they reasonably believe that the driver is unfit to drive for any reason; the driver registers a BAC of 0.05% or higher on a breath, blood or urine test; or the driver refuses, without a reasonable excuse, to provide a sample or submit to any required testing. If there is no sober, licensed passenger who is willing to drive and the vehicle cannot be safely parked, it should be towed and stored at the drivers or owners expense. When issuing a roadside suspension, police should have a duty to seize the drivers licence, and the driver should have a corresponding duty to submit his or her licence to the police. If no other suspension has been imposed, the driver should be allowed to retrieve his or her licence at the nearest police station after 24 hours. Police should also be required to keep a record of all 24-hour licence suspensions and file a report with the Registrar of Motor Vehicles. Such reports should become part of the drivers permanent record. A driver who accumulates two or more roadside suspensions within three years should be subject to additional sanctions, such as a lengthier administrative licence suspension. Moreover, the accumulation of roadside suspensions should result in a review of the drivers record by the Registrar, who may then decide whether the driver is unfit to drive or should be required to submit to an alcohol assessment or medical examination. 30-Day Administrative Licence Suspensions for Driving with a BAC of 0.05% or Higher INNOVATION Most Canadian jurisdictions already have an ALS program for drivers who have BACs of 0.08% or above, or who fail to provide breath or blood samples, without a reasonable excuse. Similarly, most jurisdictions have instituted short-term roadside licence suspensions for drivers with BACs of 0.05% or higher. However, this roadside suspension usually lasts no longer than 24 hours, and only Manitoba, Newfoundland, and Saskatchewan have record-keeping procedures and tiered penalties for repeat 0.05% BAC drivers. Thus, for most motorists, driving with a BAC between 0.05% and 0.08% entails little more than relinquishing the wheel to a sober driver or paying for a cab ride home. MADD Canada recommends that each province and territory enact a separate ALS program for drivers with BACs of 0.05% or higher. Studies consistently demonstrate that key driving-related skills are impaired at this BAC level, and that these drivers have significantly higher relative risks of crash than drivers with BACs of 0.00%. International jurisdictions that have introduced 0.05% laws have experienced significant declines in driver impairment and alcohol-related crashes. For example, the 0.05% restriction in New South Wales was estimated to have reduced fatal collisions by 8%, serious collisions by 7%, and single-vehicle nighttime collisions by 11%. This translated into the prevention of 75 fatal, 605 serious and 296 single-vehicle nighttime collisions per year. The 0.05% ALS should apply to all drivers who register a BAC of 0.05% or higher on a breath, blood or urine test. However, the program would work best in conjunction with ASD testing at systematic sobriety checkpoints. The ASDs in most provinces are already calibrated to register a warn at a 0.05% BAC level. If a driver blows a warn, police should immediately seize his or her licence and serve the driver with a notice that the licence will be suspended for 30 days. Police should have a duty to inform the driver of the right to challenge the results of the ASD test by submitting, without delay, to a test on an approved evidentiary breath-testing instrument. If the approved instrument indicates that the drivers BAC was below 0.05% at the time of driving, the drivers licence should be returned. However, the driver should also be warned that a result of 0.08% or higher on the evidentiary breath test will result in charges under section 253(b) of the Criminal Code. As well, drivers should be able to apply in writing to have the 30-day ALS reviewed by the Registrar of Motor Vehicles. The grounds for review should be limited to whether the drivers BAC reading was below 0.05%. Drivers suspended under the provincial 0.05% ALS legislation should be required to pay a $300 fee to cover the administrative costs of the program. In addition, police should be required to report the suspension and send the drivers licence to the Registrar of Motor Vehicles. If no other suspensions or conditions have been imposed, the driver should be permitted to obtain his or her licence from the Registrar at the end of the 30-day period. A record-keeping system should be implemented, with a mandatory review of the drivers record and additional sanctions imposed for drivers who accumulate multiple 30-day licence suspensions over a prescribed period of time. For example, drivers with two or more 30-day suspensions in five years should be required to undergo an alcohol assessment and complete any recommended remedial program prior to licence reinstatement. The proposed 30-day ALS legislation is similar to the 0.05% laws that currently exist in Australia and in many Western European and other countries. In these jurisdictions, violations of the 0.05% laws are subject to moderate fines and relatively short licence suspensions. These laws are akin to our provincial regulatory legislation and do not bear the social stigma or penalties of the Criminal Code impaired driving offences. Thus, the proposed provincial 0.05% ALS legislation would bring Canada into line with these countries. However, there is one major difference that should be noted. Unlike the legislation in these jurisdictions, the proposed provincial 0.05% ALS law would not create an offence. Rather, as in the case of the short-term roadside suspension, the provincial 30-day ALS law would simply create an administrative sanction. 90-Day Administrative Licence Suspensions ALS programs operate independently of criminal charges and have been shown to have significant traffic safety benefits. Studies of American ALS programs found that they reduce impaired driving among the general public, and among offenders both during and after their suspension periods. One study of ALS programs found that they resulted in a 3-14% overall reduction in nighttime crashes. Positive results were also reported for Ontarios and Manitobas ALS programs. For example, in the six years following the introduction of ALS and a vehicle impoundment program in Manitoba, there was a 12% net decrease in drinking-driver fatalities, and an approximate 32% decrease in overall impaired driving charges. There was also a 69% decrease in crash involvement among accused impaired drivers while subject to the ALS. The police should be required to issue an immediate 90-day ALS to any driver they have reasonable grounds to believe has a BAC of 0.08% or higher, based on a breath, blood or urine test. Suspensions should also be issued to drivers who, without a reasonable excuse, fail to submit to any test of impairment required by federal or provincial law. Although the driver should be entitled to a review of this suspension, the legislation should limit the grounds for a review. The Registrar should be required to confirm the suspension if he or she is satisfied that the drivers BAC was 0.08% or higher, or that the driver failed, without a reasonable excuse, to submit to the relevant test. If not, the Registrar should be required to revoke the suspension. In any case, an application for review should not delay the 90-day suspension from coming into effect. Again, the police should have a duty to seize the drivers licence, and the driver should have a corresponding duty to submit his or her licence to the police. The police should send the drivers licence to the Registrar, from whom the driver can collect it at the end of the 90-day period, if no other suspension has been imposed. Finally, the accumulation of multiple 90-day suspensions within a prescribed period should result in a mandatory review of the drivers record and additional sanctions. For example, those who accumulate two or more 90-day suspensions within ten years should be required to undergo an alcohol assessment and complete any recommended treatment prior to licence reinstatement. The ALS legislation should operate independently of any criminal proceedings. Thus, drivers should be subject to a 90-day licence suspension, regardless of whether they have been charged or convicted under the Criminal Code. Since the ALS program imposes an immediate administrative sanction, it will discourage suspects from delaying their criminal trials in an effort to keep their licences. In fact, the ALS legislation provides suspects who are likely to be convicted under the Criminal Code with an incentive to address the criminal charge as soon as possible. Drivers who delay the criminal proceedings risk having to serve the 90-day ALS separately from any subsequent federal driving prohibition and automatic provincial or territorial licence suspension. Automatic Provincial Licence Revocations for Criminal Code Offences Drivers convicted of certain Criminal Code driving offences are subject to federal driving prohibitions. However, the provinces should impose additional mandatory licence revocations on these offenders. The provincial sanctions should apply not only to the offences of driving while impaired, driving with a BAC above 0.08%, failing to provide samples, and impaired driving causing death or bodily harm, but also to dangerous driving, dangerous driving causing death or bodily harm, failing to remain at the scene of a crash, and driving while prohibited or suspended. These sanctions should also be imposed on drivers convicted of wilfully attempting to evade police pursuit, and manslaughter and criminal negligence if these offences arose from driving. It is important that criminal driving convictions result in a licence revocation, rather than a mere licence suspension. When a drivers licence is revoked, it is taken away indefinitely, and the onus is clearly on the individual to fulfil whatever conditions are imposed before re-applying for a licence. Conversely, licence suspensions often have fixed time limits and make reinstatement automatic at the expiry of the specified period. MADD Canada believes that licence revocation is necessary to drive home to criminal offenders the seriousness of impaired driving. Individuals who violate the Criminal Code should forfeit their driving privileges and should be required to establish why those privileges should be reinstated. As noted previously, a revoked licence should be seized by the police or licensing authorities to prevent people who are driving illegally from being able to present an apparently valid licence to the police. For a first offence of driving while impaired, driving with a BAC above 0.08%, failing to provide breath or blood sample, dangerous driving, driving while prohibited or suspended, and failing to remain at the scene of a crash, the offenders licence should be revoked for one year, after which he or she may apply for a new licence. A second conviction within ten years should result in a licence revocation and driving disqualification of three years. A third conviction within a ten-year period should result in a ten-year licence revocation and driving disqualification. Offenders should not be issued a new licence until they have successfully completed and paid for any required remedial program, including the installation of an alcohol interlock. A first conviction for one of the more serious Criminal Code offences impaired driving causing bodily harm or death, dangerous driving causing bodily harm or death, wilfully attempting to evade police pursuit, and manslaughter and criminal negligence causing bodily harm or death (when a motor vehicle is involved) should be subject to an automatic three-year licence revocation and driving disqualification. The disqualification period should only begin to run once the offender has finished serving any prison sentence. A second conviction within ten years should result in a ten-year revocation and driving disqualification. Again, offenders should only be allowed to re-apply for a licence if they successfully complete and pay for any prescribed remedial program and satisfy the Registrar that they do not pose an ongoing risk. The above revocations should not be subject to exceptions based on participation in education or treatment programs, or on claims of employment hardship. Studies show that restricted licences or work permits are an administrative burden and cannot be justified in terms of traffic safety. A recent Australian study examining the effectiveness of restricted licences found that offenders with restricted licences had a slightly higher rate of alcohol-related crashes than offenders who were prohibited from driving altogether. However, the restricted drivers were almost six times more likely to have a non-alcohol-related crash than the prohibited drivers. Moreover, a survey of Australian offenders who received restricted licences revealed that almost 30% admitted to successfully driving in breach of the conditions on their licences. This creates a significant enforcement burden for the police. More importantly, it decreases both the general and specific deterrent effects of the driving prohibition. Convicted offenders would not feel the entire weight of their penalties, as many would continue to drive for personal and social purposes. In addition, the general deterrent effect is undermined, because potential offenders will not perceive the penalty as either certain or severe. SECTION IV: VEHICLE SANCTIONS AND REMEDIAL PROGRAMS Alcohol Interlock Program Each province should establish an alcohol interlock program. An interlock is a small breath-testing device that is connected to the engine to prevent the vehicle from being driven if the drivers BAC is above a low pre-set level (usually 0.02% or 0.04%). The driver must blow into the instrument to provide a breath sample from which his or her BAC is determined. If the sample is above the pre-set level, the driver will not be able to start the vehicle or, depending on the type of interlock, set it in motion. The level should be set at 0.02% to reinforce the importance of separating drinking and driving. Setting the level at 0.04% may be interpreted as an approval of driving after drinking a message which is inappropriate, particularly for those already convicted of at least one drinking and driving offence. The current devices are quite sophisticated and include various anti-circumvention features. Interlocks contain computerized data logs that record all attempts to drive the vehicle, the drivers BACs and any efforts to tamper with the device. Moreover, research indicates that impaired driving offenders with interlocks on their vehicles have a significantly lower recidivism rate than offenders who do not. It has been estimated that alcohol interlocks should reduce impaired driving recidivism by better than 90% compared to licence suspensions. The use of interlocks should be a mandatory condition of licence reinstatement for all Criminal Code impaired driving offenders. After serving the required minimum term of the driving prohibition prescribed by the Criminal Code, drivers would, subject to the provincial revocation period, be eligible to participate in the interlock program. Provincial legislation should provide an incentive, in the form of a shortened licence revocation period, to encourage drivers to enter the interlock program. It has been estimated that less than 10% of eligible offenders participate in the Canadian and American interlock programs. As indicated, a majority of impaired driving offenders continue to drive, albeit perhaps only occasionally, during their suspensions. However, when they do so, they are over-represented in crashes. Consequently, there may be traffic safety benefits in encouraging offenders to install an interlock, even if this means shortening the licence suspension. While these offenders would likely drive more than if the suspension had remained in place, at least they would be much less likely to be driving while impaired. Nevertheless, it is important to balance the benefits of interlock programs with the proven benefits of licence suspensions as a general and specific deterrent. Studies consistently demonstrate that licence suspensions, more than any other sanction, reduce recidivism and subsequent crashes among impaired driving offenders. Thus, we would advise against using interlocks as a complete substitute for licence suspensions. Other ways of addressing the problem of driving while suspended need to be considered. For example, broader police enforcement powers, particularly authority to stop vehicles at random and demand documentation, and enhanced licence compliance checks will likely increase the actual and perceived risks of apprehension in driving while unlicensed or suspended. Similarly, stronger vehicle impoundment and forfeiture laws may help to deter such drivers. In our view, the fact that many suspended offenders continue to drive, at least occasionally, does not justify abandoning suspensions for interlock orders. Offenders who serve the full term of their suspensions should not be able to avoid interlock orders. Rather, these offenders should be required to install an interlock as a mandatory condition of licence reinstatement and as part of a comprehensive remedial program. Thus, MADD Canada recommends a hybrid initiative, involving both a voluntary interlock program (which provides an incentive to participate at an early stage) and a mandatory interlock program in which the use of an interlock is a condition of licence reinstatement. This approach should significantly increase participation in voluntary interlock programs, as well as help ensure that offenders do not graduate to an unrestricted licence until their underlying alcohol problems have been addressed. Furthermore, the Registrar of Motor Vehicles should have explicit authority to impose an interlock requirement on any driver that he or she reasonably believes poses a significant risk of driving while impaired. The driver should bear the cost of installing and maintaining the interlock. The minimum duration of the interlock order should be two years for a first offence, three years for a second offence within a ten-year period, and five years for a third offence within a ten-year period. Studies show that the positive effects of interlocks are often lost once the device is removed from the vehicle. For example, Ohio and North Carolina studies found that, once the interlocks were removed, the recidivism rates of offenders who had used interlocks were very similar to the rates of offenders who had not. The sharp rise in recidivism after removal of the device indicates a need to impose longer interlock orders and to incorporate interlocks into comprehensive remedial programs. Therefore, even after the prescribed interlock period ends, the interlock order should remain in place until the Registrar is satisfied that the offender no longer poses a significant risk of re-offending and that any underlying alcohol problem has been successfully addressed. In making this determination, the Registrar should be required to consider the offenders entire driving record, treatment history and, most importantly, his or her interlock data log. Research indicates that the data log is a critical tool in identifying individuals who have continuing alcohol problems and are likely to re-offend. For example, an Alberta study found that an offenders overall rate of positive BAC tests on the interlock was a strong indicator of post-interlock recidivism. The recidivism rate was even higher for offenders who had a pattern of positive BAC tests in the morning, which likely indicates very heavy consumption the preceding night. For the program to be effective, drivers subject to interlock orders should have ongoing supervision. Each province should establish an agency to monitor the offenders performance in the interlock program and prescribed remedial programs. Moreover, the interlock data log should be downloaded and reviewed every 30 to 60 days, and the accuracy of the machine should be checked. A pattern of attempts to drive with a prohibited BAC may indicate that the driver requires additional treatment and that the interlock should remain on the vehicle. Finally, the legislation should provide that driving an unequipped vehicle while subject to an interlock order constitutes driving while disqualified. Vehicle Impoundment and Immobilization As noted previously, many suspended and prohibited drivers continue to drive, at least occasionally, during the period of their licence suspension or revocation. Merely extending the length of the licence suspension may not be a sufficient deterrent for some offenders. Consequently, vehicle-based sanctions may be essential to discourage and at least temporarily prevent some unlicensed, disqualified and prohibited offenders from driving. Various North American impoundment or immobilization programs have shown positive results in terms of reduced recidivism and subsequent crashes. For example, an Ohio study found that offenders whose vehicles were not immobilized were twice as likely to re-offend as those whose vehicles were immobilized. In addition, it found that offenders whose vehicles were immobilized had lower recidivism rates for impaired driving after the end of the sanction, thereby demonstrating that immobilization programs have both a deterrent and a habituation effect. Similarly, drivers subject to Manitobas impoundment program had a 27% lower recidivism rate in the four years following their driving while suspended convictions, than drivers convicted before the impoundment law came into effect. Californias impoundment program was also reported to have reduced subsequent crashes and traffic convictions among affected offenders. In general, impoundment and immobilization programs have had much better results than other vehicle sanctions, such as zebra sticker laws. The police should be required to impound or immobilize any vehicle that they have reasonable grounds to believe is uninsured, or is being driven by an unlicensed, suspended, disqualified, or prohibited person. These drivers have shown that they are unwilling to respect provincial licensing laws, and pose a serious risk to public safety. The impoundment or immobilization period should be 45 days for a first occurrence. A second occurrence within three years involving the same owner or driver should result in a 90-day impoundment or immobilization. Further occurrences should result in vehicle forfeiture, as described in the next section. Both the driver and the owner should be liable for any towing, impounding, storage, or immobilization costs, which would constitute a lien on the vehicle. The towing and storage company should have the right to sell the vehicle to recover its costs if it has not been paid within 90 days of the end of the impoundment period. Owners of impounded vehicles should have the right to recover any costs that they have incurred from the culpable driver. Nevertheless, the vehicle should not be released before the end of the impoundment period unless the owner can prove that the car was taken without explicit or implicit permission, or that he or she took reasonable steps in attempting to verify that the driver had a valid licence. As described below, the provinces and territories should establish a system that assists vehicle owners in determining whether a prospective driver has a valid licence. Vehicle Forfeiture If a driver is responsible for three or more vehicle impoundments or immobilizations within a ten-year period, provincial legislation should authorize the forfeiture of his or her vehicle. Similarly, the vehicle of a driver who has been convicted of three or more impaired or other driving-related Criminal Code offences within a ten-year period should be forfeited. Although evaluations of vehicle forfeiture programs are scant, preliminary evidence from New York Citys program is promising. The City began confiscating vehicles driven by impaired drivers in 1999 under its power to confiscate property that is an instrumentality of a crime. Early police reports indicate that, in the first 10 months of the program, alcohol-related crashes decreased by 14.4%, alcohol-related fatalities decreased by 32.2%, and alcohol-related arrests declined by 18.3%. Drivers subject to forfeiture under this recommendation have shown a repeated willingness to endanger the public and violate vehicle licensing laws. Moreover, they often have no or limited third-party liability insurance, which puts the public at risk of serious financial loss in the event of a crash. Consequently, we believe that vehicle forfeiture is eminently justifiable. Provincial authorities should be permitted to take all steps necessary to prevent such drivers from gaining access to a vehicle. Licence Certification Program As a corollary to the vehicle impoundment and forfeiture programs, vehicle owners should have some means of verifying that a prospective driver has a valid licence. Often, the fact that a driver is able to produce what appears to be a valid licence is misleading. For example, a study of suspended drivers in the Greater Moncton Area found that 91% of suspended drivers stopped at a roadside check program were able to produce an apparently valid drivers licence for police. Drivers were only discovered to be suspended when their information was checked against a Department of Motor Vehicles computer. Given that vehicle owners may face impoundment, forfeiture and damage claims if they allow a suspended or unlicensed driver to use their vehicles, they should be able to obtain reliable information on the status of a drivers licence. This could be accomplished by providing a system whereby a driver could obtain, for a small administrative fee, certification of his or her licence as of a particular date from the Registrar of Motor Vehicles. Likewise, these certificates could be made available at the self-serve government kiosks that exist in some provinces. Vehicle owners who reasonably rely on such certification should be protected from the vehicle sanctions and adverse insurance consequences that would otherwise result from unwittingly allowing a suspended or unlicensed person to drive their vehicles. This system may be particularly useful to employers and car rental companies. Remedial Programs It has been estimated that as many as 30 to 75% of impaired driving offenders have serious alcohol problems. Many of these individuals are persistent drinking drivers and comprehensive remedial programs are needed to address their underlying alcohol problem. Therefore, the provinces and territories should establish comprehensive remedial programs involving three components education, assessment and treatment. Most education programs include information on the effects of alcohol and drugs on behaviour, the relationship between alcohol consumption and BAC, and the current impaired driving legislation. Some also provide information on making responsible lifestyle choices. The assessment programs are designed to determine if a driver has an alcohol or drug problem and, if so, the nature of that problem. The treatment programs are designed to assist the individual in recognizing that he or she has an alcohol problem, and in overcoming that problem. One study reported that between 50 and 65% of individuals receiving treatment show some evidence of improvement with their addictions at follow-up interviews. Half of these are either abstinent or have substantially reduced their consumption. While a remedial program may provide personal benefits for its participants, the primary goal is to minimize the likelihood that the individual will drive after drinking or taking drugs. Alcohol and/or drug assessment should be a mandatory condition of licence reinstatement for any driver who: is convicted of a federal impaired driving offence; receives two or more alcohol-related 30-day licence suspensions within five years; receives two or more alcohol-related 90-day ALS within ten years; or receives two or more mandatory administrative licence revocations for failing to take a required impairment test within ten years. Such drivers should also be required to successfully complete any prescribed education or treatment programs as a precondition to licence reinstatement. In addition, the Registrar should have discretion to require any driver who he or she reasonably believes has an alcohol or drug problem to undergo an assessment and successfully complete any recommended remedial program. The costs of participating in the assessment and education programs should be borne by the driver. However, the public health system should cover the cost of any medical treatment that an offender requires for his or her alcohol or drug problems. The first step is to ensure that these individuals undergo a professional alcohol and/or drug assessment. Then, depending on the results, the driver should be directed into an appropriate education or treatment program. However, drivers should not have their licences reinstated until the Registrar of Motor Vehicles is convinced that any underlying alcohol or drug problems have been adequately addressed. Successful completion of the prescribed remedial program should be a requirement, but not a guarantee, of licence reinstatement. The burden of proof should remain on the driver to satisfy the Registrar that he or she is fit to drive and will do so in a safe, sober and responsible fashion. Thus, for example, the mere fact that the individual attended all the required sessions falls short of discharging the requisite burden of proof. The legislation should clearly provide that participation in the remedial program is not an alternative to the existing federal or provincial sanctions, but rather, a requirement of licence reinstatement. Provinces should also institute a follow-up program to monitor each offenders progress after completing treatment. CONCLUSION In this model, we have attempted to outline the provincial and territorial legislative reforms that will most effectively reduce impaired driving in Canada. While these reforms are not exhaustive, they address three areas in which substantial progress needs to be made protecting young drivers, enhancing police enforcement powers, and developing effective measures for dealing with persistent drinking drivers and repeat offenders. Most of the recommendations are readily achievable, and many have already been implemented in various Canadian jurisdictions. Although some of the innovative proposals may require more careful design and drafting, we are confident that they, too, can be integrated into the existing licensing and enforcement systems. The provinces and territories should not view the model as a criticism of their current laws, but rather as a framework for legislative reform. We are committed to working with each jurisdiction and welcome opportunities to publicly support any government that demonstrates leadership. MADD Canada will be publishing brief annual assessments of the provincial and territorial traffic legislation in 2004 and 2005. We hope to be able to report in our next comprehensive review in 2006 that all of the jurisdictions have made great progress and that impaired driving deaths and injuries have fallen sharply.  The reader should assume that the word provinces includes the territories, unless otherwise stated.  R.S.C. 1985, c. C-46.  Taking Back Our Roads: A strategy to eliminate impaired driving in Canada (Mississauga: MADD Canada, 2001), online: (date accessed: 7 December 2002). MADD Canadas proposals included: introducing a 0.05% criminal blood-alcohol concentration (BAC) offence; enacting enhanced enforcement powers; redefining elements of some of the offences and defences; rationalizing sentencing; and developing more streamlined procedures for approving breath-testing instruments.  The Constitution gives provincial governments legislative authority over the administration of justice, property and civil rights, and the licensing of drivers. See Constitution Act, 1867 (U.K.), 30 & 31 Vict. c. 3, reprinted in R.S.C. 1985, App. II, No. 5, sections 92(13), (14) and (16). Taken together, these powers give the provinces broad authority over police enforcement powers and practices, civil liability, and the regulation of drivers and vehicle ownership. See, for example, Prince Edward Island (Provincial Secretary) v. Egan, [1941] 1 S.C.R. 5; Ross v. Canada (Registrar of Motor Vehicles), [1975] 1 S.C.R. 5; and Horsefield v. Ontario (Registrar of Motor Vehicles) (1999), 44 O.R. (3d) 73 (C.A.). Moreover, section 92(15) gives the provinces authority to create offences in relation to matters within their legislative competence. Provincial driving offences have been repeatedly upheld under this head of power. See, for example, Egan; Ross; and OGrady v. Sparling, [1960] S.C.R. 804.  These include comprehensive graduated licensing programs, mandatory alcohol interlocks, vehicle impoundment programs, and comprehensive assessment and treatment programs.  Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982, (U.K.), 1982, c. 11 [the Charter].  Traffic Safety Facts 2001, Young Drivers. (Washington: National Highway Traffic Safety Adminis-tration/National Center for Statistics and Analysis, 2002) at 5. For a review of the literature on the effects of minimum drinking age laws, see A.C. Wagenaar & T.L. Toomey, Effects of Minimum Drinking Age Laws: Review and Analyses of the Literature from 1960 to 2000 (2002) Supp. 14 J. Stud. Alcohol 206. The review reported that 98% of the relevant studies found that higher drinking ages reduce traffic crash rates (at 213).  See infra text at 17-18. As a result of these and other measures, 15 to 20-year-old drivers had the largest decrease (58%) in driving with BACs of 0.10% or higher among all age groups in the United States between 1982 and 1998. K. Stewart & J. Fell, Trends in Impaired Driving in the United States: Progress or Plateau in H. Laurell & F. Schlyter, eds., Proceedings of the 15th International Conference on Alcohol, Drugs and Traffic Safety T2000 (Stockholm: International Council on Alcohol, Drugs and Traffic Safety, 2000), online: (date accessed: 7 December 2002).  The minimum licensing age is 18 in Belgium, Denmark, Finland, France, Germany, Italy, Luxembourg, the Netherlands, Norway, Portugal, Russia, Spain, Sweden, and Switzerland. Austria and the United Kingdom have a minimum licensing age of 17. K. Stewart, On DWI Laws in Other Countries (Washington: National Highway Traffic Safety Administration, 2000), Table 4 [On DWI Laws in Other Countries].  In a 2001 survey of grade 7 to 13 students, 31.9% reported being a passenger in a car driven by a drunk driver in the preceding 12 months. E.M. Adlaf, R.E. Mann & A. Paglia, Drinking, cannabis use and driving among Ontario students (2003) 168(5) C.M.A.J. 565 [Adlaf].  In 1999, 72,925 people were charged with an impaired driving offence in Canada. This is a charge rate of 1 in every 287 licensed drivers. Canadian Centre for Justice Statistics, Cat. #85-205-XIE (Table 3.4) (Ottawa: Statistics Canada, 1999); and Canadian Motor Vehicle Traffic Collision Statistics 1999 (Ottawa: Transport Canada, 2001), online: (last modified: 30 May 2001). In comparison, the American charge rate for impaired driving is 1 in every 121 licensed drivers. Traffic Safety Facts 2000, Alcohol (Washington: National Highway Traffic Safety Administration, 2001), online: (date accessed: 8 December 2002). There is no reason to believe that this difference in the charge rate reflects a lower incidence of impaired driving in Canada. The per capita levels of alcohol consumption, the survey data on driving after drinking, and the percentages of fatally injured drivers who were impaired are very similar in Canada and the United States. For example, while American per capita consumption is 6.7 litres of pure alcohol annually, Canadian per capita consumption is 6.3 litres annually. Evolution of the world consumption of alcohol, online: (date accessed: 28 January 2003). In addition, a 1999 sur-vey found that 23% of Americans admit to driving within two hours of drinking within the past year, while a 2001 survey reported that 16.1% of Canadians have done so within the past 30 days. Traffic Tech No. 242 (Washington: National Highway Traffic Safety Administration, February 2001); and D.J. Beirness, H.M. Simpson & K. Desmond, The Road Safety Monitor 2002. Drinking and Driving (Ottawa: Traffic Injury Research Foundation, 2002) at 11 [Road Safety Monitor 2002].  We prefer not to use the term hard core drinking driver, as it is vague and used inconsistently. Moreover, the term is often invoked to describe a stark dichotomy between hard core drinkers and social drinkers, which tends to marginalize the impaired driving problem. See E. Chamberlain & R. Solomon, The Tooth Fairy, Santa Claus and the Hard Core Drinking Driver (2001) 7 Injury Prevention 272. Nevertheless, we recognize the need to tailor programs for those individuals who repeatedly drive after drinking, drive while suspended, or fail to be deterred by existing sanctions.  Transport Canada estimated that alcohol was a factor in more than 1,200 traffic fatalities during 1999. Road Safety Vision 2001 2000 Update (Ottawa: Transport Canada/Canadian Council of Motor Transport Administrators, 2001) at 10 [Road Safety Vision 2001]. For a more inclusive analysis of impaired driving deaths and injuries, see M. Marshall & B. Mercer, Estimating the Presence of Alcohol and Drug Impairment in Traffic Crashes and Their Costs to Canadians: A Discussion Paper (Toronto: MADD Canada, 2002) [Estimating the Presence of Alcohol and Drug Impairment]. Based on this report, MADD Canada has estimated that the total number of alcohol and drug-related crash deaths, including those occurring on and off the roads and on the water, is roughly 1500 to 1700 per year. The Magnitude of the Alcohol/Drug-Related Crash Problem in Canada Overview, online: (date accessed: 30 March 2003) [Magnitude]. In comparison, there were 538 homicides in 1999 (554 in 2001), including all murder, manslaughter and infanticide cases. Statistics Canada, Homicide Offences, Number and Rate (Ottawa: Statistics Canada, 2002), online: (last modified: 5 November 2002). Thus, alcohol and drug-related crashes claim more than two, and perhaps three times, as many lives each year as all homicides combined.  In 1997, a Health Canada report estimated that there are 74,000 alcohol-related traffic injuries in Canada annually. D.J. Beirness, D.R. Mayhew & H.M. Simpson, DWI Repeat Offenders: A Review and Synthesis of the Literature (Ottawa: Health Canada, 1997) at 11 [DWI Repeat Offenders]. This figure includes all personal injuries of any severity. Based on the M. Marshall and B. Mercer study, MADD Canada estimated that there were approximately 73,500 alcohol-related traffic injuries in 2002. Magnitude, supra note  NOTEREF _Ref39279574 \h 13. In comparison, during the 1997-98 period, there were approximately 80,000 court cases for violent Criminal Code offences (other than homicide) assault, aggravated assault, abduction, and sexual offences. The largest of these categories was assault, which accounted for 48,919 cases. C. Brookbank & B. Kingsley, Adult Criminal Court Statistics, 1997-98 (Ottawa: Canadian Centre for Justice Statistics, 1998) at 3. The number of charges is likely very much lower than the actual number of injuries caused by such criminal acts in Canada. Many cases are likely unreported or never reach the courtroom. However, even allowing for such unreported violent acts, impaired driving, with approximately 74,000 injuries, still appears to rank among the leading criminal causes of injury in Canada.  For a detailed review of the statistics, see infra the subsections entitled, Graduated Licensing Program and A 0.008% BAC Limit for all Drivers Under 21. See also Traffic Injury Research Foundation website, (date accessed: 7 December 2002).  Hunter v. Southam Inc., [1984] 2 S.C.R. 145.  See infra notes  NOTEREF _Ref39317654 \h 215- NOTEREF _Ref24461412 \h 216.  These checkpoints are known as, among other things, R.I.D.E. Programs, Counter Attack, and Operation Red Nose, depending on the jurisdiction. They are described in greater detail in the subsection entitled Authority to Establish Systematic Sobriety Checkpoints.  In particular, driving is not a liberty interest protected under section 7 of the Charter. See R. v. Pontes (1995), 100 C.C.C. (3d) 353 (S.C.C.); Horsefield, supra note  NOTEREF _Ref27041912 \h  \* MERGEFORMAT 4; and Buhlers v. British Columbia (Superintendent of Motor Vehicles) (1999), 170 D.L.R. (4th) 344 (B.C. C.A.).  Currently, only fatally injured drivers are included in the Fatality Database, from which statistics on alcohol-related crashes in Canada are derived. This means that, in a crash where the sober driver is killed but the impaired driver survives, it will be the sober drivers BAC that is recorded. As a result, the fatality will likely be entered into the Database as non-alcohol-related. This problem leads to the under-estimation of alcohol involvement in fatal crashes. See Collection of Data on Alcohol Use Among Persons Fatally Injured in Motor Vehicle Accidents in Canada During 2000, Progress Report (Ottawa: Traffic Injury Research Foundation, 2001); and Traffic Injury Research Foundation, STRID 2001 Monitoring Report: Progress in 1998 and 1999 (Ottawa: Canadian Council of Motor Transport Administrators and Transport Canada, 2000) at 3.  Raising the minimum driving age to 18 across Canada would significantly reduce crash deaths and injuries among teenagers in several ways. Obviously, it would prevent 16 and 17-year-olds from driving lawfully, thereby greatly reducing their likelihood of driving. It would also appear to reduce unlicensed driving among very young beginners. Finally, it would ensure that lawful beginners were two years older and more mature than current beginners. Nevertheless, we do not recommend adopting such a policy at this time because it is unlikely to be supported by the public and because similar benefits may be achieved by adopting a comprehensive graduated licensing system.  See H.L. Ross, Confronting Drunk Driving: Social Policy for Saving Lives (New Haven: Yale University Press, 1992) at 131.  S.A. Ferguson et al., Differences in Young Driver Crash Involvement in States with Varying Licensure Practices (1996) 28 Accid. Anal. and Prev. 171 at 174 and 177.  Ibid.  For another American study supporting the view that a young licensing age may contribute to illegal driving among youth, see A. F. Williams et al., Analysis of the Fatal Crash Involvement of 15 Year-old Drivers (1997) 28 J. Safety Research 49. This study found that, in states that had a minimum licensing age of 15, 73% of the 15-year-olds involved in fatal crashes were driving without a licence or without the required supervision.  One study reported that, after only a few years of graduated licensing, Ontario beginning drivers experienced a 31% reduction in the crash rate. D.R. Mayhew & H.M. Simpson, Youth and Road Crashes: Reducing the Risks from Inexperience, Immaturity and Alcohol (Ottawa: Traffic Injury Research Foundation, 1999) at viii [Youth and Road Crashes]. Preliminary results of the Ontario graduated licensing program were also reported in Saving Teenage Lives: The Case for Graduated Licensing (Washington: National Highway Traffic Safety Administration, 1999), online: (date accessed: 10 November 2002). The results among 16-year-old drivers were particularly dramatic, as their fatal crash rate dropped from 6.6 deaths per 10,000 licensed drivers in 1993 (before the system came into force) to 1.8 deaths per 10,000 licensed drivers in 1995 (after the system came into force).  A Traffic Injury Research Foundation (TIRF) study reported a 37% reduction in the total number of collisions among 16-year-old drivers during the first three years of Nova Scotias graduated licensing program. This reduction was not limited to young drivers, as the collision rate among new drivers aged 25 and older dropped by 42.7%. D.R. Mayhew, H.M. Simpson & M. des Groseilliers, Impact of the Graduated Driver Licensing Program in Nova Scotia (Ottawa: Traffic Injury Research Foundation, 1999).  Qubecs implementation of a revised graduated licensing system in 1997 helped to reduce crash fatalities and injuries among new drivers. J. Bouchard et al., The Qubec Graduated Licensing System for Novice Drivers: A Two-Year Evaluation of the 1997 Reform in H. Laurell & F. Schlyter, eds., Proceedings of the 15th International Conference on Alcohol, Drugs and Traffic Safety T2000 (Stockholm: International Council on Alcohol, Drugs and Traffic Safety, 2000), online: (date accessed: 7 December 2002). See also R. Simard et al., The New Graduated Licensing System in Qubec: Impact on the Number of New Drivers and on Nighttime Single Crashes in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM.  Deaths and injuries among 15 to 17-year-old drivers in New Zealand reportedly dropped by about 25% after its graduated licensing program was introduced. H.M. Simpson & D.R. Mayhew, Reducing the Risks for New Drivers: A Graduated Licensing System for British Columbia (Victoria: Ministry of the Attorney General, Motor Vehicle Branch, 1992) at 27 [Reducing the Risks]. See also J.D. Langley, A.C. Wagenaar & D.J. Begg, An Evaluation of the New Zealand Graduated Driver Licensing System (1996) 28 Accid. Anal. and Prev. 139. More recently, a review of Californias graduated licensing system, introduced in 1998, reported positive results. S.A. Bloch, H-C. Shin & S.N. Labin, Does Graduated Driver Licensing Reduce Drinking and Driving?: An Examination of Californias Teen Driving Restrictions in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM. The program helped to reduce both alcohol-related and other crashes among young drivers. Fatal and personal injury crashes decreased more than crashes involving only property damage.  See, for example, Reducing the Risks, ibid. at 9-11; and N.P. Gregersen et al., Sixteen Years Age Limit for Learner Drivers in Sweden an Evaluation of Safety Effects (2000) 32 Accid. Anal. and Prev. 25.  As indicated, the minimum age for applying for a stage-one licence should be 16.  See A.F. Williams & D.R. Mayhew, Graduated Licensing: A Blueprint for North America (Ottawa: Traffic Injury Research Foundation, 2002) at 1; L. Chen et al., Carrying passengers as a risk factor for crashes fatal to 16- and 17-year-old drivers (2000) 283 J.A.M.A. 1578; and S.T. Doherty, J.C. Andrey & C. MacGregor, The situational risks of young drivers: the influence of passengers, time of day and day of week on accident rates (1998) 30 Accid. Anal. and Prev. 45.  J.R. Treat et al., Tri-Level Study of the Causes of Traffic Accidents: Final Report, vol. 1 (Washington: National Highway Traffic Safety Administration, 1979).  Study: More Deaths with Young Drivers Associated Press (21 June 2000) [Ford Study]. See also D.F. Preusser, S.A. Ferguson & A.F. Williams, The Effect of Teenage Passengers on the Fatal Crash Risk of Teenage Drivers (1998) 30 Accid. Anal. and Prev. 217; and A.F. Williams, Teenage Passengers in Motor Vehicle Crashes: A Summary of Current Research (Arlington: Insurance Institute for Highway Safety, 2001).  A Canadian study of youth and road crashes found that nearly 80% of fatally injured teenage passengers were travelling in a vehicle operated by a young driver. Youth and Road Crashes, supra note  NOTEREF _Ref27037342 \h  \* MERGEFORMAT 26 at vi. Similarly, the Ford Study, supra note  NOTEREF _Ref31550738 \h 34, found that 70% of 14 and 15-year-old passen-gers killed in motor vehicle crashes were travelling with a driver who was 21 or younger. Fourteen per cent were passengers in vehicles driven by someone aged 12-15, who was almost always unlicensed.  In an 18-month study of all impaired driving charges in Sudbury, Ontario, the peak time period for impaired driving was between midnight and 4 a.m. This period accounted for 58% of all charges. L. Anglin et al., A Study of Impaired Drivers Stopped by Police in Sudbury, Ontario (Toronto: Addiction Research Foundation/Sudbury Regional Police, 1997) [Sudbury Study].  An American study indicated that three-quarters of nighttime crashes and more than half of the nighttime fatalities among 16 and 17-year-olds occur between 9 p.m. and midnight. A.F. Williams, Protecting New Drivers: 10 Components of Graduated Licensing that Make Sense (Arlington: Insur-ance Institute for Highway Safety, 1996) at 5. In Canada, about half of all teenage motor vehicle deaths and 64% of all such injuries occur at night, even though teens drive less during nighttime hours. Youth and Road Crashes, supra note  NOTEREF _Ref27037342 \h  \* MERGEFORMAT 26 at 14. In Ontario, drivers between the ages of 16 and 24 account for only 30.5% of all nighttime drivers, but 47.7% of nighttime driver deaths. Thus, their relative risk of a nighttime crash is significantly higher than that of older drivers. D.R. Mayhew & H.M. Simpson, New to the Road. Young and Novice Drivers: Similar Problems and Solutions? (Ottawa: Traffic Injury Research Foundation, 1990) at 61 [New to the Road].  Obviously, in provinces or territories where no such roads exist, we would not expect such restrictions to be enacted.  In addition, young drivers are more likely to be in single-vehicle crashes than older drivers. These typically occur on highways, freeways and roads with high speed limits. New to the Road, supra note  NOTEREF _Ref31550835 \h  \* MERGEFORMAT 37 at 32. Unfortunately, road restrictions for new drivers have not been widely implemented. Therefore, the impact of these restrictions is not well documented. Reducing the Risks, supra note  NOTEREF _Ref24703662 \h  \* MERGEFORMAT 29 at 33. When Ontario prohibited beginning drivers from driving on 400-series highways (high-speed), there was a 61% decrease in the collision rate for such drivers on these roads. P. Boase & L. Tasca, Graduated Licensing System Evaluation, Interim Report 98 (Toronto: Ministry of Transportation of Ontario, 1998) at 4. Of course, if the road restriction had been fully obeyed, the collision rate for beginning drivers on these highways would have been zero.  For example, a large experimental study in DeKalb County, Georgia had disappointing results. The DeKalb study, undertaken in the late 1970s and early 1980s, was intended to demonstrate the benefits of enhanced high school driver education. Sixteen thousand students were divided into three groups. The first participated in the Safe Performance Curriculum, a state-of-the-art education program lasting 72 hours. The second group took a basic driver education course called Pre-Driver Licensing, lasting only 20 hours. The control group was not formally enrolled in driver education. The results showed that driver education had only a small, short-lived positive effect on new drivers. Although students with driver education initially had fewer crashes per licensed driver than the control group, this difference was not sustained beyond six months of licensed driving. J.R. Stock et al., Evaluation of Safe Performance Secondary School Driver Education Curriculum Demonstration Project: Final Report (Columbus: Batell Columbus Laboratories, 1983). A more recent review con-firmed that the preponderance of research does not indicate that formally-trained drivers have a lower crash rate than drivers without such training. See Youth and Road Crashes, supra note  NOTEREF _Ref27037342 \h  \* MERGEFORMAT 26; and also D.R. Mayhew & H.M. Simpson, The Safety Value of Driver Education and Training (2002) 8 Injury Prevention, (Suppl II): ii3-ii8. Another study found that driver education may result in participants significantly over-estimating their skills. Young drivers taking a mandatory skid training course in Sweden were divided into two groups: insight and skills training. The strategy used with the insight group was to make drivers aware of their limited and unpredictable braking and avoidance skills. Conversely, the skills group practised braking and avoidance manoeuvres repeatedly around the same course, at increasing speeds. While the groups performed equally in terms of actually observed skill, the skills group had a higher subjective estimation of their driving capabilities. Thus, the author suggested that, in order to be effective, driver education programs must reinforce to new drivers that their skills are limited and that they need to drive with larger safety margins. The author also noted that most young drivers believe themselves to be more skilled than other drivers and, therefore, underestimate their risk. See N.P. Gregersen, Young Drivers Overestimation of their Own Skill An Experiment on the Relation Between Training Strategy and Skill (1996) 28 Accid. Anal. and Prev. 243 at 245-48. See also L. Potvin, F. Champagne & C. Laberge-Nadeau, Mandatory Driver Training and Road Safety: The Quebec Experience (1988) 78 A.J.P.H. 1206; L.S. Robertson, Crash Involvement of Teenaged Drivers when Driver Education is Eliminated in High School (1980) 70 A.J.P.H. 599; and N.P. Gregersen, Systematic Co-operation Between Driving Schools and Parents in Driver Education, an Experiment (1994) 26 Accid. Anal. and Prev. 453.  We acknowledge that such a requirement would be difficult to enforce, in that parents could sign the relevant forms without the driver having completed the required hours of supervised driving. However, it would likely raise the consciousness of parents and new drivers about the importance of supervised driving. In any event, such a requirement would not have any negative effects, even if falsified by some.  In 1997, 40% of all teenaged drivers killed in road crashes in Canada had been drinking. Three-quarters of these drivers had a BAC above 0.08%, and 44% had a BAC in excess of 0.15%. Youth and Road Crashes, supra note  NOTEREF _Ref27037342 \h  \* MERGEFORMAT 26 at vii. Similarly, figures from 2000 indicate that 39.2% of 16 to 19-year-old crash fatalities were alcohol-related. See D.R. Mayhew, S.W. Brown & H.M. Simpson, The Alcohol-Crash Problem in Canada: 2000 (Ottawa: Traffic Injury Research Foundation/Transport Canada, 2002) at 14.  Research on unsafe driving behaviour has shown that perceptual, cognitive and vehicle-handling skills are less developed in beginning drivers than in experienced drivers. In addition, beginning drivers have less ability to detect and recognize imminent hazards in the driving environment. J.A. Groeger & I.D. Brown, Assessing Ones Own and Others Driving Ability: Influence of Sex, Age and Experience (1989) 21 Accid. Anal. and Prev. 155 [Groeger].  Reducing the Risks, supra note  NOTEREF _Ref24703662 \h  \* MERGEFORMAT 29 at 6. Presumably, this latter figure was drawn from jurisdictions in which the minimum driving age is 15 or younger.  Because single-vehicle nighttime fatal crashes have such a high rate of alcohol involvement, they are often used as a surrogate measure for estimating alcohol-related crashes.  R. Hingson, T. Heeren & M. Winter, Lower Legal Blood Alcohol Limits for Young Drivers (1994) 109 Public Health Reports 738 [Hingson, Heeren & Winter].  Ibid. at 744. In addition, one study concluded that BAC restrictions on young drivers have even greater results if they are combined with extensive public education campaigns. For instance, while Marylands 0.02% BAC restriction resulted in a 21% decrease in the number of young crash-involved drivers judged to have been drinking, the addition of a public education campaign resulted in a further 30% decrease. R.D. Blomberg, Lower BAC Limits for Youth: Evaluation of the Maryland .02 Law (Washington: National Highway Traffic Safety Administration, 1992) at 67. See also A.C. Wagenaar et al., Lowered legal blood alcohol limits for young drivers: Effects on drinking, driving, and driving-after-drinking behaviours in 30 states (2001) 91 A.J.P.H. 801.  J.H. Lacey, R.K. Jones & C.H. Wiliszkowski, Zero Tolerance Laws for Youth: Four States Experience (Washington: National Highway Traffic Safety Administration, 2000) at 24 [Lacey et al.].  A survey of licensed grade 11 and 12 students in Ontario before and after the introduction of graduated licensing reported a 25% reduction in the number of males who reported driving after drinking any alcohol. R.E. Mann et al., Graduated Licensing in Ontario: Impact of the 0 BAL Provision on Adolescents Drinking-Driving in C. Mercier-Guyon, ed., Alcohol, Drugs and Traffic Safety, T97 (Annecy, France: Centre dtudes et de recherche en mdecine du trafic, 1997) at 1055. Moreover, the graduated licensing system has been reported to have had some effect on attendance at heavy drinking events, such as bush parties, among affected drivers. In a relatively recent survey of Ontario students, only 38.4% of drivers in the graduated licensing program reported attending a bush party in the last 12 months, down from 57% prior to the introduction of the program. G. Stoduto, E.M. Adlaf & R.E. Mann, Adolescents, Bush Parties and Drinking-Driving (1998) 59 J. Stud. Alcohol 544 at 546.  Since studies show that the risk of collision declines significantly after two years of driving experience, a minimum two-year period is justifiable. See, for example, D.R. Mayhew, H.M. Simpson & A. Pak, Changes in Collision Rates among Novice Drivers during the First Months of Driving (Arlington: Insurance Institute for Highway Safety, 2000) at 14. However, because the risk continues to decline during the first five years of driving, an even longer period of supervised or restricted driving may also be warranted. Reducing the Risks, supra note  NOTEREF _Ref24703662 \h  \* MERGEFORMAT 29 at 48.  In some jurisdictions, this official is known as the Superintendent of Motor Vehicles. For ease of reference, we will use the term Registrar of Motor Vehicles to refer to all such officials in the remainder of this report.  Reducing the Risks, supra note  NOTEREF _Ref24703662 \h  \* MERGEFORMAT 29 at 16.  See supra note  NOTEREF _Ref27141668 \h 15.  Youth and Road Crashes, supra note  NOTEREF _Ref39319539 \h 26 at 5.  Ibid. at 6-7. Young drivers are also at far greater risk of being killed per kilometre driven. In 2001, 43.4 Canadian drivers aged 16 to 19 were killed per billion kilometres driven. This is more than twice the rate of 20 to 24-year-old drivers, seven times the rate of 25 to 34-year-old drivers, and twelve times the rate of 34 to 44-year-old drivers. Canadian Council of Motor Transport Administrators, Road Safety Vision 2010 The 2001 Update (Ottawa: Minister of Public Works and Government Services, 2002) at 15 [Road Safety Vision 2010 The 2001 Update]. For comparable American data, see A. F. Williams, Nightime Driving and Fatal Crash Involvement of Teenagers (1985) 17 Accid. Anal. and Prev. 1.  Youth and Road Crashes, supra note  NOTEREF _Ref39319539 \h 26 at 8-9.  Ibid. at 19. Of the fatally injured drinking drivers, 75% had a BAC above 0.08%, and 44% had a BAC above 0.15%. Ibid. at vii.  Ibid. at 22.  Canadian Centre on Substance Abuse (CCSA), and Centre for Addiction and Mental Health (CAMH), Canadian Profile: Alcohol, Tobacco and Other Drugs, Table 2.6. Type of drinker by age (Ottawa: CCSA, 1999) at 41 [Profile].  Canadian Institute for Health Information (CIHI), Health Indicators, May 2002, Catalogue no. 82-221-XIE, Volume 2002, No. 1, Frequency of drinking 5 or more drinks (Ottawa: Statistics Canada, 2002) [Health Indicators, May 2002].  Almost 86% of 20 to 24-year-olds reported drinking in the past year. Profile, supra note  NOTEREF _Ref39319833 \h 59.  Among 20 to 24-year-olds who drink, 77% of the males and almost 62% of the females report binge drinking in the past 12 months, with just over 37% doing so 12 or more times per year. Health Indicators, May 2002, supra note  NOTEREF _Ref39319863 \h 60.  See supra text at 13, 17-18.  Youth and Road Crashes, supra note  NOTEREF _Ref39319539 \h 26 at 21.  Ibid. at 22.  Beginning drivers have a decreased ability to detect and recognize imminent hazards in the driving environment. They are less aware of what constitutes a hazard, and their lack of experience in diverse driving situations makes them more likely to respond inappropriately to hazards. See Groeger, supra note  NOTEREF _Ref24773057 \h  \* MERGEFORMAT 43.  Some researchers have suggested that this dual lack of experience leads to a far greater inability to drive after drinking. D.R. Mayhew et al., Youth, Alcohol and Relative Risk of Crash Involvement (1986) 18 Accid. Anal. and Prev. 273 at 283 [Relative Risk]. Many young drivers may not have acquired a tolerance for even low doses of alcohol. Thus, they cannot compensate adequately for the alcohols effects on driving. In addition, inexperienced drivers must focus more attention on the task of driving, whereas experienced drivers perform the task automatically or instinctively. Consequently, even a small amount of alcohol can substantially decrease an inexperienced drivers ability to drive safely.  Research indicates that young drivers are more likely to speed, follow too closely, allow less time to merge with traffic, cross traffic lanes, and pass other vehicles. They also tend to over-estimate their driving abilities. J. Bergeron, Behavioural, attitudinal and physiological characteristics of young drivers in simulated driving tasks as a function of past accidents and violations (Paper presented at the New to the Road Symposium, Halifax, 17-20 February 1991). Moreover, they are less likely to wear seatbelts. See also J.J. Arnett, D. Offer & M.A. Fine, Reckless Driving in Adolescence: State and Trait Factors (1997), 29 Accid. Anal. and Prev. 57 at 59-60.  See P.L. Zador, S.A. Krawchuk & R.B. Voas, Alcohol-Related Relative Risk of Driver Fatalities and Driver Involvement in Fatal Crashes in Relation to Driver Age and Gender: An Update Using 1996 Data (2000) 61 J. Stud. Alcohol 387 at 390 [Zador]; and R.G. Stewart & R.W. Sanderson, The Measurement of Risk on Canadas Roads and Highways in S. Yager, ed., Transport Risk Assessment (Waterloo: University of Waterloo Press, 1984).  Canadian researchers have also documented the greater relative risks of death among young drinking drivers. See Relative Risk, supra note  NOTEREF _Ref27039357 \h  \* MERGEFORMAT 67 at 282-83; and R.E. Mann et al., Assessing the Potential Impact of Lowering the Legal Blood Alcohol Limit to 50 mg% in Canada (Toronto: Addiction Research Foundation, 1998) at 20 [Potential Impact].  Relative Risk, supra note  NOTEREF _Ref27039357 \h 67 at 281-83.  Lacey et al., supra note  NOTEREF _Ref30678626 \h  \* MERGEFORMAT 48 at 29.  The number of 15 to 19-year-olds in Canada is expected to increase 12.2% by 2011, to a percentage of the population comparable to that of the late 1970s and early 1980s. Youth and Road Crashes, supra note  NOTEREF _Ref39319539 \h 26 at 2.  See supra note  NOTEREF _Ref31552096 \h  \* MERGEFORMAT 16.  It should be noted that these licence sanctions would not apply to drivers who commit minor traffic violations. Suspending the licences of such drivers would only detract from their ability to improve their skills and gain experience. Rather, the suspensions would apply to drivers who deliberately violate key conditions of their licences, or who show a pattern of disregarding licensing laws and traffic safety. It has been suggested that young drivers do not necessarily apply the skills they learn in driver education programs. Therefore, it is important to motivate new drivers to demonstrate their skills through clean driving records, rather than allowing them to graduate to a full licence without regard to traffic violations. D.R. Mayhew et al., Effectiveness and Role of Driver Education and Training in a Graduated Licensing System (1998) 19 J. Public Health Policy 51 at 57.  R.P. Compton, R.G. Ulmer & V.I Shabanova, Effectiveness of the Use and Lose Law in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM.  A licence suspension is seen as a more serious punishment than a fine, for example, which has less direct impact on teenagers. In addition, few teens are sent to prison for breaching highway traffic or alcohol possession legislation. Ibid.  Ibid.  Criminal Code, supra note  NOTEREF _Ref28663687 \h  \* MERGEFORMAT 2, s. 254(2).  Ibid., s. 254(3).  Ibid., ss. 254(5) and 255(1).  Ibid., s. 258(1)(c) and (g).  Ibid., s. 254(3)(b)(ii).  Ibid., ss. 254(5) and 255(1).  R. v. MacMillan (1989), 19 M.V.R. (2d) 137 at 142 (P.E.I. S.C.); and R. v. Brooke (1999), 39 M.V.R. (3d) 293 (Alta. Prov. Ct.).  There is nothing in the Criminal Code requiring or authorizing doctors to provide such confidential patient information to the police.  The provincial legislation regulating health practitioners typically makes any unauthorized disclosure of confidential patient information, without the patients consent, grounds for a finding of professional misconduct. See, for example, Regulated Health Professions Act, S.O. 1991, c. 18, s. 36(1); and M.T. v. College of Physicians and Surgeons (Nova Scotia), [1996] N.S.J. No. 256 (N.S. S.C.), online: QL (NSJ). Although this area of the law is evolving, the wrongful disclosure of confidential patient information may also give rise to a civil suit. See, for example, X v. Y, [1988] 2 All E.R. 648 (Q.B.); and Hollinsworth v. BCTV (1996), 34 C.C.L.T. (2d) 95 (B.C. S.C.). Finally, the unauthorized disclosure of confidential information, without the patients consent, may render the test results inadmissible for violating section 8 of the Charter. See R. v. Dyment, [1988] 2 S.C.R. 417; R. v. Dersch, [1993] 3 S.C.R. 768; and R. v. Erickson (1992), 72 C.C.C. (3d) 75 (Alta. C.A.), affd (1993), 81 C.C.C. (3d) 447 (S.C.C.).  See Criminal Code, supra note  NOTEREF _Ref28663687 \h 2, s. 254(3), and s. 258(1)(d), (d.1) and (h).  Section 257(1) and (2) of the Criminal Code specifically exempts doctors from criminal liability for refusing to draw a blood sample. It also exempts doctors from civil and criminal liability if they draw a sample. Despite these latter protections, some doctors refuse to become involved, and there is nothing that the police can do other than attempt to find a doctor who is willing to assist them.  McMillan, supra note  NOTEREF _Ref39410271 \h 85; and R. v. Evans (1990), 9 W.C.B. (2d) 639 (B.C. Co. Ct.).  See, for example, MacMillan, supra note  NOTEREF _Ref39410271 \h 85 (impracticability must relate directly to the suspects physical condition at the time); R. v. Campeau (1990), 24 M.V.R. (2d) 191 (Ont. Gen. Div.) (breath test is impracticable if delay would interfere with the suspects medical treatment); R. v. Lipka (1989), 20 M.V.R. (2d) 298 (Ont. Dist. Ct.) (breath test is not impracticable where the suspects injuries were not life-threatening and the suspect was waiting for x-rays); R. v. Pearce (1988), 13 M.V.R. (2d) 116 (Man. Q.B.) (impracticability connotes a degree of reason and involves some regard for practice); and R. v. Wytiuk (1989), 17 M.V.R. (2d) 18 (Man. Q.B.) (provisions are intended to give police flexibility in determining whether breath test is impracticable).  Supra note  NOTEREF _Ref28663687 \h 2, s. 256.  Unfortunately, these provisions are also technically complex and contain numerous pitfalls for police officers or justices of the peace who are unfamiliar with them. See, for example, R. v. Petkau, [1998] A.J. No. 1645 (Q.B.), online: QL (AJ); and R. v. Sattelberger (1995), 106 Man. R. (2d) 252 (Q.B.).  Criminal Code, supra note  NOTEREF _Ref28663687 \h 2, s. 257(1).  See, for example, R. v. Clark, [1995] S.J. No. 566 (Q.B.), online: QL (SJ), where Gerein J. found that the odour of alcohol on the suspects breath could be equally attributable to a medical condition, such as diabetes.  House of Commons Debates (20 December 1984) at 1384 and 1388 (Mr. Crosbie), and at 1396 (Mr. Waddell).  Indeed, two American studies found that less than one-third of the drivers who had BACs of 0.10% or higher when hospitalized were charged with impaired driving and only half of those charged were eventually convicted. Moreover, impaired drivers involved in the most serious crashes were the most likely to escape criminal liability. R.K. Cydulka et al., Injured Intoxicated Drivers: Citation, Conviction, Referral and Recidivism Rates (1998) 32 Ann. Emerg. Med. 349 (abstract); and J.W. Runge et al., Enforcement of Drunken Driving Laws in Cases Involving Injured Intoxicated Drivers (1996) 27 Ann. Emerg. Med 66 (abstract). The charge rates in these situations are likely to be similar or even lower in Canada, since police in most American jurisdictions have far broader authority to demand blood samples than their Canadian counterparts.  Road Safety Monitor 2002, supra note  NOTEREF _Ref39410505 \h 11 at 11.  Ibid. Emphasis in original.  Ibid. at 11 and 12. Based on the number of times respondents admitted to driving when they thought they were over the legal limit, the authors estimated that over 8 million impaired driving trips are made each year in Canada.  Based on the estimated 8 million annual impaired driving trips, an average of roughly 21,900 such trips are made each day. Ibid. Even accounting for impaired drivers who make more than one trip, it seems reasonable to assume that there are thousands of impaired drivers on the roads each night.  DWI Repeat Offenders, supra note  NOTEREF _Ref27040181 \h 14 at 4. Unfortunately, this estimate is based on data from the late 1980s. Given the sharp decreases in the estimated number of self-reported impaired driving trips per year, the current charge rate is likely considerably higher than 1 per every 445 impaired trips. See Road Safety Monitor 2002, supra note  NOTEREF _Ref39410505 \h 11 at 11.  R. F. Borkenstein, A Panoramic View of Alcohol, Drugs and Traffic Safety (1972) 16 Police 6.  B. Jonah et al., Front-line Police Officers Practices, Perceptions and Attitudes about the Enforce-ment of Impaired Driving Laws in Canada (1999) 31 Accid. Anal. and Prev. 421 at 426 [Police Perceptions].  Ibid.  Ibid.  Ibid. at 435.  Police Services Division, Safe Roads, Safe Communities (Victoria: Ministry of the Attorney General, Public Safety and Regulatory Branch, 2000) at B-4.  For a brief discussion of these problems see E. Chamberlain and R. Solomon, Taking Back Our Roads: A Comprehensive Review of Federal Impaired Driving Legislation (Toronto: MADD Canada, 2001) at 12 and 13.  Only 23% of those charged with impaired driving causing death pleaded or were found guilty in the provincial courts between 1994 and 1998. The conviction rate rose to 33% for impaired driving causing bodily harm, 56% for failing to provide breath or blood samples, and 61% for driving with a BAC above 0.08%. Canadian Centre for Justice Statistics, Adult Criminal Court Survey (Ottawa: Canadian Centre for Justice Statistics, 1999) [ACCS]. While statistics indicate that only 31% of those charged with impaired driving pleaded or were found guilty in the provincial courts, this low rate of guilty dispositions is misleading. Typically, the police will, if possible, charge impaired drivers with both impaired driving and driving with a BAC above 0.08%. The Canadian courts have held that, once a driver is convicted of one of these offences, he or she cannot be convicted of the other, if both arose out of the same incident. R. v. Kienapple (1975), 15 C.C.C. (2d) 524 (S.C.C.). Consequently, once a driver pleads or is found guilty on the first charge laid (driving with a BAC above 0.08%), the other charge (impaired driving) is invariably stayed or withdrawn. Indeed, 61% of the charges for impaired driving were stayed or withdrawn in the provincial courts between 1994 and 1998.  Ibid.  The 1999 and 2000 amendments: doubled the fines for first impaired driving offenders; increased the minimum driving prohibitions for first, second and subsequent offenders to one, two and three years, respectively; increased the maximum penalty for driving while prohibited to five years imprisonment, when tried by indictment; increased the maximum penalty for leaving the scene of a crash to ten years imprisonment, if the crash caused injury, and to life imprisonment, if the crash caused death; and increased the maximum penalty for impaired driving causing death to life imprisonment. Of the other amendments, three related to sentencing, and two related to enforcement. See An Act to amend the Criminal Code (impaired driving and related matters), S.C. 1999, c. 32; and An Act to amend the Criminal Code (impaired driving causing death and other matters), S.C. 2000, c. 25.  [1985] 2 S.C.R. 2.  As indicated, these programs have different names depending on the jurisdiction. We will use the term sobriety checkpoint for the remainder of this study.  Dedman, supra note  NOTEREF _Ref24198695 \h  \* MERGEFORMAT 113 at 35-36.  Ibid.  In 2000, Parliament enacted the Criminal Code offence of flight, which applies whenever a driver, while being pursued by a peace officer operating a motor vehicle, fails, without reasonable excuse and in order to evade the peace officer, to stop the vehicle as soon as is reasonable in the circumstances. The offence is punishable by a maximum of five years imprisonment if no injuries result. If, during the chase, the drivers dangerous driving causes bodily harm or death, then the maximum sentence is 14 years imprisonment or life, respectively. Criminal Code, supra note  NOTEREF _Ref28663687 \h  \* MERGEFORMAT 2, section 249.1, as am. by An Act to Amend the Criminal Code (flight), S.C. 2000, c. 2. However, this offence is limited to situations where a chase is already in progress. It does not encompass situations where the officer has simply signalled the driver to stop, the offender has fled on foot, or the offender was not fleeing for the purpose of evading the officer (e.g., if he was rushing himself or an injured party to the hospital). Nor would it apply to a driver who fails to stop for an officer who is standing at roadside, directing vehicles into a sobriety checkpoint. A broader provincial offence is needed to promote respect for checkpoint programs, improve their general deterrent effect, and protect law enforcement officers.  More than one-quarter of such pursuits result in damage, injuries or fatalities. Fact Sheet: Failing to Stop for Police (Ottawa: Canadian Police Association, 2000), online: Canadian Police Association homepage (date accessed: 4 November 2002). Statistics from Ontario indicate that, between 1991 and 1997, 33 people were killed and 2,415 were injured as a result of police chases. See Proceedings of the Standing Committee on Legal and Constitutional Affairs (Ottawa, 1 March 2000) (Senator Lorna Milne, chair).  See, for example, Prince Edward Island, Highway Traffic Act, R.S.P.E.I. 1988, c. H-5, s. 10(1)(d); and Newfoundland, Highway Traffic Act, R.S.N. 1990, c. H-3, s. 9(1)(d).  See infra text at 38-39.  An early study indicated that three-quarters of all impaired drivers were apprehended during what was described as routine patrol activities and random roadside checks. A.C. Donelson, D.J. Beirness & D.R. Mayhew, Characteristics of Drinking Drivers. Impaired Driving Report No. 1 (Ottawa: Department of Justice, 1985) at 27-29.  Sudbury Study, supra note  NOTEREF _Ref24199174 \h 36.  R. v. Therens, [1985] 1 S.C.R. 613 at 645.  In R. v. Ladouceur, [1990] 1 S.C.R. 1257, the Supreme Court of Canada upheld the constitutionality of the Ontario legislation authorizing random stops. Although the stop constituted an arbitrary detention, contrary to section 9 of the Charter, it was upheld under section 1 because it was prescribed by law and was a reasonable limit that was demonstrably justified in a free and democratic society. The Court specifically noted that random checks were the only means by which the police could detect and prosecute unlicensed, prohibited or uninsured drivers (at 1281).  J.E.L. Malenfant, R. Van Houten & B. Jonah, A Study to Measure the Incidence of Driving Under Suspension in the Greater Moncton Area (2002) 34 Accid. Anal. and Prev. 439 at 441 [Greater Moncton Area].  Ibid.  The roadside component of the study did not address the percentage of suspended drivers who continued to drive. Meanwhile, in the questionnaire component of the study, only roughly a third of the drivers admitted to driving while suspended when directly asked. However, when the question was rephrased, nearly half of the drivers stated that they drove more carefully while suspended. At least theoretically, the 57% exposure rate could be the result of either only 57% of the suspended drivers driving exactly the same amount as prior to the suspension or 100% of the suspended drivers driving 57% as much as they had previously driven. As we shall discuss, it is more likely that a very high percentage of suspended drivers continued to drive, with most driving less than they had prior to the suspension.  See, J.R. Snortum, Deterrence of Alcohol-impaired Driving: An Effect in Search of a Cause in M.D. Laurence, J.R. Snortum & F.E. Zimring, eds., Social Control of the Drinking Driver (Chicago: The University of Chicago Press, 1988) at 198; H.L. Ross & P. Gonzales, Effects of License Revocation on Drunk-Driving Offenders (1988) 20 Accid. Anal. and Prev. 379 at 383; and T.R. Neuman et al., National Cooperative Highway Research Program Report 500, Volume 2: A Guide for Addressing Collisions Involving Unlicensed Drivers and Drivers with Suspended or Revoked Licences (Washington: Transportation Research Board, 2003) at III-1.  D.J. DeYoung, R.C. Peck & C.J. Helander, Estimating the Exposure and Fatal Crash Rates of Suspended/Revoked and Unlicensed Drivers in California (1997) 29 Accid. Anal. and Prev. 17 [Estimating Exposure].  Since their analysis relies on fatal accidents data, it reflects the prevalence of such drivers during the times and at places where fatal accidents occur. Consequently, their estimates do not necessarily reflect the percentage of such drivers on California roads throughout the entire day. Ibid. at 22.  Ibid.  R.A. Scopatz et al., Unlicensed to Kill: The Sequel (Washington: AAA Foundation for Traffic Safety, 2003) at 7.  If one assumes that only 65% of Canadas suspended impaired drivers continue to drive and applies this percentage to the annual number of federal impaired driving convictions (which carry a minimum one-year driving prohibition), there would be at least 30,000 recent offenders driving while prohibited each year. ACCS, supra note  NOTEREF _Ref524766210 \h  \* MERGEFORMAT 110. This figure does not include offenders who were in the second, third or subsequent year of a lengthier federal driving prohibition or provincial licence suspension for a federal impaired driving offence. Nor does it include unlicensed drivers or offenders who never attempted to get their licences reinstated after their suspensions or prohibitions ended. Finally, it does not include the tens of thousands of drivers whose licences were suspended or revoked for: driving while prohibited, disqualified, suspended or uninsured; leaving the scene of an accident or wilfully attempting to evade police pursuit; or dangerous or careless driving. The average annual number of Criminal Code charges for driving while prohibited or suspended for a federal impaired driving offence between 1994-1998 was only 8,207. ACCS, supra note  NOTEREF _Ref524766210 \h  \* MERGEFORMAT 110. As the Malenfant study indicates, this represents a small fraction of the prohibited and suspended offenders who continue to drive. Greater Moncton Area, supra note 125.  A recent survey of American impaired driving offenders found that over half (54.5%) thought it was not at all likely that they would be caught driving while suspended. Another 21% thought that being caught was unlikely. MADD USA, Sanction Issues Compendium (Irving, Texas: MADD USA, 2002) at 15.1, online: (last modified: December 2002). Given this low perceived risk of apprehension, it is not surprising that so many offenders continue to drive during their suspension periods.  In such programs, a driver can only be asked for a breath sample if the police have the requisite grounds to believe that he or she has been drinking or is impaired. This is the type of checkpoint program that is currently used in Canada and the United States, where constitutional issues have, so far, inhibited the introduction of random breath testing (RBT) at sobriety checkpoints. In RBT programs, every driver is asked to provide a breath sample, and the police need not suspect that the driver has been drinking or is impaired. RBT programs are in effect in Australia, New Zealand and most western European jurisdictions.  R.W. Elder et al., Effectiveness of Sobriety Checkpoints for Preventing Alcohol-involved Crashes in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM [Elder].  For full details, see J.H. Lacey, R.K. Jones & R.G. Smith, Evaluation of Checkpoint Tennessee: Tennessees Statewide Sobriety Checkpoint Program (Washington: National Highway Traffic Safety Administration, 1999), online: (date accessed: 21 May 2002).  Ibid.  D.J. Beirness, R.D. Foss & B. Mercer, Roadside Breath Testing Surveys to Assess the Impact of an Enhanced DWI Enforcement Campaign in British Columbia in C. Mercier-Guyon, ed., Alcohol, Drugs and Traffic Safety, T97 (Annecy, France: Centre dtudes et de recherche en mdecine du trafic, 1997) at 955.  E. Vingilis, Problems in Detecting DWIs in Alcohol, Drugs and Driving, vol. 7 (Los Angeles: UCLA Brain Information Service/Brain Research Institute, 1991).  For guidelines on effective checkpoint programs, see Office of Enforcement and Emergency Services, The Use of Sobriety Checkpoints for Impaired Driving Enforcement (Washington: National Highway Traffic Safety Administration, 1990).  T.R. Miller, M.S. Galbraith & B.A. Lawrence, Costs and Benefits of a Community Sobriety Checkpoint Program (1998) 59 J. Stud. Alcohol 462.  Some jurisdictions also have mobile RBT, which allows police to demand breath samples on an ASD from any driver who is stopped during routine patrol activities. We do not presently advocate this sort of RBT in Canada, because it is more likely to be found unconstitutional as an unreasonable seizure, and because of concerns over its arbitrary use. For similar concerns regarding American civil liberties and RBT, see R. Homel, Random Breath Testing the Australian Way: A Model for the United States? (1990) 14 Alcohol Health and Research World 70.  The Scandinavian countries and Finland introduced RBT in the mid-1970s. For a brief review of RBT in selected countries, see On DWI Laws in Other Countries, supra note  NOTEREF _Ref39412874 \h 9, Table 6. While the studys coverage is limited, it indicates that most western European countries provide for some form of RBT.  J. Henstridge, R. Homel & P. Mackay, The Long-Term Effects of Random Breath Testing in Four Australian States: A Time Series Analysis (Canberra: Federal Office of Road Safety, 1997), Table 7.1 [Henstridge, Homel & Mackay]. Since all four states introduced RBT during the 1980s, the authors had a lengthy follow-up period in which to observe the long-term effects of RBT. In addition, the authors statistically controlled for variables such as season, time of day, day of week, weather, road usage, and the effects of 0.05% BAC laws (at 27). RBT was found to reduce serious, fatal and single-vehicle nighttime collisions in each state in which it was introduced. For example, in New South Wales, RBT was estimated to have prevented 522 serious, 204 fatal, and 686 single-vehicle nighttime crashes in its first year (at 104). Moreover, the positive effects of RBT were found to have extended throughout the evaluation period on most of the indicators that were examined. See also R. Homel, P. Mackay & J. Henstridge, The Impact on Accidents of Random Breath Testing in New South Wales: 1982-1992 in C. Kloeden & A. McLean, eds., Alcohol, Drugs and Traffic Safety T-95, vol. 2 (Adelaide: NHMRC Road Accident Research Unit, 1995) at 849-55; L.R. Sutton, The Effectiveness of Random Breath Testing: A Comparison Between the State of Tasmania, Australia and Four States in the Eastern United States in P.C. Noordzij & R. Roszbach, eds., Alcohol, Drugs and Traffic Safety T86 (Amsterdam, 1987).  Henstridge, Homel & Mackay, ibid., Table S.2.  R. Homel, Random Breath Testing and Random Stopping Programs in Australia in R.J. Wilson & R.E. Mann, eds., Drinking and Driving: Advances in Research and Prevention (New York: Guilford Press, 1990) at 175 [Homel].  Solicitor General for Alberta, Impaired Driving Program (Briefing Paper) (Edmonton: Department of the Solicitor General, 1989). In New South Wales, public support for RBT increased from 64% in 1982 to 97% in 1987.  See Henstridge, Homel & Mackay, supra note  NOTEREF _Ref30754198 \h 144; and R. Homel, Random breath testing in Australia: getting it to work according to specifications (1993) 88 (Suppl.) Addiction 27S at 28S-29S.  L. berg, Routine Breath Testing and Drivers Perceived Probability of Breath Test in P.C. Noordzij & R. Roszbach, eds., Alcohol, Drugs and Traffic Safety T86 (Amsterdam, 1987) 447 at 448-49.  H.M. Simpson, D.J. Beirness and D.R. Mayhew, National Opinion Poll on Drinking and Driving (Ottawa: Traffic Injury Research Foundation, 1999) at 15.  As indicated at supra note  NOTEREF _Ref31271506 \h 141, the savings attributable to checkpoint programs involving selective breath testing more than outweigh their costs. Although adding RBT to a checkpoint program would increase the costs, it would also increase the benefits.  See Elder, supra note  NOTEREF _Ref31272227 \h 135.  Criminal Code, supra note  NOTEREF _Ref28663687 \h  \* MERGEFORMAT 2, s. 254(2).  Moreover, almost 80% of officers believed that drivers sometimes or frequently escaped conviction because of legal technicalities. Presumably, this would include a significant number of cases in which judges ruled that evidentiary breath and blood-test results were inadmissible because the officer lacked the required grounds for the demand. Police Perceptions, supra note  NOTEREF _Ref39414949 \h 104 at 435.  Some judges have held that the odour of alcohol on a drivers breath is sufficient to create a reasonable suspicion to demand that the driver provide a breath sample for analysis on an ASD. R. v. Brown, [1986] N.S.J. No. 521 (Co. Ct.), online: QL (NSJ); and R. v. Mitruk, [1998] O.J. No. 2444 (Gen. Div.), online: QL (OJ). However, other judges have held that the odour of alcohol is not sufficient. R. v. Lindsay, [1998] O.J. No. 2587 (Gen. Div.), online: QL (OJ); and R. v. Stephens (1967), 2 C.R.N.S. 103 (N.S. C.A.). Finally, some courts have held that the odour of alcohol, coupled with one other symptom of impairment, can create a reasonable suspicion that a driver has alcohol in his or her body. R. v. Clarke, [1996] N.S.J. No. 123 (S.C.), online: QL (NSJ); R. v. Wolthauser, [1996] O.J. No. 3567 (Prov. Div.), online: QL (OJ); and R. v. Doyle, [1998] O.J. No. 1091 (Gen. Div.), online: QL (OJ).  These studies involved roadside surveys conducted downstream from police spot checks. Presumably, some of these undetected drivers had developed a tolerance for alcohol and failed to exhibit visible signs and symptoms of impairment, or had otherwise learned to avoid raising police suspicions. See S.A. Ferguson, J.K. Wells & A.K. Lund, The role of passive alcohol sensors in detecting alcohol-impaired drivers at sobriety checkpoints (1995) 11:1 Alcohol, Drugs and Driving 23 [Ferguson]; and I.S. Jones & A.K. Lund, Detection of alcohol-impaired drivers using a passive alcohol sensor (1986) 14 J. Police Sci. Administration 153 [Jones]. This earlier study found that 52% of legally intoxicated drivers interviewed by police officers at sobriety checkpoints were not apprehended. See also J.K. Worden et al., Preventing Alcohol-Impaired Driving Through Community Self-Regulation Training (1989) 79 A.J.P.H. 287; and E. Vingilis, E.M. Adlaf & L. Chung, Comparison of Age and Sex Characteristics of Police-Suspected Impaired Drivers and Roadside Surveyed Impaired Drivers (1982) 14 Accid. Anal. and Prev. 425 [Vingilis]. The authors of the latter study estimated that 95% of legally impaired drivers do not risk apprehension at police spotchecks. See also E. Vingilis et al., Psychosocial Characteristics of Alcohol-Involved Seriously Injured Drivers (1994) 26 Accid. Anal. and Prev. 195 at 203. Escaping detection not only allows intoxicated drivers to continue driving, but also reinforces their belief that they are not really impaired. This, in turn, encourages subsequent impaired driving. DWI Repeat Offenders, supra note  NOTEREF _Ref27040181 \h  \* MERGEFORMAT 14 at 62.  J.K. Wells et al., Drinking Drivers Missed at Sobriety Checkpoints (1997) 58 J. Stud. Alcohol 513 at 516.  Ethyl alcohol, the active ingredient in alcoholic beverages, has no scent of its own. Rather, it is the other components of the beverages that give them their distinctive smells. DWI Repeat Offenders, supra note  NOTEREF _Ref27040181 \h 14 at 63.  See H.A. Moskowitz, M.M. Burns & S.A. Ferguson, Police officers detection of breath odors from alcohol ingestion (1999) 31 Accid. Anal. and Prev. 175, where the authors found that officers estimates of BAC were not better than would have been expected from random guesses.  For example, one study reported that women are routinely missed more often than men, young drivers are missed more often than older drivers, and drivers without passengers are more likely to be missed than drivers with passengers. Vingilis, supra note  NOTEREF _Ref519656913 \h 156.  One study indicated that only 36% of drivers with BACs above the legal limit believed it was very likely that they would be arrested if stopped at a checkpoint when driving drunk. R.D. Foss & M.W. Perrine, Perceived Risk of Arrest for DUI (Paper presented at the 118th Annual Meeting of the American Public Health Association, New York, October 1990).  Although developed and initially tested in the early 1980s, passive alcohol sensors were not routinely used until 1991 when they were donated by MADD U.S.A. to police agencies in Massachusetts, Florida, Rhode Island, California, Georgia, Louisiana, Minnesota, Tennessee, and Texas.  These sensors are classified as passive devices because they do not require the person to perform any act, such as blowing into a mouthpiece. J.P. Manak, Constitutional Aspects of the Use of Passive Alcohol Screening Devices as a Law Enforcement Tool for DWI Enforcement (Winter 1986) 19 The Prosecutor 29 [Manak].  A.J. Farina, Laboratory Evaluation of Two Passive Sensor Devices (Washington: National Highway Traffic Safety Administration, 1998). See also R.B. Voas, Laboratory and Field Tests of a Passive Alcohol Sensing System (1983) 4:3 Abstracts and Reviews in Alcohol & Driving 3; D.C. Lestina & A.K. Lund, Laboratory Evaluation of Two Passive Alcohol Sensors (Paper presented to the 70th Annual Meeting of the Transportation Research Board, Washington, D.C., 15 January 1991); and Ferguson, supra note  NOTEREF _Ref519656913 \h 156.  R.D. Foss, R.B. Voas & D.J. Beirness, Using a Passive Alcohol Sensor to Detect Legally Intoxicated Drivers (1993) 83 A.J.P.H. 556 at 558 [Foss].  Jones, supra note  NOTEREF _Ref519656913 \h 156.  Foss, supra note  NOTEREF _Ref40165698 \h 165 at 557.  C.M. Farmer et al., Field Evaluation of the PAS III Passive Alcohol Sensor (1998) 1 Journal of Crash Prevention and Injury Control 55.  This challenge would likely be based on the fact that passive alcohol sensors would be used routinely, without prior authorization or specific evidence that a driver had been drinking.  R. v. Collins, [1987] 1 S.C.R. 265 at 278.  See supra note  NOTEREF _Ref519666271 \h  \* MERGEFORMAT 123.  See S.L. Grey, Passive Alcohol Sensors and the Fourth Amendment in The Impaired Driving Update (Spring 2001) (Kingston, NJ: Civil Research Institute, Inc., 2002). The issue has not apparently been addressed by American appellate courts. See also M. Fields & A.R. Hricko, Passive Alcohol Sensors - Constitutional Implications (Summer 1986) 20 The Prosecutor 45; and Manak, supra note  NOTEREF _Ref24382586 \h  \* MERGEFORMAT 163, where the respective authors concluded that the courts would likely find the use of passive alcohol sensors to be constitutional.  Ibid. This is based on an analogy with an individuals voice and face, which were held not to be subject to a reasonable expectation of privacy because they are constantly exposed to the general public. U.S. v. Dionisio, 410 U.S. 1 (1973). This should be contrasted with breath samples provided into an ASD or evidentiary instrument, which require a deep lung sample not normally exposed to public observation.  This is a variation on the more well-known plain view doctrine, which states that police do not offend the Fourth Amendment when they seize an object which is in their plain view, provided they come across it inadvertently. Similarly, the American courts have upheld a plain smell doctrine, finding that a defendant is not protected from search by an officers inquisitive nostrils. See U.S. v. Johnson, 497 F.2d 397 (9th Cir. 1974). In Canada, a version of the plain smell doctrine was adopted by Major, Gonthier and LHeureux-Dub JJ. in R. v. Evans, [1996] 1 S.C.R. 8 at para. 50. They found that the police conduct of smelling marijuana while standing at the accuseds front door did not constitute a search within the meaning of section 8 of the Charter. The majority of the Court reached the opposite conclusion, but emphasized the special privacy interests associated with police intrusions in residences. The majority might well reach a different decision on the section 8 issue in impaired driving cases, given the lessened expectation of privacy while driving and the public interest in attempting to keep impaired drivers off of the roads. See R. v. OConnor (2002), 62 O.R. (3d) 263 at 287-288 (C.A.), where the Court specifically stated that the expectation of privacy in a vehicle is lower than that in a residence.  See, for example, U.S. v. Knotts, 460 U.S. 276 (1983); U.S. v. Dubrovsky, 581 F.2d 208 (9th Cir. 1978); U.S. v. Minton, 488 F.2d 37 (4th Cir. 1973); and U.S. v. Bronstein, 521 F.2d 459 (2d Cir. 1975).  See M. Burns & H. Moskowitz, Psychophysical Tests for DWI Arrest (Washington: National Highway Traffic Safety Administration, 1977); and J. Stuster & M. Burns, Validation of the Standardized Field Sobriety Test Battery at BACs below 0.10 Percent (Washington: National Highway Traffic Safety Administration, 1998).  The examiner looks for three indicators of impairment in each eye: (i) the eye cannot follow a moving object smoothly; (ii) jerking is distinct when the eye is at maximum deviation (i.e., farthest from centre); and (iii) once the eye begins jerking, its angle of onset is within 45 degrees of centre (i.e., halfway to the edge of the eye). Ibid.  The suspect will accumulate points if he or she exhibits any of the following eight indicia of impairment: (i) losing balance during the instruction phase; (ii) beginning before the instructions are finished; (iii) stopping while walking to regain balance; (iv) failing to touch heel-to-toe; (v) stepping off the line; (vi) using arms to balance; (vii) making an improper turn; and (viii) taking an incorrect number of steps. Ibid.  The four indicia of impairment are: (i) swaying while balancing; (ii) using arms to balance; (iii) hopping to maintain balance; and (iv) putting the foot down. Ibid.  Ibid. See also M. Burns & E.W. Anderson, A Colorado Validation Study of the Standardized Field Sobriety Test (SFST) Battery. Final Report to the Colorado Department of Transportation (Denver: Colorado Department of Transportation, 1995).  Currently, immediate access to an ASD is often critical to the legality of the demand for a sample. In R. v. Grant, [1991] 3 S.C.R. 139, the Supreme Court of Canada held that a 30-minute delay in the arrival of an ASD provided the driver with a reasonable excuse for refusing the test. Consequently, the suspect had to be acquitted of failing to provide a breath sample. However, if the police had been authorized to demand SFST, the results may have quickly provided them with grounds for demanding an evidentiary breath test.  In R. v. Pierman; R. v. Dewald (1994), 92 C.C.C. (3d) 160 (Ont. C.A.), Arbour J.A. (as she then was) suggested that a delay in ASD testing would only be justified if the police had reason to believe that an immediate test would provide a breath sample that was unsuitable for analysis, due to recent alcohol consumption. (In order to avoid false high readings due to alcohol in the mouth, ASD tests should be taken at least 15-20 minutes after the individuals last drink.) Arbour J.A.s reasoning was affirmed by the Supreme Court of Canada in R. v. Dewald, [1996] 1 S.C.R. 68.  Ironically, delay in administering the breath test will almost invariably benefit a suspected impaired driver. Toxicological evidence shows that a persons BAC usually peaks approximately 20 minutes after his or her last drink, and then decreases by an average of 0.015% per hour. As a result, breath tests performed two to three hours after the alleged offence will likely provide a reading that is considerably lower than the drivers BAC at the time of driving. In some cases, drivers who were legally impaired while driving may obtain readings below 0.08%, thereby escaping criminal charges. T.A. Loomis, Blood Alcohol in Automobile Drivers: Measurement and Interpretation for Medicolegal Purposes. I. Effect of Time Interval Between Incident and Sample Acquisition (1974) 35 Quart. J. Stud. Alc. 458 at 468-71.  See, for example, R. v. Smith (1996), 28 O.R. (3d) 75 (C.A.); and R. v. Bonin (1989), 47 C.C.C. (3d) 230 (B.C. C.A.). The Courts stated that the police have the power to take steps to determine if there are reasonable and probable grounds to demand a breath sample, and that this implicitly includes the power to request a field sobriety test, even though the driver does not have to comply. While the test constitutes a detention, entitling the accused to the right to counsel under section 10(b) of the Charter, the Courts ruled that the officers infringement of the accuseds section 10(b) rights was justifiable under section 1 of the Charter. Given the tests short duration and minimal inconvenience, the infringement was held to be a reasonable limit that was justifiable in a free and democratic society. Thus, according to these cases, the police do not have to inform a driver of his or her right to counsel before requesting or conducting a field sobriety test.  See, for example, R. v. Gallant (1989), 48 C.C.C. (3d) 329 (Alta. C.A.); and R. v. Baroni (1989), 49 C.C.C. (3d) 553 (N.S. C.A.). Consistent with Smith and Bonin, these cases held that the police have the power to request a field sobriety test. However, unlike Smith and Bonin, Gallant and Baroni held that a denial of a drivers right to counsel under section 10(b) of the Charter was not justified under section 1. Therefore, according to Gallant and Baroni, the police must inform drivers of their right to counsel and give them an opportunity to contact counsel before conducting field sobriety tests.  If the police think that they might wish to use the results and videotape of the test in evidence to establish the accuseds impairment, they should comply with section 10(b) of the Charter. The accused should be informed of his or her right to counsel and given an opportunity to contact counsel before the test is undertaken. In R. v. Milne (1996), 28 O.R. (3d) 577 (C.A.), the Ontario Court of Appeal ruled that field sobriety tests constituted a detention, thereby triggering the accuseds right to counsel under section 10(b) of the Charter. The Court also indicated that an infringement of this right was justified under section 1 of the Charter only if the purpose of the tests was to form the necessary reasonable and probable grounds to demand an evidentiary breath test. However, the infringement would not be justified under section 1 if the police wished to use the results of the field sobriety test to establish the accuseds impairment at trial. Moldaver J.A. stated that field sobriety tests were not meant to provide the police with a means of gathering evidence that could later be used to incriminate and convict the motorist of impaired driving at trial (at 587). Therefore, the results of field sobriety tests performed in violation of the accuseds right to counsel will be ruled inadmissible if introduced to establish the accuseds impairment. This decision was followed by the New Brunswick Court of Appeal in R. v. Oldham (1996), 49 C.R. (4th) 251. The Court held that a suspect should not be asked to give incriminating evidence without first being informed that he or she does not have to comply with the police officers request. In the absence of a statutory obligation to comply with an SFST demand, this reasoning will likely be followed in other jurisdictions as well.  A recent Australian study found that SFST was reliable in detecting impairment by marijuana, both at moderate and high levels. The One Leg Stand test was a particularly good indicator of drug impairment, and the authors also supported the use of a new component head movements/jerks to assess impairment. See K. Tzambazis and C. Stough, The SFST and Driving Ability. Are they related? in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM.  As indicated, 29% of Canadian police officers who participated in a recent survey felt that judges give greater credibility to defence witnesses than to the police at impaired driving trials, and three-quarters felt that impaired drivers sometimes or frequently escape conviction due to legal technicalities. Police Perceptions, supra note  NOTEREF _Ref39414949 \h 104 at 426, 432 and 435.  Ibid.  See B. Jones, In-vehicle Videotaping of Drinking Driver Traffic Stops in Oregon (1999) 31 Accid. Anal. and Prev. 77 at 83. In addition, a recent survey polled police, Crowns, defence counsel, and judges on a pilot project that installed video cameras in police cruisers and breath-testing rooms. Sixty-one per cent recommended the adoption of video equipment in cruisers, and 92% recommended its use in breath-testing rooms. The respondents felt that the video cameras had a calming effect on the arrestee, reducing the incidence of disruptive behaviour. Finally, 86% of those surveyed felt that video evidence increased the proportion of guilty pleas. B. Pavic et al., Fast-track Courts and Video Cameras as Drinking-Driving Countermeasures in C. Mercier-Guyon, ed., Alcohol, Drugs and Traffic Safety, T97 (Annecy, France: Centre dtudes et de recherche en mdecine du trafic, 1997) 73 at 76-77.  These tests are known by different names in different jurisdictions. To avoid confusion, we will use the more generic term, drug evaluation and classification test.  As described with respect to SFST, Qubec is currently the only province that requires drivers to comply with a demand for field sobriety tests. While the courts in other provinces have found that police have authority to request such a test, there are no sanctions for drivers who refuse to comply.  Criminal Code, supra note  NOTEREF _Ref28663687 \h  \* MERGEFORMAT 2, s. 253(a).  As explained earlier, police have very limited authority to demand a blood sample from an impaired driving suspect. The provisions governing judicial warrants authorizing the taking of blood samples from unconscious impaired driving suspects are even more narrowly defined. See supra text at 28-31. Nevertheless, if a blood sample has been lawfully obtained under these provisions or with the accuseds consent, section 258(5) of the Criminal Code enables a qualified technician to test it for drugs.  Since some drugs and their metabolites remain in the body long after their psychoactive and impairing effects have worn off, the presence of a drug in the drivers blood or urine does not necessarily mean that the driver was impaired at the time of testing. Depending on the drug, a positive test may simply indicate that the driver had used that drug sometime in the recent past. S.B. Karch, The Pathology of Drug Abuse (Boca Raton: CRC Press, 1993) at 18. It is for this reason that the suspect must be subject to a detailed examination for evidence of impairment.  H. Richter, Marijuanas effect on traffic safety in Canada: an educators guide (Developed by Drug Awareness Service, RCMP, 1999) [unpublished] at 77.  O.H. Drummer, Drugs and Accident Risk in Fatally-Injured Drivers, online: Schaffer Library of Drug Policy, (date accessed: 9 August 2001). See also D. Brookoff et al., Testing Reckless Drivers for Cocaine and Marijuana (1994) 331 N.E.J.M. 518.  G.W. Mercer & W.K. Jeffery, Alcohol, Drugs, and Impairment in Fatal Accidents in British Columbia (1995) 27 Accid. Anal. and Prev. 335 [Mercer & Jeffery]. See also Estimating the Presence of Alcohol and Drug Impairment, supra note  NOTEREF _Ref39279574 \h 13 at 15-24.  C. Dussault et al., The Contribution of Alcohol and Other Drugs Among Fatally Injured Drivers in Quebec: Some Preliminary Results in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM [Dussault et al.].  Ibid.; and Mercer & Jeffery, supra note  NOTEREF _Ref31618674 \h 198 at 335.  Adlaf, supra note  NOTEREF _Ref31621750 \h 10. See also G.W. Walsh & R.E. Mann, On the High Road: Driving Under the Influence of Cannabis in Ontario (1999) 90 C.J.P.H. 260.  Dussault et al., supra note  NOTEREF _Ref39473284 \h 199.  Youth and Road Crashes, supra note  NOTEREF _Ref39319539 \h 26 at 2.  This brief summary of the procedure is based on T.E. Page, The Drug Recognition Expert (DRE) Response to the Drug Impaired Driver: An Overview of the DRE Program, Officer and Procedures, online: Los Angeles Police Department, (date accessed: 29 August 2001).  Because the test will undoubtedly be classified as a detention, the officer should first inform the driver of the right to counsel and allow him or her a reasonable opportunity to consult counsel. This is particularly important because the test is used to gather evidence of drug impairment, which will be introduced in court. After the driver has had a reasonable opportunity to consult counsel, the officer can proceed with the test.  G.E. Bigelow, Identifying Types of Drug Intoxication: Laboratory Evaluation of a Subject Examination Procedure (Washington: National Highway Traffic Safety Administration, 1984).  R.P. Compton, Field Evaluation of the Los Angeles Police Department Drug Detection Procedure (Washington: National Highway Traffic Safety Administration, 1986).  Section 11(c) of the Charter provides that an accused cannot be compelled to testify at his or her own trial. Pursuant to section 13, self-incriminating testimony that a witness has given cannot be used against him or her in a subsequent criminal proceeding.  For example, in R. v. Curr, [1972] S.C.R. 889, the Supreme Court of Canada held that the privilege against self-incrimination, then contained in section 2(d) of the Canadian Bill of Rights, was limited to statements, and not to incriminating conditions of the body. Similarly, sections 11(c) and 13 of the Charter have been found to be restricted to testimonial compulsion and not to encompass compulsory breath tests. See R. v. Altseimer (1982), 1 C.C.C. (3d) 7 (Ont. C.A.); and R. v. Gaff (1984), 15 C.C.C. (3d) 126 (Sask. C.A.), leave to appeal to S.C.C. refused (22 November 1984). An accused might also argue that the legislation violates the principles of fundamental justice in section 7 of the Charter. That section states that [e]veryone has the right to life, liberty and security of the person and the right not to be deprived thereof except in accordance with the principles of fundamental justice. The Supreme Court of Canada has held that fundamental justice includes the right not to be forced to provide incriminating evidence, and thereby supplements the right to silence contained in sections 11(c) and 13. See R. v. White, [1999] 2 S.C.R. 417. Nevertheless, given that the courts have summarily dismissed arguments to the effect that compulsory breath tests violate section 7, similar arguments regarding drug testing are likely to receive short shrift.  See supra text at 28-32.  Testing of crash-involved drivers is already required in many jurisdictions. In addition, several jurisdictions, including Australia, Belgium, Finland, France, the Netherlands, New Zealand, and Sweden, authorize random breath testing of drivers. Accordingly, the police can demand that any driver, including one involved in a crash, provide breath samples for analysis. Thus, the testing provisions in these jurisdictions are far broader than those proposed here. K. Stewart, Preventing Impaired Driving Around the World: Lessons Learned (2000) 6 Injury Prevention 80 at 81; and On DWI Laws in Other Countries, supra note  NOTEREF _Ref39412874 \h 9, Table 6.  Criminal Code, supra note  NOTEREF _Ref28663687 \h  \* MERGEFORMAT 2, s. 254(4) and (5).  For example, although a 1997 Health Canada survey reported that 1,680 people are killed as a result of impaired driving each year, there were only 133 charges for impaired driving causing death in 1996. DWI Repeat Offenders, supra note  NOTEREF _Ref27040181 \h 14; and S. Tremblay, Impaired Driving in Canada, 1996 (1997) 17:2 Juristat at 10. Moreover, only a relatively small percentage of these charges lead to a conviction for this offence. During the 1994-1998 reporting periods, only 23% of the charges for impaired driving causing death decided in the provincial courts resulted in convictions. Although only 3% resulted in acquittals, the charges were stayed or withdrawn in 43% of the cases. Presumably, a significant number of these cases involved the accused pleading guilty to a lesser offence in exchange for the dropping of the original charge. The remaining 28% of the charges were transferred to the Superior Courts. Since no statistics were available regarding the outcome of these cases, the number of convictions is unknown. ACCS, supra note  NOTEREF _Ref524766210 \h 110. See also B. Carroll & R. Solomon, Understanding Drinking and Driving Reforms: A Profile of Ontario Statistics (2000) 6 Injury Prevention 96 at 97.  While it has been estimated that approximately 74,000 Canadians are injured in alcohol-related crashes each year, the average annual number of charges for impaired driving causing bodily harm between 1994 and 1998 was only 1,365. Of these, only 33% pleaded or were found guilty in the provincial courts. ACCS, supra note  NOTEREF _Ref524766210 \h  \* MERGEFORMAT 110.  The accused was also acquitted of dangerous driving causing death and bodily harm, because the judge was not convinced beyond a reasonable doubt that she was driving in a dangerous manner. See R. Exner, Dangerous driving acquittal sends wrong message The [Edmonton] Journal (7 November 1998).  In another case, the accused driver crossed the yellow centre line, collided head-on with Robert Swinsons car and killed him. The accused was charged with impaired driving causing death. Even though the accused had the odour of alcohol on her breath, admitted to drinking, caused a fatal collision and there was no other explanation for the crash, the judge ruled that the police did not have sufficient grounds to arrest her and demand an evidentiary breath test. Thus, the breath test results, which established that the accuseds BAC was 0.20%, or two-and-a-half times the Criminal Code limit, were excluded and she was acquitted. MADD Canada, News Release, MADD Canada National Board Outraged at Attorney Generals Refusal to Appeal Unjust Court Ruling (18 September 1995).  Road Safety Vision 2001, supra note  NOTEREF _Ref39279574 \h 13 at 11; and Magnitude, supra note  NOTEREF _Ref39279574 \h  \* MERGEFORMAT 13.  The exception would be in cases where the crash was obviously the result of some other factor, such as inclement weather, wandering animals or road obstacles.  Provincial legislation authorizing the police to demand an evidentiary breath test from such drivers would more likely be struck down as an unreasonable search under section 8 of the Charter. In R. v. Bernshaw, [1995] 1 S.C.R. 254, the Supreme Court of Canada held that, in order for evidence obtained through a warrantless search, such as an evidentiary breath test, to be admissible in criminal proceedings, it must be based on reasonable and probable grounds, not mere suspicion. While the occurrence of a crash may create a suspicion that one of the drivers had been drinking, it would rarely, by itself, create reasonable grounds to believe that a specific driver had committed an impaired driving offence. It is for this reason that MADD Canada recommends that the provincial provision be limited at first instance to an ASD. Similarly, MADD Canada proposes that police be empowered to demand an evidentiary breath or blood test in crashes only if it is not possible to conduct an initial ASD.  As indicated, section 254(2) of the Criminal Code authorizes the police to demand a breath sample for analysis on an ASD from any driver they reasonably suspect has any alcohol in his or her body. They need not believe that the driver is drunk, impaired or committing an offence. While this provision is helpful in routine impaired driving cases, it is problematic in crash situations for several reasons. For example, in some cases, it may be difficult for the police to immediately determine who was driving. Section 254(2) only authorizes police to demand an ASD test from known drivers, not suspected drivers. Moreover, the police may have difficulty determining whether a driver has alcohol in his or her body, and thereby forming the requisite grounds to demand an ASD test. If the driver is injured, unconscious or otherwise unable to communicate, the police may not be able to rely on the traditional indicia of alcohol consumption, such as slurred speech, unresponsiveness, glazed eyes, or the odour of alcohol on the drivers breath.  The problems with the current breath-testing provisions are exacerbated when a suspect is taken to hospital following a crash. In such cases, the police do not generally have the benefit of having administered an ASD test. Nevertheless, they must still prove that they had reasonable and probable grounds to demand an evidentiary breath test. As indicated, this is a high threshold to satisfy. Moreover, the statutory time constraints for testing are more difficult to meet if a suspect is hospitalized. The police must secure the scene, arrange transportation for the injured, get to the hospital, find the suspect, form the requisite grounds for demanding an evidentiary breath test, advise the suspect of his or her legal rights, and provide him or her with an opportunity to exercise those rights. As will be discussed, if the breath sample is taken more than two hours after the commission of the alleged offence, the Crown will be unable to rely on the presumption of identity. See infra note  NOTEREF _Ref39474413 \h 225. Finally, even if police establish the necessary grounds to demand a breath test within the appropriate time period, administering such a test in a hospital is inherently difficult. See supra text at 28-32.  Several Canadian cases indicate that the police cannot administer an ASD test away from the roadside, unless the suspect is being taken to a location for immediate ASD testing. See R. v. Cote (1992), 70 C.C.C. (3d) 280 (Ont. C.A.); and R. v. Pierman; R. v. Dewald, supra note  NOTEREF _Ref39474587 \h 182. This is particularly problematic in a crash situation, where the suspect has or feigns illness or injury, and is taken to hospital. Since ASDs cannot be used in these cases to screen drivers for impairment, the police must attempt to establish the necessary grounds to demand evidentiary tests by other means, which is extremely difficult. Without preliminary ASD results, the officers grounds for demanding evidentiary breath or blood tests are more likely to be challenged by defence counsel, which increases the possibility that a suspect will be acquitted of the most serious impaired driving charges.  Section 254(2) requires that a breath sample be provided forthwith, but the meaning of this word has not been precisely defined by the Canadian courts. Some courts have taken a strict approach and require that an ASD test must be administered immediately after the demand, while other courts have taken a more flexible approach. For example, in Grant, supra note  NOTEREF _Ref24709545 \h 181, the Supreme Court of Canada held that the term forthwith suggests that the breath sample is to be provided immediately. Thus, a demand by a police officer who had to wait 30 minutes for an ASD to arrive does not satisfy this condition. In R. v. Payne (1994), 91 C.C.C. (3d) 144, the Newfoundland Court of Appeal held that the determination of whether the ASD test was administered forthwith is not a question of the number of seconds that have elapsed from the time of the demand to the administration of the test. Rather, the question is one of circumstance. The Court held that the demand would be invalid in most situations if the time between the demand and the administration of the test is more than 30 minutes. If the time period between the demand and the administration of the test is less than 30 minutes, then the court must examine the circumstances. In Bernshaw, supra note  NOTEREF _Ref39474647 \h 219, the Supreme Court of Canada revisited the issue of the meaning of forthwith, and held that it does not require the ASD test to be administered immediately after a demand. The Court further stated that an ASD test should be administered as soon as possible, but it did not specify what this meant. This uncertainty is of particular concern since not all police vehicles are equipped with ASDs. Consequently, even if police have the reasonable suspicion necessary for an ASD demand, they do not always have the means to administer the test. If and when an ASD finally arrives at the scene, the question then turns to whether a breath sample was provided forthwith, which is determined on a case-by-case basis. This situation provides defence counsel with ample grounds to challenge the validity of ASD results, which may result in the exclusion of subsequent evidentiary tests.  The police can only demand evidentiary breath tests under section 254(3) if they reasonably believe that the driver has committed an impaired driving offence within the previous three hours. This not only imposes difficult time constraints on police, but also impedes investigations, particularly in the most serious crashes. It may take 30 minutes or more for police to arrive on the scene of a crash. Upon arrival, they are faced with several, often conflicting, tasks. Their first priority is, and should be, to ensure the victims well-being. They are also concerned with securing the scene, gathering evidence, and investigating the causes of the crash. While this transpires, potential suspects may be rushed to hospital for treatment before police have an opportunity to fully investigate the matter. Thus, in the most serious crashes, the police may not have sufficient time to develop the requisite grounds to demand an evidentiary breath test.  Even if the police successfully establish that they have reasonable grounds to believe that a suspect committed an impaired driving offence within the previous three hours, the presumption of identity in section 258(1)(c) does not necessarily apply. This section presumes that an accuseds BAC at the time of the breath test reflects his or her BAC at the time of driving. However, when Parliament extended the time limit to three hours for making demands for breath and blood samples in 1999, it failed to make a corresponding amendment to the presumption of identity. Therefore, the results of the breath or blood analyses will only be presumed to reflect the offenders BAC at the time of the alleged offence if the samples were taken within two hours. If the test is administered after two hours, the Crown will be required to call an expert toxicologist to relate the accuseds BAC at the time of testing back to the time of driving. This is time-consuming, expensive and provides the defence with additional opportunities to challenge the results.  In Therens, supra note  NOTEREF _Ref519666271 \h 123, the Supreme Court of Canada held that a motorist who is required to provide an evidentiary breath sample is detained within the meaning of section 10, and thus, has the right to consult counsel. While the suspect consults counsel, even more time elapses between the commission of the alleged impaired driving offence and the time of testing. As indicated, this additional delay may be critical if the presumption of identity is to apply. Moreover, by exercising the right to counsel, an accused learns of his or her legal options. An accused who has caused a crash may prefer to be convicted of the lesser offence of refusing to provide a breath sample than to provide an evidentiary breath sample and potentially be convicted of a more serious offence, such as impaired driving causing death or bodily harm.  For example, the laws in Australia, New Zealand and the United Kingdom attempt to ensure that impaired drivers who kill or injure do not escape criminal liability. Although the legislation in these three nations differs, there are some common elements. First, in the event of a crash, the police are authorized to demand a preliminary breath test from any driver or suspected driver. They do not have to believe or suspect that the person has any alcohol in his or her blood. Second, all three jurisdictions have been mindful of the realities of crash investigations and the difficulties that police face in obtaining evidentiary breath and blood samples. This is exemplified by the authority granted to police to demand an evidentiary breath or blood sample from any driver or suspected driver involved in a crash. Moreover, New Zealand, the United Kingdom, and most Australian states have recognized the impracticalities of administering evidentiary breath tests in a hospital. Although they have taken a different approach in addressing this concern, they share one fundamental element the express authority to demand a blood sample from a suspect admitted to hospital for treatment. See Road Traffic Act 1988 (U.K.), 1988, c. 52, ss. 6 and 7; New Zealand, Transport Act 1962, s. 58A, 58B, 58D, as am. by Transport Amendment Act (No. 3) 1992, s. 16(1)(a) and (b); Road Transport (Alcohol and Drugs) Act 1977 (ACT), ss. 8, and 15AA; Traffic Act 1949 (Qnld), s. 16A(2); Road Safety (Alcohol and Drugs) Act 1970 (Tas), ss. 7A(1), and 8(3); Road Safety Act 1986 (Vic), s. 53(1); Road Traffic Act 1961 (SA), s. 47E, and 47I; Road Traffic Act 1974 (WA), s. 66; Road Transport (Safety and Traffic Management) Act 1999 (NSW), ss. 13(1)(a) and (b), and 20(2); and Traffic Act 1987 (NT), s. 23.  The occurrence of the crash would be the basis for the demand. Given the high rates of alcohol involvement in fatal and personal injury crashes, it can hardly be said that the demand is random or arbitrary. Road Safety Vision 2001, supra note  NOTEREF _Ref39279574 \h 13 at 11; and Magnitude, supra note  NOTEREF _Ref39279574 \h 13. While our laws generally require individualized suspicion for searches, it must be emphasized that driving is a highly-regulated activity that occurs on public roads. It is also important to stress that, currently, the police only need a reasonable suspicion that a driver has alcohol in his or her body to demand an ASD test. This is a much lower standard than reasonable and probable grounds. Furthermore, ASD results are not admissible in court as evidence of the suspects BAC at the time of the offence. Rather, they are merely an investigative tool that serves the crucial purpose of providing an officer with the requisite grounds to demand an evidentiary breath or blood sample.  See supra note  NOTEREF _Ref39474944 \h 97.  Runge et al., supra note  NOTEREF _Ref39474944 \h 97, found that, while 33% of those with non-severe injuries were charged with impaired driving, only 7% of those with severe injuries were charged. See also K.R. Krause et al., Prosecution and Conviction of the Injured Intoxicated Driver (1998) 45 J. Trauma 1069; and R.E. Lucke, Medical Reporting of BACs to the Police (paper presented to the 27th International Forum on Traffic Records and Highway Information Systems, New Orleans, 1 August 2001). The situation in Canada is likely worse, since Canada lacks the implied consent laws that exist in American states. Such laws provide that, on lawful arrest for impaired driving, the suspect consents to having a breath, blood or urine sample taken to determine his or her BAC. Thus, Canadian police may be even less likely than their American counterparts to obtain BAC evidence from injured drivers.  For example, in South Australia, when a person over 14 attends or is admitted to a hospital with injuries suffered in a motor vehicle crash, the medical practitioner must take samples of his or her blood and seal them in two containers. See Road Safety Act 1961, s. 47I. The practitioner must then make the samples available to the patient and the police. Once analyzed by a qualified professional, the blood sample can be used in evidence in a criminal trial for impaired driving. Similar provisions exist in Victoria, Tasmania, and the Australian Capital Territory.  See Egan, Ross, and Horsefield, supra note  NOTEREF _Ref27041912 \h 4, and also R. v. Boggs, [1981] 1 S.C.R. 49.  Unfortunately, some of the provinces also use the term prohibition in reference to their 90-day administrative licence suspensions. See, for example, Motor Vehicle Act, R.S.B.C. 1996, c. 318, s. 94.1(1)(a); and Highway Traffic Act, R.S.P.E.I. 1988, c. H-5, s. 277.2(1). Consequently, one cannot assume that the term driving prohibition is always meant to refer to a federal driving prohibition imposed under the Criminal Code. In our discussion, however, we will reserve the use of the term prohibition for the federal provisions.  Depending on the jurisdiction, the grounds for imposing suspensions may include: specified provincial traffic offences; too many demerit points; unfitness to drive; a poor driving record; and unpaid fees, fines and civil judgments.  Both laboratory and field studies establish that various driving-related skills are significantly impaired at BAC levels of 0.05% or lower. Drivers with low and moderate BACs have impaired psychomotor skills, such as steering and braking, as well as impaired visual function and reaction time. However, the skills most affected by small amounts of alcohol are information processing and divided attention skills, which are crucial if drivers are to respond quickly and accurately to various traffic hazards. Drivers with low and moderate BACs have difficulty recognizing traffic hazards and deciding the best way to respond. See, for example, H. Moskowitz & D. Fiorentino, A Review of the Literature on the Effects of Low Doses of Alcohol on Driving-Releated Skills (Washington: National Highway Traffic Safety Administration, 2000); Moskowitz et al., Driver Characteristics and Impairment at Various BACs (Washington: National Highway Traffic Safety Administration, 2000); and P. Howat, D. Sleet & I. Smith, Alcohol and Driving: Is the 0.05% Blood Alcohol Concentration Limit Justified? (1991) 10 Drug and Alcohol Review 151 [Howat]. See also the following recent papers in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM: M. Burns & D. Fiorentino, The Effects of Low BACs on Driving Performance; K. Stewart, Low Blood Alcohol Content: Overview of Performance, Safety, and Policy Implications; and H. Moskowitz & D. Fiorentino, A Review of the Experimental Studies of Low BAC Effects on Skills Performance. Due to this impairment of skills, drivers with BACs in the 0.05% to 0.079% range are at considerably increased relative risk of crash. See Zador, supra note  NOTEREF _Ref31621217 \h  \* MERGEFORMAT 69; and Relative Risk, supra note  NOTEREF _Ref27039357 \h  \* MERGEFORMAT 67. See also the following recent papers in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM: D.F. Preusser, BAC and Fatal Crash Risk; H. Moskowitz et al. Methodological Issues in Epidemiological Studies of Alcohol Crash Risk; and R.P. Compton et al., Crash Risk of Alcohol Impaired Driving. For a summary of the evidence on skills impairment and relative risk at low and moderate BAC levels, see E. Chamberlain & R. Solomon, The Case for a 0.05% Criminal Law Blood Alcohol Concentration Limit for Driving (2002) 8 (Suppl. III) Injury Prevention iii1 [The Case for 0.05%].  A recent New Brunswick study indicated that 91% of suspended drivers stopped at a roadside checkpoint were able to produce an apparently valid licence. Police only learned of a drivers suspension if they checked the licence with a Ministry of Transportation database. Greater Moncton Area, supra note  NOTEREF _Ref31621366 \h  \* MERGEFORMAT 125 at 441- 442. This undermines the deterrent impact of the impaired driving laws and makes it difficult to enforce driving prohibitions and suspensions.  See supra note  NOTEREF _Ref26510034 \h 235.  See, for example, R.E. Mann et al., The effects of introducing or lowering legal per se blood alcohol limits for driving: an international review (2001) 33 Accid. Anal. and Prev. 569; Howat, supra note  NOTEREF _Ref26510034 \h  \* MERGEFORMAT 235; P.C. Noordzij, Decline in drinking and driving in the Netherlands in The Nature of and the reasons for the worldwide decline in drinking and driving. Transportation research circular number 422 (Washington: National Academy Press, 1994) at 44-49; T. Norstrm & H. Laurell, Effects of lowering the legal BAC-limit in Sweden in C. Mercier-Guyon, ed., Proceedings of the 14th International Conference on Alcohol, Drugs and Traffic Safety T97 (Annecy, France: Centre dtudes et de recherche en mdecine du trafic, 1997) at 87-94; and G. Bartl & R. Esberger, Effects of lowering the legal BAC-limit in Austria in H. Laurell & F. Schlyter, eds., Proceedings of the 15th International Conference on Alcohol, Drugs and Traffic Safety T-2000, online: (date accessed: 4 January 2003). For a more thorough review of the evidence on the impact of a lower BAC limit, see The Case for 0.05%, supra note  NOTEREF _Ref26510034 \h  \* MERGEFORMAT 235.  See Henstridge, Homel & Mackay, supra note  NOTEREF _Ref30754198 \h 144, Table 7.4.  If a driver cannot produce his or her licence for the officer, the driver should be given two days to bring the licence to the police station. Drivers failing to surrender their licences within this period should have their licences revoked.  Because the drivers BAC will decline as time passes, these confirmatory tests on approved instruments should be conducted as soon as possible. Otherwise, the driver may be able to delay the evidentiary test long enough for his or her BAC to fall below the 0.05% threshold. In Canada, it is generally accepted that an average persons BAC declines by 0.015% per hour. H.R. Fisher, R.I. Simpson & B.M. Kapur, Calculation of Blood Alcohol Concentration (BAC) by Sex, Weight, Number of Drinks and Time (1987) Can. J. Public Health 300 at 301. Thus, an individual whose BAC was just below 0.08% would have a BAC below 0.05% after two hours. Based on the less conservative metabolism rate (0.017% per hour) used by NHTSA, the drivers BAC would fall below 0.05% even faster. In order to address this concern, it is important that the driver have the burden of proving that his or her BAC was below 0.05% at the time of driving.  These jurisdictions include: Argentina, Australia, Austria, Belgium, Bulgaria, Croatia, Denmark, Finland, France, Germany, Greece, Iceland, Israel, Italy, Macedonia, the Netherlands, Norway, Portugal, Slovenia, Spain, and Sweden.  See On DWI Laws in Other Countries, supra note  NOTEREF _Ref39412874 \h 9, Tables 1 and 2.  The provinces and territories may wish to create a 0.05% provincial offence to supplement the 0.05% ALS law. The offence could provide meaningful sanctions for unlicensed drivers, with BACs of 0.05% or higher, who are largely unaffected by the licence suspensions.  K. Stewart, P. Gruenewald & T. Roth, An Evaluation of Administrative Per Se Laws: Final Report (Washington: National Institute of Justice, 1989). See also A.K. Lund, Effectiveness of Administrative License Revocation (ALR) Laws (Arlington: Insurance Institute for Highway Safety, 1992).  The authors suggested that this reduction was a result of the general deterrent effect of the ALS provisions on all drivers. J.L. Nichols & H.L. Ross, The effectiveness of legal sanctions in dealing with drinking drivers (1990) 6:2 Alcohol, Drugs and Driving 33 [Legal Sanctions].  A preliminary evaluation of Ontarios ALS law found that it helped to reduce both alcohol-related fatalities and self-reported driving after drinking. See G. Stoduto et al., The Impact of the Admini-strative Drivers Licence Suspension Law in Ontario in H. Laurell & F. Schlyter, eds., Proceedings of the 15th International Conference on Alcohol, Drugs and Traffic Safety T2000 (Stockholm: International Council on Alcohol, Drugs and Traffic Safety, 2000), online: (date accessed: 29 January 2002).  Finally, impaired drivers who received an ALS had a 44% lower recidivism rate over the next four years than drivers without an ALS. D.J. Beirness, H.M. Simpson & D.R. Mayhew, Evaluation of Administrative Licence Suspension and Vehicle Impoundment Programs in Manitoba (Ottawa: Transport Canada, 1997) at 34, 46, 55, and 59 [Manitoba Evaluation].  General deterrence is accepted to be the product of three factors: the certainty, celerity and severity of the penalty. ALS satisfies the first two, in that it takes effect immediately and without exception. These are important features which distinguish ALS from the provincial licence suspensions that only apply after a federal criminal conviction. Because these latter suspensions are tied to the outcome of criminal proceedings, they may not take effect until months after the alleged offence. Moreover, due to the various evidentiary and other problems that may arise in such criminal prosecutions, the driver may never be convicted. Thus, although the provincial suspensions imposed for federal convictions may be more severe, they are neither as certain nor as swift as immediate ALS. See B.M. Sweedler, Strategies for Dealing with the Persistent Drinking Driver (Washington: National Transportation Safety Board, 1995).  This includes breath and blood tests, SFST, and drug evaluation and classification tests.  It should be noted that the ALS laws have withstood various constitutional challenges. See Leclair v. R. (1990), 25 M.V.R. (2d) 252 (Man. Q.B.); White v. Nova Scotia (Registrar of Motor Vehicles) (1996), 20 M.V.R. (3d) 192 (S.C.); Horsefield, supra note  NOTEREF _Ref27041912 \h  \* MERGEFORMAT 4; Buhlers, supra note  NOTEREF _Ref31550317 \h  \* MERGEFORMAT 19; Gonzalez v. Alberta (Driver Control Board) (2001), 96 Alta. L.R. (3d) 324 (Q.B.).  See supra note  NOTEREF _Ref30048273 \h  \* MERGEFORMAT 236.  It is important that the ALS last for a substantial amount of time for two reasons. First, this helps to keep suspected offenders off the roads until the date of trial. Second, studies have found that ALS laws with relatively weak penalties have little or no deterrent effect. F.J. Chaloupka, H. Saffer & M. Grossman, Alcohol-Control Policies and Motor Vehicle Fatalities (1992) 22 J. Legal Stud. 161 at 162.  For example, after Manitoba introduced its ALS program, the average number of days between the offence and the conviction was reduced from 114 to 55. Manitoba Evaluation, supra note 249 at 55.  In 1999, the Criminal Code was amended to increase the length of the federal driving prohibitions for the offences of impaired driving, driving with a BAC above 0.08%, and failing to provide breath or blood samples. The minimum driving prohibition is generally one year for a first offence, two years for a second offence and three years for any subsequent offence. However, these mandatory driving prohibitions will be reduced to three, six and twelve months respectively, if the driver agrees to participate in a provincial alcohol interlock program, if one is available. Criminal Code, supra note  NOTEREF _Ref28663687 \h 2, s. 259(1.2).  See supra note  NOTEREF _Ref30048273 \h  \* MERGEFORMAT 236.  Some studies suggest that the effectiveness of licence suspensions in reducing recidivism for impaired driving declines with longer periods of suspension. In addition, drivers with lengthy suspensions are more likely to drive while suspended. DWI Repeat Offenders, supra note  NOTEREF _Ref27040181 \h  \* MERGEFORMAT 14 at 70. Nevertheless, licence suspensions continue to be associated with decreases in impaired driving recidivism and crash rates for both first and repeat offenders. R. Mann et al., Sentence Severity and the Drinking Driver: Relationships with Traffic Safety Outcome (1991) 23 Accid. Anal. and Prev. 483 at 486 [Sentence Severity]. The problem of driving while suspended can, perhaps, be best addressed by combining licence suspensions with increased licence inspections and vehicle-based sanctions, such as seizure, impoundment and forfeiture.  The work permits typically allow the offender to drive for employment purposes only.  B. Watson, V. Siskind & M. King, The Effectiveness of Licence Restriction For Drink Driving Offenders: The Australian Experience in H. Laurell & F. Schlyter, eds., Proceedings of the 15th International Conference on Alcohol, Drugs and Traffic Safety T2000 (Stockholm: International Council on Alcohol, Drugs and Traffic Safety, 2000), online: (date accessed: 29 January 2002) at 4.  Ibid.  Ibid., at 2.  Some types of interlocks are attached to the ignition to prevent the vehicle from being started, while other types are attached to the transmission to prevent the vehicle from being shifted into drive.  For example, many interlock devices require rolling retests repeat breath tests at random intervals after the vehicle has been started. These retests are designed to discourage the impaired driver from getting a sober friend or passenger to start the car, to detect drivers whose BACs are still rising, and to prevent a driver from leaving the vehicle idling while he or she drinks. Failing to provide a sample within the allotted time may result in a warning or an alarm, and the driver may be required to report immediately to a service centre or program manager. Failure to do so will cause the interlock to prevent further use of the vehicle. Other anti-circumvention features include: temperature and pressure sensors that prevent the use of filtered breath samples; hum-tone recognition; and breath pulse codes. These features are intended to prevent a bystander from providing a sample for the impaired driving offender. They are used to identify the offender and require some practice to master. Thus, it is highly unlikely that a bystander could successfully provide a sample on a first attempt. Moreover, repeated failed tests result in a lockout of the system, so the driver will be unable to start the vehicle or shift it into drive. Finally, the wires and circuits on the interlock are sealed so that it is easy to detect any attempts to tamper with the device. D.J. Beirness, Alcohol Ignition Interlocks (Ottawa: Traffic Injury Research Foundation, 1996) at 7-8 [Alcohol Ignition Interlocks].  Various American studies show that the recidivism rate of impaired driving offenders in an interlock program is 28% to 65% lower than that of offenders who are not in a program. The studies are summarized in D. Beirness et al., Evaluation of the Alberta Interlock Program: Preliminary Results (Ottawa: Traffic Injury Research Foundation, 1997) at 1. In Alberta, roughly 10% of convicted impaired drivers have participated in the interlock program since it was introduced in 1994. Only 7.2% of participants had an impaired driving conviction in the 36 months following the program, compared to the 22.2% re-conviction rate among non-participants (at 3-4).  See D.W. Collier, F.J.E. Comeau & I.R. Marples, Experience in Alberta with Highly Sophisticated Anti-Circumvention Features in a Fuel Cell Based Ignition Interlock in C.N. Kloeden & A.J. McLean, eds., Proceedings of the 13th International Conference on Alcohol, Drugs and Traffic Safety, vol. 2 (Adelaide, South Australia: University of Adelaide, NHMRC Road Accident Research Unit, 1995) at 673. Similarly, a recent position paper from leading interlock researchers reported that interlocks reduce impaired driving recidivism by 40-95% relative to simple licence suspension. ICADTS Working Group on Alcohol Interlocks, Alcohol Ignition Interlock Devices. I: Position Paper (Calverton, MD: International Council on Alcohol, Drugs and Traffic Safety, 2001) at 10 [Interlock Position Paper].  Many researchers have suggested that it is best to include as many offenders as possible in interlock programs. See, for example, D.J. Beirness & R.D. Robertson, Best Practices for Alcohol Interlock Programs: Findings from Two Workshops in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM [Best Practices].  As indicated, the mandatory federal driving prohibitions of one, two and three years for first, second and subsequent convictions are reduced to three, six and twelve months, respectively, if the offender participates in a provincial interlock program. Criminal Code, supra note  NOTEREF _Ref28663687 \h 2, s. 259(1.2).  See Best Practices, supra note  NOTEREF _Ref25315493 \h  \* MERGEFORMAT 266. See also R.B. Voas et al., Evaluation of a program to motivate impaired driving offenders to install ignition interlocks (2002) 34 Accid. Anal. and Prev. 449.  See Actions to Reduce Fatalities, Injuries and Crashes Involving the Hard Core Drinking Driver (Washington: National Transportation Safety Board, 2000) at 17 [Actions to Reduce Fatalities]; and supra notes  NOTEREF _Ref31621366 \h 125- NOTEREF _Ref39761201 \h 134.  Moreover, research demonstrates that the longer the suspension, the less likely it is that the offender will apply to have his or her licence reinstated. In many cases, the offender simply chooses to drive without a licence. Thus, shortening the provincial revocation period would likely encourage more offenders to seek licence reinstatement. See Best Practices, supra note  NOTEREF _Ref25315493 \h  \* MERGEFORMAT 266.  See, for example, R. Homel, Policing and Punishing the Drinking Driver: A Study of General and Specific Deterrence (New York: Springer-Verlag, 1988) at 219; Sentence Severity, supra note  NOTEREF _Ref31622532 \h  \* MERGEFORMAT 257; and R.E. Hagen, The Efficacy of Licensing Controls as a Countermeasure for Multiple DUI Offenders (1978) 10 J. Safety Res. 115.  It has been argued that the costs of installing and maintaining an interlock are prohibitive for some offenders. This has been countered by the argument that the costs of the interlock would, in many cases, be off-set by the offenders reduced alcohol consumption. See Interlock Position Paper, supra note  NOTEREF _Ref39761269 \h 265 at 18-19. That said, the provinces may wish to subsidize offenders who truly cannot afford an interlock. In addition, insurance companies should consider reducing the premiums or surcharges they normally charge an impaired driving offender if he or she participates in an interlock program.  Alcohol Ignition Interlocks, supra note  NOTEREF _Ref27097303 \h 263 at 33. Similar results were observed in an evaluation of Qubecs interlock program, which was introduced in 1997. L. Vzina, The Qubec Ignition Interlock Program: Impact on Recidivism and Crashes in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM.  Whatever the reason for this increased recidivism, it does not warrant foregoing the introduction of an interlock program. Even the short-term benefits of the current interlock programs provide a sufficient justification for their enactment. Moreover, we believe that, with appropriate modifications, the programs can provide far greater long-term safety benefits.  In this way, the interlock will be seen as an essential component of a comprehensive remedial program. See Best Practices, supra note  NOTEREF _Ref25315493 \h 266; and Interlock Position Paper, supra note  NOTEREF _Ref39761269 \h 265 at 17. The authors of the latter report noted that the focus of rehabilitative programs, such as interlocks, should be on positive changes in the offenders behaviour, not simply the completion of a pre-determined period of time in the program.  Similarly, recidivism is higher among offenders whose rate of attempted starts declines over the period the interlock is on the vehicle. In these cases, offenders may have chosen to drive less frequently or to drive a vehicle that was not equipped with an interlock. See P.R. Marques, A.S. Tippetts & R.B. Voas, During Alcohol Interlock Use, Elevated BACs in Morning or Declining Vehicle Use Over Time Predicts Post-Interlock Recidivism in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM [Post-Interlock Recidivism].  In Sweden, participants in the interlock program are required to attend medical appointments every three months to monitor their alcohol consumption and lifestyle. See B. Bjerre, A Preliminary Evaluation of the Swedish Ignition Interlock Programme and Recommended Further Steps in D. Mayhew & C. Dussault, eds., Proceedings of the 16th International Conference on Alcohol, Drugs and Traffic Safety (Montreal: Socit de lassurance automobile du Qubec, 2002), CD-ROM. This type of monitoring may be a positive step toward ensuring that interlocks are part of a comprehensive remedial program, which addresses the offenders underlying alcohol problems. Indeed, a preliminary study reported that participants in the Swedish interlock program showed a steady decrease in alcohol consumption, and a marked decrease in potentially dangerous or harmful alcohol habits. It will be interesting to see whether these effects of the Swedish program extend beyond the interlock period.  See Post-Interlock Recidivism, supra note  NOTEREF _Ref25415210 \h  \* MERGEFORMAT 275.  As noted, the above study found that some offenders had a declining rate of attempted starts on the interlock vehicle. The authors suggested that these offenders may have started driving vehicles not equipped with an interlock device. In such cases, the benefits of the interlock program would be lost. Therefore, it is important that the legislation have the appropriate teeth to deter offenders from simply switching to a non-interlock vehicle. In order to make such enforcement possible, it is important that there be some indication on the offenders licence that he or she is only permitted to drive a vehicle equipped with an alcohol interlock. See Interlock Position Paper, supra note  NOTEREF _Ref39761269 \h 265 at 9.  As described, a recent study in the Greater Moncton Area attempted to determine the driving exposure of suspended drivers by checking drivers licences at systematic roadblocks. The study suggests that many offenders routinely drove during their suspension periods. Greater Moncton Area, supra note  NOTEREF _Ref31621366 \h  \* MERGEFORMAT 125. This is consistent with American research, which indicates that up to 75% of convicted impaired drivers continue to drive at least occasionally during their suspensions. See supra notes  NOTEREF _Ref27095291 \h 128- NOTEREF _Ref39761201 \h 131; R.E. Hagen, E.J. McConnell & R.L. Williams, Suspension and Revocation Effects on the DUI Offender (Sacramento: Business and Transportation Agency, Department of Motor Vehicles, 1980); Legal Sanctions, supra note  NOTEREF _Ref39761482 \h 246; and A.T McCartt, L.L. Geary & W.J. Nissen, Observational Study of the Extent of Driving While Suspended for Alcohol-Impaired Driving (Washington: National Highway Traffic Safety Administration, 2002).  Generally, impounded vehicles are secured in a locked storage facility, whereas some form of boot or club is attached to immobilized vehicles to prevent them from being driven. Immobilization programs tend to be less costly than impoundment programs, because the vehicle can be kept on the owners property and no storage costs are incurred.  R. B. Voas, A.S. Tippetts & E. Taylor, Temporary Vehicle Immobilization: Evaluation of a Program in Ohio (1997) 29 Accid. Anal. and Prev. 635 at 639. Because the study reviewed all offenders eligible for immobilization within a two-year period, the follow-up period varied considerably. For example, offenders whose vehicles were immobilized near the beginning of the study were tracked for nearly two years, while those immobilized near the end of the study had a much shorter follow-up period. A second study reported that Ohios impoundment program reduced driving among suspended drivers by 50%, even 23 months after the end of the impoundment period. DWI Repeat Offenders, supra note  NOTEREF _Ref27040181 \h  \* MERGEFORMAT 14 at 72.  This longer-term effect was confirmed in a follow-up study by the same authors. R.B. Voas, A.S. Tippetts & E. Taylor, Temporary Vehicle Impoundment in Ohio: A Replication and Confirmation (1998) 30 Accid. Anal. and Prev. 651 [Voas 1998]. However, they were unable to determine whether this was a result of specific deterrence or extended incapacitation by being denied access to a vehicle (at 654-55).  Manitoba Evaluation, supra note  NOTEREF _Ref31623195 \h 248 at 48, 57, and 59. The program also had a general deterrent effect among suspended drivers. During the six years following its implementation, there was a 35% decrease in the number of charges for driving while suspended. In addition, recidivism during the three months following conviction was reduced from 10.3% prior to the introduction of impoundment to 6.3% after its enactment.  In addition to reducing recidivism for impaired driving and driving while prohibited, Californias impoundment program resulted in an 18-22% reduction in traffic convictions and a 25-38% reduction in subsequent crashes for both first and repeat offenders whose vehicles were impounded. Offenders subject to impoundment also had 24-34% fewer convictions for driving while suspended in the year following the 30-day impoundment. D.J. DeYoung, An Evaluation of the Specific Deterrent Effects of Vehicle Impoundment on Suspended, Revoked, and Unlicensed Drivers in California (Washington: National Highway Traffic Safety Administration, 1997) at 40-43.  Under these laws, those caught driving while suspended have a zebra sticker affixed to their licence plates. This identifies the vehicle as one that is owned by a suspended driver. These programs have not been particularly effective. R.B. Voas, A.S. Tippetts & J.E. Lange, Evaluation of a Method for Reducing Unlicensed Driving: The Washington and Oregon License Plate Sticker Laws (1997) 29 Accid. Anal. and Prev. 627 at 629. The zebra sticker provides police with reason and authority to stop the vehicle and determine if the driver is licensed. The owner of the vehicle could have the zebra sticker removed only if he or she paid a small fee and presented a valid drivers licence. However, two-thirds of suspended owner-offenders merely transferred title to their vehicles to a family member in order to obtain clean plates, and there was relatively minimal delay in getting their vehicles back on the highway. Nevertheless, the study did show reductions in both moving violations and crash rates among suspended drivers. Oregon and Washington, the states that had introduced zebra sticker laws, have allowed those laws to lapse and they are no longer in force. M.M. Levy & J.F. Frank, A Review of Research on Vehicle Sanctions in the U.S.A. in H. Laurell & F. Schlyter, eds., Proceedings of the 15th International Conference on Alcohol, Drugs and Traffic Safety T2000 (Stockholm: International Council on Alcohol, Drugs and Traffic Safety, 2000), online: (date accessed: 7 December 2002).  Individuals who drive while suspended generally have higher crash rates than members of the general driving population. For example, suspended or revoked drivers in California have close to four times the fatal crash rate of licensed drivers. See Estimating Exposure, supra notes  NOTEREF _Ref31632202 \h  \* MERGEFORMAT 129 and  NOTEREF _Ref39761201 \h 131. Moreover, of all the repeat offenders involved in fatal crashes in the United States in 1991, nearly half (48%) were driving while suspended at the time of the fatal crash. Voas 1998, supra note  NOTEREF _Ref31623436 \h  \* MERGEFORMAT 282 at 651. See also Sanction Issues Compendium (Texas: MADD U.S.A., 2002) at 15.5, online: (date accessed: 4 December 2002). Similarly, statistics from New Brunswick indicate that suspended drivers are more than twice as likely to be involved in fatal crashes as licensed drivers. Greater Moncton Area, supra note  NOTEREF _Ref31621366 \h  \* MERGEFORMAT 125 at 3.  An evaluation of Manitobas vehicle impoundment program revealed that 20% of owners whose vehicles are impounded have previously had a vehicle impounded. DWI Repeat Offenders, supra note  NOTEREF _Ref27040181 \h  \* MERGEFORMAT 14 at 71.  In some cases, the proceeds from the sale of the vehicle will not be sufficient to recover the towing and storage costs. This is particularly true for lengthy impoundment periods. If costs are a potential problem, the province or territory may wish to use immobilization, rather than impoundment. See supra note 282.  An evaluation of Manitobas impoundment program indicated that 20% of the vehicles seized from repeat offenders belonged to someone other than the offender. Manitoba Evaluation, supra note  NOTEREF _Ref31623195 \h  \* MERGEFORMAT 248.  See Actions to Reduce Fatalities, supra note  NOTEREF _Ref39761691 \h 269 at 24. However, this early data should be treated with caution, as it has not been subject to scientific evaluation.  Greater Moncton Area, supra note  NOTEREF _Ref31621366 \h 125 at 441- 442. This raises two issues. First, licensing authorities need to be stricter in seizing drivers licences from individuals who are suspended. Second, the fact that an individual is able to produce a licence does not mean that he or she is legally entitled to drive. A better system needs to be in place to verify the validity of a drivers licence.  In some cases, the insurance company may be exempt from responsibility for certain damages caused by an unlicensed or suspended driver. The company may also have a right to recover from the vehicle owner any damages that it had to pay out because of such drivers. Thus, the owner may be held unexpectedly responsible for most or all the damages arising from the crash.  H.M. Simpson, D.R. Mayhew & D.J. Beirness, Dealing with the Hard Core Drinking Driver (Ottawa: Traffic Injury Research Foundation, 1996) at 57 [Dealing with the Hard Core].  DWI Repeat Offenders, supra note  NOTEREF _Ref27040181 \h  \* MERGEFORMAT 14 at 80.  Research indicates that individualized assessment is preferable to a system that assigns offenders to remedial programs based solely on the number of prior convictions or the offenders BAC at the time of arrest. This allows experts to consider other relevant factors, such as the offenders personality, lifestyle and risk-taking behaviour. Ibid. at 79. See also Dealing with the Hard Core, supra note  NOTEREF _Ref39566579 \h 293 at 47-48.  M. Eliany & B. Rush, How Effective are Alcohol and Other Drug Prevention and Treatment Programs? A Review of Evaluation Studies (Ottawa: Health and Welfare Canada, 1992).  Impaired drivers who participated in some form of remedial program had an 8% to 9% lower recidivism and alcohol-related crash rate than drivers who did not. E. Wells-Parker et al., Final Results From a Meta-Analysis of Remedial Interventions with Drink/Drive Offenders (1995) 90 Addiction 907. Another study found that impaired drivers who did not participate in remedial programs were 1.5 to 1.7 times more likely to re-offend than those who did. D.J. DeYoung, An Evaluation of the Effectiveness of Alcohol Treatment, Driver License Actions and Jail Terms in Reducing Drunk Driving Recidivism in California (1997) 92 Addiction 989 [DeYoung 1997]. One Canadian study found that remedial programs combining educational and therapeutic elements reduced impaired driving recidivism by as much as 24% for low problem offenders. G. Stoduto et al., Existing Programs for Convicted Drinking-Drivers in Canada (Toronto: Addiction Research Foundation, 1998) at 4. For a comprehensive review of the literature on remedial programs, see R. Mann et al., A Critical Review on the Effectiveness of Drinking-Driving Rehabilitation Programmes (1983) 15 Accid. Anal. and Prev. 441; and R.K. Jones & J.H. Lacey, Alcohol and Highway Safety 2001: A Review of the State of Knowledge (Washington: National Highway Traffic Safety Administration, 2001), Chapter 5, online: (date accessed: 4 December 2002).  In addition to its positive impact on individual offenders, increased levels of treatment have also been shown to have positive effects on public health in general. Increased treatment rates have been associated with declines in problems such as liver cirrhosis, pancreatitis, alcoholism, alcohol intoxication, and alcohol psychosis, as well as lower rates of alcohol-related and total collision fatalities. R.G. Smart & R.E. Mann, The Impact of Programs for High Risk Drinkers on Population Levels of Alcohol Problems (2000) 95 Addiction 37 at 42. See also R.G. Smart & R.E. Mann, Are Increased Levels of Treatment and Alcoholics Anonymous Membership Large Enough to Create the Recent Reductions in Liver Cirrhosis? (1990) 85 British J. Addiction 1385; and R.E. Mann, R.G. Smart & L. Anglin, Factors Influencing Aggregate Indicators of Drinking-Driving in the United States in C.N. Kloeden & A.J. McLean, eds., Alcohol, Drugs and Traffic Safety, T95, vol. 1 (Adelaide: NHMRC Road Accident Research Unit, 1995) at 71.  Educational programs are most suited to offenders with few or only early-stage alcohol problems. Dealing with the Hard Core, supra note  NOTEREF _Ref39566579 \h 293 at 60.  We prefer not to comment on the specific structure or content of the treatment program, which would be designed by health and addictions professionals. A summary of various types of assessment and treatment programs can be found in Dealing with the Hard Core, ibid. at 46-68. It may also be preferable to use a variety of existing treatment programs, and to direct offenders to the program that best suits their needs.  A study prepared for the Ontario government recommended that successful completion of a remedial program should entail full attendance, sobriety while in attendance, and some form of client-initiated follow-up six months after completing the program. Only after such a follow-up would the program provider endorse the offenders application for licence reinstatement. R. Mann et al., Remedial Programs for Convicted Drinking Drivers. Part 2: Implementing a Systematic Program in Ontario (Toronto: Addiction Research Foundation, 1997) at 21-22 [Remedial Programs].  This burden of proof is not intended to be so severe that it is impossible to meet. Nevertheless, there should be some affirmative duty on the individual to prove his or her fitness to drive, rather than an automatic licence reinstatement upon completion of a prescribed remedial program.  It is important that remedial programs (except alcohol interlocks) are assigned in addition, and not as an alternative, to licence suspension or revocation. When remedial programs are used to reduce or replace a licence suspension, recidivism rates are similar to those for offenders receiving licence suspensions. In many instances, drivers who attended remedial programs instead of receiving a licence suspension had higher recidivism rates than drivers who received suspensions. R.E. Mann, Effectiveness of DUI Treatment and the Importance of Screening and Matching Clients to Appropriate Treatment (Paper presented to Drinking and Driving Prevention Symposium, Los Angeles, 15-17 November 1992) at 117. However, combining licence suspensions and remedial programs has been found to be more effective than either component alone. In a California study of convicted impaired driving offenders, those assigned to treatment and some form of licence sanction had lower recidivism rates than those who received only a licence sanction, treatment or jail. The author suggests that suspensions plus treatment are best because suspensions reduce driving exposure and encourage more cautious driving, and treatment allows offenders to address their alcohol problems. DeYoung 1997, supra note  NOTEREF _Ref27095953 \h  \* MERGEFORMAT 297 at 994 and 997. See also Remedial Programs, supra note  NOTEREF _Ref25314429 \h  \* MERGEFORMAT 301. PAGE  1 -  PAGE 43 "#'(8:CUV~! + Y y |   4 = ] {     ? ^ ~  " # % w hxh+~h7V h0T5h0T *hAv *hAvfHq hAv5hAv5CJaJhGjh&(UmHnHuh&(h'hAvCJ8OJQJ^JaJ8hAvOJQJ^JhAv: !"#$%&')*+,-./0123$$$$$$I=$$$$$$$$$$$$$$$$$$$a$"2333456789:;<=>?@ABCUVeg~$$$$$$$$$$$$$$$$$t$$$$$$ h*&>^>a$ $ &0^0a$$ #a$/ z { |  5 ^  @  $$$$$$$$$$$$$$$$$ h*&>^>a$$ h*&>`>a$gds;$ h*&>`>a$$ h*&>^>a$$ h*&0`0a$ $ h*(a$$ *&>^>a$ & $ |  @a$$$$$$$$$$$$$$$$ h*&>^>`a$ $ h*&a$$ h*(>`>a$$ h*&>`>a$$ h*&>^>a$$ h*&0`0a$ $ h*(a$$ *&>^>a$  # |  ?@`n !Vjkopquww?@5ihhh4~jhAv0JU hAv6]hhAv6] hAvaJ( hAv6 *hAvhxhAv hAv5 hx5HaAg%#$$$$$$$$t$$$q$$$$E($$$$4Idh$a$$ h*&0^0a$ $ h*(a$$ >*&@>^>a$$ *&>^>a$$ h*&>`>a$$ h*&>^>a$$+,FGgk5 6 !!!/!!!!!&&&&(()(+++u--9/]///̸̸̨̨̨̨̨̨̟̖̊~ho$hmh\hAv6]hmH sH hlhmH sH hS!hmH sH jhAv0JUhmH sH h|-hmH sH hAv6hmH sH hAvhmH sH  hAv6hS!hS!hS!6h|- hAv6]hAvjhAv0JU0%#'+-a136k8888\<@CGOJM1QSAS5WoXpXX$$4$,$$E($ $$$RF$RF$,$,$,$,$$E($E($$RF$,$O $RF$RFU  U  p~ @ $ p~ @ a$$a$Idh/////0000011_1`1e1{1116 6 6666666777l88888*989<<<==== >>>>?@@BB$B%BzB{BF%FGGpHxHHH5I6IIIh+h`phQ# hAv6 *hAvCJaJhAvCJaJhAvCJ(aJ(jhAv0JU h%] hAv]h% hAv6]hh fhAvhmBIKJNJJJYL[LLLLLNN%O(OSSSASSSTT]U^UVV0W1WoXpXXXXXXXXXXAYBYYYZZb[c[]]]]__{_|_bbbbUdVdeeeee{f|fggӾh!hEh-L hAv6hT"CJaJhAvCJaJjhAv0JUhh/jh%NjhAv0JU\ hAv6]hAvFXZ\_c}fjnrtex|||i}~^_~π$$$q$$q$$E($!0$E($'8$$E($!0$$RF$$O $"$$ѱ2$s%$ /0v8 xHX "L^`La$$ ,08 xHX a$ $ dha$ Ug>hBhJhKhBiPiQiTiVinioijj.k/kkkknn(o)oooqprpqqrrss2w3w y yyyyy2z3z1{2{||||||h}i}HI\]^_`zjhAvUh]hN2hk6 *hAvCJaJhAvCJaJh( hE6h Vh H3 hAv6jhAv0JUhoQLh/>hAv>z{|}356PQRSTzρ efۇ܇IJˊ̊&'Ǎȍ̍ıĥěěěěěěěěėěěěěěěěčhAvCJOJQJh/4jhAv0JUh:j#hM>U%jGB hM>CJUVaJmH sH hAvhAv56CJaJhAv5CJaJjhAvUjhIU%jB hICJUVaJmH sH 4π345T g$s$s$$$0|5x$s$s$s$$${ v$"v$ U%$ /0v8 xHX "L^`La$$ ,08 xHX a$$ ,08 xHX dha$%$ /0v8 xHX "hL^h`La$ gdˍ̍&2>J$,$$$$$CsLsLsLsLs $ ,p@ P !$$Ifa$$ ) p@ P !a$$a$U JKmn̗͗WX"#2?JW=>خظشحإؙhEhAvCJ(aJ(hECJ(aJ( hAv6]hWhAvCJaJh6, hAv] hAv6jhAv0JUhAvhAv0JOmH sH hAv0JO5\mH sH hAv5OJQJ\^J?Z[JK789cd8?Bdoȵիաիիիի՝իիի՝ՙ՝Ցh oVh oV\h oVhAv\h%h oVjhAvH*UjhAv0JUjhAv0JU] hAv6] hAv6hAvCJaJhAvhAv6mH sH h oVmH sH jhAv0JUmH sH hAvmH sH 6RSpeg1$$RF$E($$4$ $$$RF$$q$$$E($O $$RF$ $Idh1$ & Fq  v@=^`=dh`U`gd*e & Fq   ^`U$ & Fq  V^`Va$7DKjkVWXY!:>aede,-.1=mѫ짣h@)8h,h.hZjho$0JUh shu+ho$hAvCJaJhA jhAv0JUh7Vht)h oV hAv6hAv h oVhAvh oVhAv\>+,67=q ()_b %94止 hAv6] hAvaJ"jhAv0JUaJhmH sH hEaJhmH sH hAvaJhmH sH hAvCJaJhEjh s0JUjhAv0JUjh 0JUhW#hAvh@)8jhk0JUh.hk1cefn- $"$!0$ $,$$RF$g*<$E($$$q$$$RF$O #$ ,0* xHX dh`a$$ & Fq  *:^`:a$ $dh`a$Idh1$$ & Fq  #^#`a$UJdh]^`45_`cf}GHBCGJi45jknѳ㳃hAv6\]aJhmH sH h{\aJhmH sH %jhAv0JU\aJhmH sH hAv\aJhmH sH  *hAvCJaJhECJaJhAvCJaJjhAv0JUhAv hEaJ hAvaJjhAv0JUaJ.>EGNQstuxy-    7 8      z+0k  ] ^   !h:h/ih(2 hAv5h{jhAv0JUjhAv0JUmH sH hAvCJaJhE hAv6]hAvjhAv0JU\E  /!#T&b)--.$q$$q$$'8$q$$$q$$,$$ $q$$$4$r$RF & Fq  B^`B#$ ,0* xHX dh`a$#$ ,0* xHX *dh`*a$  *dh`U&$ ,0* xHX dh`a$gdE !!##$$n$u$1%8%9%:%%%%%&&$'+','''>(](^(_(`(((K)S)**** ++++----.n/o/F0G0115566t7u777};;;;<<G=H=>?LAMAAABhD| hAv6hvzAhAvCJaJ hD|5 hAv]hT)hE hAv6]h: hAv5hAvjhAv0JUI.0357J:>aBGHHImJNOR:UCXK]`dgjjj{m$$E($$$$,$!0$$4$"$$RF$O $$4$E($q$$q$$'8$E($!0$$q$$$RF$  & Fq  ^` & Fq  b^`Udh`BBBB{C|C~CDDDDEEHFNFqFxFFFF GH IIIIIkJlJ4KAKKKLLLLLLEMFMnNpNqNIOJOSPTP9Q:QeQfQQQMRNR$S%S2S3SSSTTTUUUXXH^j^^^^^_ __0_Y_h  *hAvCJaJhAvCJaJhlhD|hV* hAv6hAvjhAv0JUPY_f_____a(a*a^bzbbbbbbbccc d dddff{g|gghhhhhjjjjjjjk:lHlll:mGmHmmmn ooƽƵhAv6]mH sH hAv5mH sH jhAv0JUmH sH hAv6mH sH hAvmH sH  *hAvCJaJhAvCJaJ hAv6]jhAv0JUh\h[h[6 h\hAvh[hAv hAv65oo-.0{jhAv0JUmH sH  h[6h[jhAv0JU *hAvCJaJhAvCJaJ hAv6 hAv6]hAvhAvmH sH h[mH sH D{mpVW45| !TU  Z[hkh$#jh~$0JUjh$#0JU hAv6]hN5h;jhAv0JUhEhAvCJaJhAvCJ(aJ(hAvhlk5EJLs1bwx [ $!0$ $ $Y$RF$"$!0$ $E($$RF$E($$$RF$E($ $$RF & Fs ^` $dh`a$ & Fs ^ & Fs ^gd$# & Fs @@^$da$U{|GHJKLscd\]FG^_45x     j k   [ T`#'-7qrh"1hV;mh]5OJQJ\^JaJhAv5OJQJ\^JaJhAh] hAv6jh$#0JUhAvCJaJhAvmH sH jhAv0JUhAvA  ij892t}wxbcpqITU].delp&  y z !!;!?!Z![!""jhi;0JCJUaJhHhPWh, hh,hxjhAv0JUh]hV;mh"1hAvJITUI "#*#%%')):*P,.H0[1F482;:<$"$,$!0$$4$t$$q$$$$$dA$$s$^s$^s$$$dA$`Ugd,$dh`a$gd,$dh`a$gd,gd[U"###*#+#,#v#%%&%'%%%'='>'?'A'E'Z'^'_'`'b''''''(((((((()))))))**0*7*:*;*=*f*P,Q,S,W,\,-...//'0E0H0I0[1\1U22q33F4G4V8m8n8s8888882; hi;6 hi;] hi;6]hi;CJaJjhi;0JCJUaJhi;T2;3;4;;;<5<<=7=m=n=o=E>>>>z??@@@GAAAABBBBBBBBB.C;CD/DEEEFFGGG&G=G>GCGIGOGPGiGjGkGnGoGpGqGGG張jehi;Ujhi;U hi;5\jhi;0JUjohi;Ujhi;U hi;6 hi;6]hi;hi;CJaJjhi;0JCJUaJ>:<@EG=GGHIGMOmPQQRxVpX[]`Xaac+dehzikmm$o$$s$s$ $s$$^s$$$dA$i$ $$$^s$$aZ $$s$^s`GGGGGGGHHHHHI I II8I;IJIOIPIiIjIzI{I|I}IIIIIIIIIKPLLLGMHMIMOOP:PmPnP}PPQQQQQQQQQRRSST,TTTxVyVzVW;XpXqXsXjhi;0JUjhi;U hi;6 hi;6]hi;CJaJjhi;0JCJUaJhi;jhi;UJsXxX YYYYZZM[[[[[\\I][]1__``````XaYaZaaaabDbIbcccccc+d,dDeeeemffffffffffffffffg g g g ghh hhhhjhi;Uj[hi;Ujhi;Uhi;CJaJjhi;0JCJUaJ hi;6 hi;6]hi; hi;aJHh ijiwizi{i|iCjjpkkkkkkkkkkkkl4mnm}mmmmmmnnnnnnnnnnnnooooooooooooppqq9uu+v.vvvvvv hi;>*jGhi;Ujhi;UjQhi;Ujhi;Uhi;CJaJjhi;0JCJUaJhi; hi;6]Bmquw"|}tg.~w7iÐ@ Yڔȕ.1P$$$ $s$s$$dA$$^s$^s$^s$g($^s$$dA$ $$$$g($$^s$$ $ $$ $$$vvvvvv}}tu.KQRkl|}$Mghi./|}~,j͇҇vwx܈.378 !ڵڵhi;mH sH  hi;6hi;CJaJjhi;Uj=hi;U hi;6]jhi;0JCJUaJjhi;Uhi;jhi;UB@-FLMfgwxz{ijkÐĐɐΐԐՐ45689@ABGےMv  $&+12Kjhi;Uj)hi;U hi;6jhi;Uhi;CJaJjhi;0JCJUaJj3hi;Ujhi;Uhi; hi;6]AKLMOPYZ[`̔єڔ۔ܔI|ŕȕɕ '()+,-./ *+,./0127<OPQRouvگjhi;Uhi;CJaJjhi;Uhi;mH sH jhi;U hi;6jhi;0JCJUaJhi;jhi;Ujhi;U@ؙޙ ?@áɡΡϡ{âƢǢ٢ޢ޿jhi;Ujhi;Ujhi;0JUhi;CJaJj hi;U hi;6] hi;6jhi;0JCJUaJhi;mH sH jhi;Uhi;>?ƢaV<2&.ڪ ..µ+$aZ $g($$$s$$$ $$z $s$s$s$s$$$$$$$^s$^s$$$aZ $$g($$ %*+DEUVXYabc *01JKLNOVW\bgh23 $&'(,˶˧˶ hi;6jhi;0JUjrhi;Ujhi;UmH sH jhi;UmH sH hi;6mH sH hi;mH sH j|hi;U hi;6]jhi;0JCJUaJhi;jhi;U7,./0=?EJKdeuvwxڪ۪ܪ&\^_`bh !#3U[]cdeiûûûûûûûû hi;6hi;mH sH hi;6mH sH jhi;Ujhi;U hi;6]hi;CJaJjhi;0JCJUaJhi;Hu|̯گ./04HNOhijklǰ./08:@EF_`acdjkmnpuԼjhi;UmH sH jhi;UmH sH jhhi;Ujhi;U hi;6hi;hi;CJaJjhi;0JCJUaJhi;mH sH hi;6mH sH ?Ųʲ˲CGIP۳߳%'()+2µõŵʵеѵ*+,.BHIbŪ hi;6hi;CJaJjhi;UmH sH hi;jhi;0JCJUaJj^hi;UmH sH jhi;UmH sH hi;mH sH hi;6mH sH >bcdefstvRSTl!067PQRTU]^_c~cdfξϾоҾӾιʴʬʡʬιδιʴιʬʖʬjJhi;Ujhi;Ujhi;U hi;6hi;CJaJhi;6mH sH hi;jhi;0JCJUaJhi;mH sH jhi;UmH sH jThi;UmH sH ;+sR]~c/fpz% Bq$$ $^s$$$$ $ $aZ $$ $$$$^s$ $z $s$$$$^s$$$a$s``t`tgd~t "#$&'./02fgimpqswz{}%&(tWr    !uBCDESXY͵hi;CJaJ hi;6]hi;6mH sH jhi;0JCJUaJhi;mH sH jhi;Ujhi;Uhi; hi;6GYrs46^qrY()*,TgӨӨhi;6mH sH hi;mH sH hi;CJaJmH sH j hi;U hi;6] hi;6hi;CJaJjhi;0JCJUaJjhi;Uj@ hi;Uhi;<)EMvH{$dA$^s$$m $$$dA$ $$ $s$$ $ $$^s$$s$s D`Dgd s `gd s`gd `gdk^`gd$k^`gd* Z^Z`gdo$`CDEFH\bc|}~!zиббРvvvh2Rhi;CJaJh2Rhi;6aJh2Rhi;6]aJh2Rhi;aJ!jh2Rhi;0JCJUaJ hi;6]j6!hi;Ujhi;U hi;6hi;jhi;0JCJUaJhi;mH sH hi;CJaJmH sH hi;aJmH sH .8uvwHInr{|}~5AFGabrsvw*j."h2Rhi;UaJj!h2Rhi;UaJjh2Rhi;UaJh2Rhi;6]aJh2Rhi;aJmH sH h2Rhi;CJaJ!jh2Rhi;0JCJUaJhv,hi;CJaJh2Rhi;aJh2Rhi;6aJ1p-./;@ [\][ .~$jk&'(sƾԷƾƾԷԷƾԷԲԷԷԷƾԲԪԟԪj"hi;Ujhi;U hi;6 hi;6]hi;CJaJjhi;0JCJUaJhi;h2Rhi;6]aJh2Rhi;aJjh2Rhi;UaJ hi;aJ>-[~j&sG\mH*>u  $dA$$ $aZ $$aZ $^s$ $^s$dA$ $$s$$ $ $^s$^s$^s$ $$$aZ $$H$s```stu_FGH H\]^{:mnoHIjpuv*+!N忸j#hi;U hi;6 hi;5\ hi;\j&#hi;Ujhi;U hi;\aJ hi;6]hi;hi;CJaJjhi;0JCJUaJB,-./>?@z}~;> hi;] hi;6j%hi;Uj$hi;U hi;6]hi;CJaJjhi;0JCJUaJj$hi;Ujhi;Uhi;D589;@uv  {            " # 0 5     Hdijƺj&hi;UmH sH jhi;UmH sH hi;6mH sH hi;mH sH hi;CJaJmH sH hi;CJaJj%hi;Ujhi;Ujhi;0JCJUaJ hi;6]hi; hi;64 "  /OC{D-rQ"d#$&'_(R)k+$s$^s$ $ $ $aZ $ $aZ $dA$^s$$$^s$^s$$$dA$$z $$^s$aZ $aZ $$^s$ $dA```/01OPQ}CDFMRXYstuxy{|ιιιʴʴʴʬʡʬιܒʴʬj}'hi;UmH sH j'hi;Ujhi;U hi;6hi;CJaJhi;6mH sH hi;jhi;0JCJUaJhi;mH sH jhi;UmH sH j&hi;UmH sH 5DEF-./4:?@Z[kloprsu%&)*}QR輵輵j(hi;U hi;]ju(hi;U hi;6] hi;6hi;CJaJhi;6mH sH hi;mH sH jhi;0JCJUaJhi;jhi;Uj'hi;U@,B5BBEoFGGI2LLMOfPPqQQR7STUOV$dA$$i$dA$$s$$aZ $$ $s$^s$ $$$dA$^s$ $ $aZ $$^s$$$aZ $$^s$^s``<<<J>K>?T@@@AAAABBBBB,B-B.B/B3B5B6B7BBBDDBEiEEEEoFpFqFGGGGGGGGGGGGGGGGGGIIIIIIKK2L3L4LLLMMMMNNj)hi;U hi;6hi;CJaJjm)hi;Ujhi;Ujhi;0JCJUaJ hi;6]hi;JNNNNNOOOOOOOO#OO PP]PcPePfPgPiPnPsPtPPPPPPPPPPPPPPPPPPPPPP˸Ӧ˔˅˔˔v˔jY+hi;UmH sH j*hi;UmH sH jhi;UmH sH  hi;aJ hi;6aJhi;6aJmH sH hi;aJmH sH hi;6mH sH hi;mH sH hi;CJaJjhi;0JCJUaJjc*hi;Ujhi;Uhi;.PqQrQtQ}QQQQQQQQQQQQQQQQQQQQRR7S8STTUbUUUU VOVPVgVnVVWWWiWlWmWoWsWWXXXYYY YY=Y@YAYCYGYZ%Z꺦꺦꺦꺦 hi;6] hi;6jO,hi;UmH sH hi;j+hi;UmH sH jhi;UmH sH hi;6mH sH hi;CJaJjhi;0JCJUaJhi;mH sH >OVVWY\\__dIfvgj$kkmoosxR|ÈR6$ $$s$$$$l$aZ $aZ $g($$^s$z $s$s$$l$$!q'$s$i$$$z `gd[ `gd[``%Z[[[[[[\\\\\\-_R_X_Y_r_s_t_u_v__________________`````````d d dddd2d3d4d7d8dTddddf hi;aJj;.hi;Uj-hi;UjE-hi;U hi;6]hi;CJaJj,hi;Ujhi;Ujhi;0JCJUaJ hi;6hi;?f f ffff1f2fBfCfFfGfIfJfLfLgbgvgwgxgiijjjjjjjjjjjjjjj k kkkk k$k%k&kkkkklllllllhi;6mH sH hi;mH sH j/hi;UaJ hi;aJjhi;UaJj5/hi;U hi;6hi;CJaJjhi;0JCJUaJj.hi;Ujhi;Uhi; hi;6]8lloooooosssswwwwwwwwwwwxxxxxoypyrysyuyyyyyyyyyyyyyyyyyyyyyyyyR|S|U|}}}}}}}ԽԲԽԽԧԽԽj0hi;Uj+0hi;Ujhi;U hi;6hi;aJmH sH hi;hi;CJaJjhi;0JCJUaJhi;mH sH hi;6mH sH @}}}}}~~~~~PXZ`efÈĈ #$RSTM_}  IeyΖܖ@P^_`1HJûûûj2hi;Uhi;CJaJjhi;0JCJUaJj1hi;U hi;6] hi;6hi;jhi;Uj!1hi;UE6^l7vs{Spl׷|eƼҽ$s$s$$s$$z $s$dA$$dA$ $$s$s$s$$$l$$ $i$ $$ $ ````JOUVopqstlmsx~֛כ؛ڛۛ78,Avwy}ʥj4hi;Uhi;CJaJj4hi;U hi;6]j3hi;Ujhi;0JCJUaJj3hi;Uj2hi;Ujhi;Uhi; hi;6<¡͡ϡСѡӡءǢܢKQS56FGIJaglm:­$pq hi;5\jv5hi;Uj4hi;U hi;6]hi;CJaJjhi;0JCJUaJhi;mH sH  hi;6hi;jhi;UCΰϰҰӰlmnrw}~±djophb0KQVWpqʵʦʵ🋟jb7hi;Uhi;6]aJj6hi;U hi;6]jl6hi;UmH sH jhi;UmH sH hi;6mH sH hi;mH sH hi;CJaJjhi;0JCJUaJjhi;Uhi;j5hi;U2϶жѶԶն׷طŸ|}~efkƼǼѽҽӽ3}99:;GAB hi;6 hV 0hi;!jhgchi;0JCJUaJjX8hi;Uhi;CJaJ hi;\j7hi;U hi;6]jhi;0JCJUaJhi;jhi;U>9AeI INWf2.-$ $dA$aZ $i$$aZ $$$s$$$$$$$$^s$P'$$$$ $^s$ $aZ $aZ $dA$gd$#gd;B18:c =efkqvwIJjhi;0JUhi;CJaJj9hi;UjN9hi;Uj8hi;Ujhi;U hi;6] hQVhhi; hi;6jhi;0JCJUaJhi;=  )Ocij   IJKH\`NOPT޷޷޷טhi;6]aJj:;hi;Uj:hi;Uhi;CJaJjD:hi;Ujhi;U hi;6 hi;6]jhi;0JCJUaJ hr*Nhi;hi;hEhi;6xj>hi;Ujhi;0JCJUaJj=hi;B*Uphjhi;B*Uphhi;B*phhi;6B*]ph!jh~$hi;0JCJUaJj&=hi;Uj<hi;Ujhi;U hi;6 hi;6]hi;/>VX^cd}~7tz{|TUw}j@hi;Uj?hi;Uhi;6]aJj?hi;Ujhi;0JCJUaJj>hi;Ujhi;Uhi; hi;6]B{T| l RD -7$aZ $$$i$$z $s$ $s$s$$s$$$$/$s$$ $aZ $^s$^s$aZ $aZ $^s$$$^s`Ogiotu%ZjBhi;UjoBhi;UjAhi;U hi;6jyAhi;Uj@hi;U hi;6]hi;CJaJjhi;0JCJUaJjhi;Uhi;j@hi;U8u  lm)l!'-.GHXY\]bc|}~ΰjDhi;Uj[Dhi;Uhi;CJaJjChi;UjeChi;Ujhi;0JCJUaJ h$#hi;jhi;0JU hi;6]jhi;Uhi;<>DIJcdtuxyBHRST ,jhi;B*Uphhi;B*phhi;6B*]phjGFhi;Uhi;CJaJjhi;0JCJUaJjEhi;UjQEhi;Ujhi;U hi;6]hi;5,-=>ABCDEJfhmst    "(-.GHILM-.Z+,./78׺ׯ׺ɧ׺ל׺׺ב׺׺j3Hhi;UjGhi;Uhi;CJaJj=Ghi;Ujhi;U hi;6]jhi;0JCJUaJhi;jhi;B*Uphhi;B*phjFhi;B*Uph9{]b !!@!!!n""##'C'''''(( (((,(-(.(1(2(:(;(%)?)A)G))))a+f+i++ ,,, ,!,C-D-E-//L2R2W2X2q2 hi;]j)Ihi;Uhi;CJaJjhi;0JCJUaJjHhi;Ujhi;U hi;6hi; hi;6]H= v!#':(),C-0222223333333$a$s$s$$^s$$dA$ $$sv:$Ugd,Lh1*&]hgdkmLh]hL&`#$q2r2222222222222222222222233 3 3 3333333ʹh,hAvhd0JKmHnHu hi;0JKjhi;0JKUh"1jJhi;U hi;6jhi;Uhi;jIhi;U", 001h:pUx/ =!"#$%n$-8ҸЊPNG  IHDRR@PLTE!!!!)!!)))11!!111911999B)1BBBJJJRRRZZZccckkksss{{{!!))))1B1RBZ9ZJk焔9999B!B!J)J)R1R1Z9Z9cBcJcJkRkRsZsZ{c{cks{{bKGDH cmPPJCmp0712HsIDATx^s۸YWGwJ=u\nn.8~D-es1۲<=2I#P<2ydxG9yM<&$TG*=Q5ІB#ʛ8ӔWʲZ'DZyw5LcMuYqGդWM)(o7- ]?Emd*fg~Zd`k,~|s~vۓWggof2.3!FfL+)2)Z+IˇWL^=?f6o8Ţfb6QNب-*V[py:&8,[nvkܾzWDx5m{fn$ؑ hZcXLu}0ȭ%yNak|ޘ＾U萚6- b5:ANHO>b& hׯηoޟ/d{rfksޜRFmEVjtK'YdR19^ٓr(2)o//D&?-Ӌr7v(VY d2ـWicqƮnz_DsK 3~@G~ 0* A]ܼ{Ni/%Ia QskS9Qxm*'0*p/0.:̆&LvoԞ@uLF+/6';"hsVvs&ZKhԽݜQB1;ê(9[\XhRǤd夯j)01/ q́Pz|S2}NbD7p1XK/pr(kEa30 e vX46Z,p8nG8զ+_B?^qS(Ye*9&zIęp0*N4\НBC1e˳+&]f_vV [gx% ZPlzvMU2!uRyd&-L'RiAğTK#xc;WqΫ-m1ާNNv72 I(;+5LQh#gɄH.]U`g¢xX x<o-":3'>Tdd, H6N1¡L^S?b5"US{&<`OR>pK.ͮf } JO6''Hj=,L*7؏ABڹ*L N ~cu]ʞNm=(DyvlJ&Y,a ҝS(5~.0 [܅0J& N_t69h]_2!.GdI#u֗$j7?Md"k!}t=" oJ .XȲ+ "?[mQWLifLҺCD^A|>9*OGKJ(,:_y,!(OEĄ3NJWEӲd2 On)k$"&j &7ͷ ?e Ogu M0Dk/(1H^~[;P3})E1HwefA[R:(C>pȟP()1!H^Hüs%TfL@G4PK_,X4yϝ:9 s5H+1KT%ahT;6432^%"+DRvVdJ9q9ש´cRU668\R;jA5ЂI6tkF;ݤ0Fb"IgTCZ>|HLU'fShDXneޥI0j?!cٞdR\I)o1[l:1Iؓ6=]KYͶ_b\q0X4Ӹ .?sQ TAw32#tU{ =0H0Fn4 H3h-C|dqlWjD*S`l_$Y63A1_-芷a/b(񆽖Uƻ9ɔ9:*x|?Ds& X\)ؿCNpn+P~մ4Uʣc~uL=x-;D*"#9{R!D^*;NP`èb$fPܢiV:,LQ%]TpM_1:또`PHb;w@w &{ ݾ]W_} *oX aèts &N9VNu5ystL]PHi :0iٓJ7~/zQ((ͶWD2~h^Y'%YNT;Aµn_[^&MgO,"}h"o^`a\":ԶHLjS ,Z9ZS]qG -V^ždr(̕Vb ynDGnos&vrr\#S nCD6l[!9o|)H~PZW=ʿmʄ7"cUѝk 3(n[OƿmOx(Xz_D\ڲTQENZM~PI:{9arC&,90g4ߜHnh@c,6?) y 6P%L}at=bbIZƍO\Lg#^qfӓ" ,/YٔC,i`aQ,V1Wljr7`b~:(䛱yAE(5} | TNPM@ v+( ZN#9)ךrLћT} Dy0~Y$AAaSn!&[FI-e"&,Ki5b %2wkdւ$qH>}C2MtǠ工 +zZ.BӲ^ɪCW{eDLgZ]wZi-F^Wpd6{[29jhgҴ,EK;Z$Qdh)7Lsjk &w&繲/ =`h2;'1_{g`Ey5AP(6P$փlǯ₮ij`f‘d vܰb5{6—H:pq K9^P_;?a[TYY$Q~/P$^ϊhIFωl#yE2aF$#?{(2 pRsRM\&Yݲfw'TI[Sϗ?rLZ2$z}8fTdH5g+Yh4Xibhmp"k"0,s%ՓdlҼ =B%^ BIv)GUqtO,`pyvIYEy).\)eNa,-yz5կ֐ÝeuRCԅ0K=Ga)Ul ]Հխ,n.c J\2 {Z#̒(I1LB*X0ĄaOb*H eGX{AHݠqV1aϑa&IAQ Iއ2!Q)G`aB0ƭwǥ#lЮQF@dq?}țbQxh {1NWoqg}bqP`P[Mo nY)䣺}򣶃s=1˙ᎵHtQh՚eL(O* ɒOЋXoi/wv6e̒@A᳓gA|4t|G cʤeGFITv4Kc_Ś4us8 CHWzi}ohf"g#Gk&t\0AOV&S(o.'I_&>Xs҆>$JV2}۫*L/ ƄeSwgQbINRm3=1찫],%F8-ۻO[ZP&#EQSoSk>R2iDoFմrBkԺDO LZ1IҪLϷחؼ#VHtiǏ5+q6|p.8CIOc=-1 i66F{]TL^p$&#ݨCQMnJs{ba\7%!O Iӳ=_u`~r;dk܇1@m,L(gbaDJ󯑡 OK&@cYVEw6|cRNLus{C39$C/s&͈IܝVÚဇ?<0"Cc7Ʀݱ?p035m/h^ط AөLFlׅqpL'7'v&qLIS+ ,qbd-XnK\Ӟf+F8MFܷLԅ Ni ,%Q==v۞:zk8rkQ+'U[ZN:u M0^GJZNqezhSOJck^S罙ΛhH+G #I&6ͪ("ZL&dѕ ݤ)[ YhDpMUʌo^_0ӕ {{y= JP'AFԷ[ QSE5J VZ B l_VffGU~ǧ{4]菲A RhY?\ O(vs"?Mysb2QeЫh~gpN%|aaAPLw0C}+ZjW~DvDG{#zbK ٦'؆3C^vHgcAֶq'6]WԴxEtN"N٧&݁*}# w*t0?ӟ04z+OYoffQiPaԛ Ϯ{{LSVCߝ5ާ.4wlUUvG`UPc`[l _f{[g}d4 l&7|f M|f0=`>cg3CaE{xOd/G8'JB{]S'e[~7~o#+ 8J9cor tq+w)FH[Pp~X)~tJ'ؗ~-m'{{A8;Ϸ?8X\Dd b  c $A? ?3"`?2Po~kq,g`!$o~kqqND|5xXAhGW֊I6R6mz@V$;`psh jRXRm: KAP]T!zp S|HB(饸 m W뵢$R*f}Wc8P{'BfQj QAC7 VKϏn 4=:V[{족!‘sŅ[ŏ\ B^bh/smVs4~[;`7mذV'stӂi #tjc*dyfA43 |D%υ/$4n{:MȂ 'Ε?:?K $vN/JU~%@_JݤVkoߚ{V[l[m,dMV&ؚ`kxH+*U~ ɥFWnQMfҩL6q\T;w;Ke)fs\\.5gUvOceI:x_LSU8F7H=]_ͼDtbWS:{7ެ*w^RL_k|V\x=yP.\.$GǒntNqP Yuv9n@DLgwM9hRo׭ x3Ot:$$If!vh55$5$5$5$#v#v$:V l06,55$/  4aT$$If!vh55$5$5$5$#v#v$:V l06,55$/  4aT$$If!vh55$5$5$5$#v#v$:V l06,55$/  / 4aT$$If!vh55$5$5$5$#v#v$:V l06,55$/  4aT{DyK  _Ref39279574{D  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~kRoot Entry FP/@ Data JWordDocumentZ'ObjectPoolZP/_1120389570 FZZOle PRINTCompObje  !"#$%&'()*+,-./012347:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijkmnopqrtuvwxyz{|}~ FMicrosoft Graph ChartGBiff5MSGraph.Chart.89q B""$"#,##0;\-"$"#,##0"$"#,##0;[Red]\-"$"#,##0"$"#,##0.00;\-"$"#,##0.00#"$"#,##0.00;[Red]\-"$"#,##0.005*0_-"$"* #,##0_-;\-"$"* #,##0_-;_-"$"* "-"_-;_-@_-,)'_-* #,##0_-;\-* #,##0_-;E0 $    h   '' Times New Roman-"System----Times New Roman--------------------------------------------------'- g -'- g -'- dG --  $cBY B c- $ccmBc$BmY mY BB-cBBY #cY cKKY +cY cTTY 4cY c\\Y =cY ceeY EcY cmmY -$Y B cBY B-$ccmBc$BmY mY BB-MM- MMMM$ttmBt- $tt-ss- ssss$tmmt-MM- MMMM$&B- $&&-ss- ssss$&+&&-MM- MMMM$MMB- $(((-ss- ssss$Ms(-MM- MMMM$II.BI- $ooIIo-ss- ssss$I..oI-MM- MMMM$TB- $-ss- ssss$TT-MM- MMMM$  " " B - $  -ss- ssss$ " H  ---c ccK#c#KcK+c+KcK4c4KcK=c=KcKEcEKcK 2 d 02 2  22 2 l 42 2  62 2 u 82 2 1022 2 }1222 2 1422 2 1622 2 1822 2 2022------'--   Times New Roman-"2 $Deaths per 100,000H,2!8'8,,222222-----'-- dG c  cc99  2 15-19D22!222 20-24D22!222 !25-34D22!222 h35-44D22!222 45-54D22!222 55-64D22!22------'--  3]= 2 xK Age GroupH2,M,288----'-- dG --'-- g -------'--  J 2 gTMotor Vehicle Death Rates: Canada 1997 f6$60N0<00N06$<N6$0*$N6<6<66666----'-- g --'-- g --' ' 'ObjInfo Workbook8 _1120389703 FZZOle 5_-* "-"_-;_-@_-=,8_-"$"* #,##0.00_-;\-"$"* #,##0.00_-;_-"$"* "-"??_-;_-@_-4+/_-* #,##0.00_-;\-* #,##0.00_-;_-* "-"??_-;_-@_-"$"#,##0_);\("$"#,##0\)!"$"#,##0_);[Red]\("$"#,##0\)""$"#,##0.00_);\("$"#,##0.00\)'""$"#,##0.00_);[Red]\("$"#,##0.00\)72_("$"* #,##0_);_("$"* \(#,##0\);_("$"* "-"_);_(@_).)_(* #,##0_);_(* \(#,##0\);_(* "-"_);_(@_)?:_("$"* #,##0.00_);_("$"* \(#,##0.00\);_("$"* "-"??_);_(@_)61_(* #,##0.00_);_(* \(#,##0.00\);_(* "-"??_);_(@_)1Arial1Arial1Arial1.Times New Roman1.Times New Roman1.Times New Roman1Arial= P,##0.00_< ` >` J ` J 883ffff̙̙3f3fff3f3f33333f33333\R3&STU15-1920-2425-3435-4445-5455-644@2@&@333333!@ @%@WY P -w= ,>>X4H^3qdd 3Q  3-D Column 1QQQ3_4E4D $% M`3O&Q4$% M`3O&Q4FAUt 3OUt 3 b#M!   43*@@#M! M! M NM4% - OM3OW!Q  Age Group'4% ]BMZ3O!Q (Deaths per 100,000'432" :dd44% dT nM23O^"&Q TMotor Vehicle Death Rates: Canada 1997 '44  FMicrosoft Graph ChartGBiff5MSGraph.Chart.89qPRINT  CompObj6eObjInfo 8Workbook9 0 &    {   '' Times New Roman-"System----------------Times New Roman---------------'- z -'- z -'- Z --  $k s k- $kkKk$K K -k dk## krr k Qk k^^ k =k kKK -$ s k -$kkKk$K K -f- ff$'- $''-- $'''-f- ff$=55- $==-- $=5Q==-f- ff$66p6- $RR66R-- $6ppR6-f- ff$E- $-- $EE-f- ff$9 9 - $-- $9 U-f- ff$: : @  : - $U U @: @: U -- $: @  U @: @---k kkPdkdPkPkPQkQPkPkP=k=PkP 2 02 2 )200222 2 x400222 2 600222 2 800222 2 d{10002222 2 {12002222 2 {14002222 2 Q{16002222------'--  Z Times New Roman-%2 Injuries per 100,000'8!8,,'8,,222222-----'-- Z k s k0k00o0o0 0 s 0s 2 J15-1922!222 J20-2422!222 JP25-3422!222 J35-4422!222 J45-5422!222 JS 55-6422!22------'-- jt 2  Age GroupH2,M,288----'-- Z --'-- z -------'--  ' 2 EUMotor Vehicle Injury Rates: Canada 1997 h7%71O1=11+=%=18O7%1+%O7=7=77777----'-- z --'-- z --' ' ' B""$"#,##0;\-"$"#,##0"$"#,##0;[Red]\-"$"#,##0"$"#,##0.00;\-"$"#,##0.00#"$"#,##0.00;[Red]\-"$"#,##0.005*0_-"$"* #,##0_-;\-"$"* #,##0_-;_-"$"* "-"_-;_-@_-,)'_-* #,##0_-;\-* #,##0_-;_-* "-"_-;_-@_-=,8_-"$"* #,##0.00_-;\-"$"* #,##0.00_-;_-"$"* "-"??_-;_-@_-4+/_-* #,##0.00_-;\-* #,##0.00_-;_-* "-"??_-;_-@_-"$"#,##0_);\("$"#,##0\)!"$"#,##0_);[Red]\("$"#,##0\)""$"#,##0.00_);\("$"#,##0.00\)'""$"#,##0.00_);[Red]\("$"#,##0.00\)72_("$"* #,##0_);_("$"* \(#,##0\);_("$"* "-"_);_(@_).)_(* #,##0_);_(* \(#,##0\);_(* "-"_);_(@_)?:_("$"* #,##0.00_);_("$"* \(#,##0.00\);_("$"* "-"??_);_(@_)61_(* #,##0.00_);_(* \(#,##0.00\);_(* "-"??_);_(@_)1Arial1Arial1Arial1. Times New Roman1.Times New Roman1.Times New Roman1.Times New Roman= 1)!,##0.00_P ` K` J ` J 883ffff̙̙3f3fff3f3f33333f33333\R3&STU15-1920-2425-3435-4445-5455-64D@@@ @@@WY<-w=  >X4`3ddn  3Q  3-D Column 1QQQ3_ O (  MMd4E4D$% M`3O&Q4$% M`3O&Q4FAYA 3O[< 3 b#M!   43*#M! M! M NM4% t 7M3Oq+Q  Age Group'4% TIMZ3O+Q ,Injuries per 100,000'432" :dn44% 7& ]M23O.&Q UMotor Vehicle Injury Rates: Canada 1997 '44 Oh+'0  0 < H T`hpx ssErika ChamberlaiyK  _Ref39317654{DyK  _Ref24461412{DyK  _Ref27041912{DyK  _Ref27037342{DyK  _Ref31550738{DyK  _Ref27037342{DyK  _Ref31550835{DyK  _Ref24703662{DyK  _Ref27037342{DyK  _Ref27037342{DyK  _Ref24703662{DyK  _Ref24703662{DyK  _Ref24703662{DyK  _Ref27141668{DyK  _Ref39319539{DyK  _Ref39319539{DyK  _Ref39319833{DyK  _Ref39319863{DyK  _Ref39319539{DyK  _Ref24773057{DyK  _Ref27039357{DyK  _Ref27039357{DyK  _Ref30678626{DyK  _Ref39319539{DyK  _Ref31552096{DyK  _Ref28663687{DyK  _Ref28663687{DyK  _Ref39410271{DyK  _Ref39410271{DyK  _Ref28663687{DyK  _Ref28663687{DyK  _Ref39410505{DyK  _Ref27040181{DyK  _Ref39410505{DyK  _Ref24198695{DyK  _Ref28663687{DyK  _Ref24199174}DyK _Ref524766210}DyK _Ref524766210{DyK  _Ref39412874{DyK  _Ref30754198{DyK  _Ref31271506{DyK  _Ref31272227{DyK  _Ref28663687{DyK  _Ref39414949{DyK  _Ref27040181{DyK  _Ref27040181}DyK _Ref519656913}DyK _Ref519656913}DyK _Ref519656913{DyK  _Ref40165698}DyK _Ref519666271{DyK  _Ref24382586{DyK  _Ref39414949{DyK  _Ref28663687{DyK  _Ref39279574{DyK  _Ref31618674{DyK  _Ref31621750{DyK  _Ref39473284{DyK  _Ref39319539{DyK  _Ref39412874{DyK  _Ref28663687{DyK  _Ref27040181}DyK _Ref524766210}DyK _Ref524766210{DyK  _Ref39279574{DyK  _Ref39279574{DyK  _Ref39474413{DyK  _Ref39474587{DyK  _Ref24709545{DyK  _Ref39474647}DyK _Ref519666271{DyK  _Ref39279574{DyK  _Ref39279574{DyK  _Ref39474944{DyK  _Ref39474944{DyK  _Ref27041912{DyK  _Ref31621217{DyK  _Ref27039357{DyK  _Ref31621366{DyK  _Ref26510034{DyK  _Ref26510034{DyK  _Ref26510034{DyK  _Ref30754198{DyK  _Ref39412874{DyK  _Ref27041912{DyK  _Ref31550317{DyK  _Ref30048273{DyK  _Ref28663687{DyK  _Ref30048273{DyK  _Ref27040181{DyK  _Ref28663687{DyK  _Ref25315493{DyK  _Ref31621366{DyK  _Ref39761201{DyK  _Ref25315493{DyK  _Ref31622532{DyK  _Ref39761269{DyK  _Ref27097303{DyK  _Ref25315493{DyK  _Ref39761269{DyK  _Ref25415210{DyK  _Ref39761269{DyK  _Ref31621366{DyK  _Ref27095291{DyK  _Ref39761201{DyK  _Ref39761482{DyK  _Ref27040181{DyK  _Ref31623195{DyK  _Ref31632202{DyK  _Ref39761201{DyK  _Ref31623436{DyK  _Ref31621366{DyK  _Ref27040181{DyK  _Ref31623195{DyK  _Ref39761691{DyK  _Ref31621366{DyK  _Ref27040181{DyK  _Ref39566579{DyK  _Ref39566579{DyK  _Ref27095953{DyK  _Ref253144291TableSummaryInformation(lDocumentSummaryInformation8sCompObjj      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijmlnoqrstuvwxyz{|}~V@@@ NormalCJ_HaJmH sH tH T@T Heading 1$$@&a$5CJ$OJQJ\^JV@V Heading 2 $@&5CJOJQJ\]^JaJN@N Heading 3 $@&5OJQJ\^JaJ~@~ Heading 4K$$ ,0* xHX dh@&`a$5\F@F Heading 5$$@&a$ 5CJ(aJDA@D Default Paragraph FontVi@V  Table Normal :V 44 la (k(No List V+V  Endnote Text dx^ `CJaJ>*> Endnote ReferenceH*nB@n Body Text=$ ,08 xHX da$aJ\@"\  Footnote Text$(*$^`(a$CJaJ@&@1@ Footnote ReferenceH*XC@BX Body Text Indent 1$^ `CJaJfORf 1AutoList1#$ 01$^`0a$CJ_HmH sH tH VObV 2AutoList1$`1$^`a$CJ_HmH sH tH VOrV 3AutoList1$`1$^`a$CJ_HmH sH tH VOV 4AutoList1$`1$^`a$CJ_HmH sH tH VOV 5AutoList1$`1$^`a$CJ_HmH sH tH VOV 6AutoList1$`1$^`a$CJ_HmH sH tH VOV 7AutoList1$`1$^`a$CJ_HmH sH tH VOV 8AutoList1$`1$^`a$CJ_HmH sH tH fOf 1Paragraph#$ 01$^`0a$CJ_HmH sH tH VOV 2Paragraph$`1$^`a$CJ_HmH sH tH VOV 3Paragraph$`1$^`a$CJ_HmH sH tH VOV 4Paragraph $`1$^`a$CJ_HmH sH tH VOV 5Paragraph!$`1$^`a$CJ_HmH sH tH VO"V 6Paragraph"$`1$^`a$CJ_HmH sH tH VO2V 7Paragraph#$`1$^`a$CJ_HmH sH tH VOBV 8Paragraph$$`1$^`a$CJ_HmH sH tH tORt 11B%$ ,h8 xHX (#h1$^ha$CJ_HmH sH tH tObt 21B&$ ,h8 xHX (#h1$^ha$CJ_HmH sH tH tOrt 31B'$ ,h8 xHX (#h1$^ha$CJ_HmH sH tH tOt 41B($ ,h8 xHX (#h1$^ha$CJ_HmH sH tH tOt 51B)$ ,h8 xHX (#h1$^ha$CJ_HmH sH tH tOt 61B*$ ,h8 xHX (#h1$^ha$CJ_HmH sH tH tOt 71B+$ ,h8 xHX (#h1$^ha$CJ_HmH sH tH tOt 81B,$ ,h8 xHX (#h1$^ha$CJ_HmH sH tH <O< Default Para mH sH ulOl _levnl90. h8 xH 1$^ CJ_HmH sH tH lOl _levnl80/ p@ P@ 1$^@ CJ_HmH sH tH pOp _levnl730 # h8 xH 1$^ CJ_HmH sH tH pOp _levnl631 # p@ P p1$^pCJ_HmH sH tH rO"r _levnl562 & h8 xH1$^CJ_HmH sH tH rO2r _levnl463 & p@ P 1$^CJ_HmH sH tH vOBv _levnl394 ) h8 xHX 81$^8CJ_HmH sH tH vORv _levnl295 ) p@ P !1$^CJ_HmH sH tH xObx _levnl1<6 ,h8 xHX (#h1$^hCJ_HmH sH tH lOrl _levsl907 h8 xH 1$^ CJ_HmH sH tH lOl _levsl808 p@ P@ 1$^@ CJ_HmH sH tH pOp _levsl739 # h8 xH 1$^ CJ_HmH sH tH pOp _levsl63: # p@ P p1$^pCJ_HmH sH tH rOr _levsl56; & h8 xH1$^CJ_HmH sH tH rOr _levsl46< & p@ P 1$^CJ_HmH sH tH vOv _levsl39= ) h8 xHX 81$^8CJ_HmH sH tH vOv _levsl29> ) p@ P !1$^CJ_HmH sH tH xOx _levsl1<? ,h8 xHX (#h1$^hCJ_HmH sH tH lOl _level90@ h8 xH 1$^ CJ_HmH sH tH lOl _level80A p@ P@ 1$^@ CJ_HmH sH tH pO"p _level73B # h8 xH 1$^ CJ_HmH sH tH pO2p _level63C # p@ P p1$^pCJ_HmH sH tH rOBr _level56D & h8 xH1$^CJ_HmH sH tH rORr _level46E & p@ P 1$^CJ_HmH sH tH vObv _level39F ) h8 xHX 81$^8CJ_HmH sH tH vOrv _level29G ) p@ P !1$^CJ_HmH sH tH 4 @4 Footer H !hR@h Body Text Indent 2I$dh1$`a$aJhmH sH TOT double indentJ$]^a$aJ.)@. Page NumberF@F HeaderL !1$CJaJmH sH 6U@6 Hyperlink >*B*phFV@F FollowedHyperlink >*B* ph4O4 15CJOJQJ\^JaJ44 TOC 1P$ $ a$BBTOC 2Q $ ^ mHnHu@@ TOC 3R $ dh^.. TOC 4 S^HBH  Balloon TextTCJOJQJ^JaJOR :Style Justified First line: 0.5" Line spacing: 1.5 linesU$dh*$`a$aJp ? 5( ..6/58:BKKL]MN0OPPPPQbSUUW{WU\]]{^_b.ccf(ggqhij2o2r1sHww\xz{|}~~eI˂&Dž̏W"ئU`UEON?ݼ¾)޿m>ZJ7cXd-( 4_GB4j7      ]9_ "" ##n')-t//34G5L99::{;<=AkBCDEEpFSH9IIMJ$KKLP[^{_`bdQqtsIuTu{{|t}J~ Ԁ4-؇D0E!I'y´w'^.ϾEYUV4  {Gc\F^4jqi 8   wbp+  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./01h{o  6>#d$$$%B'*F--k.s.+035@9=>?@A C]EFJOKO+[\^`a6bbefujlmnKnnbqq0r?rsttuuuvjw6z|~et|ۇ0KkvNɔdԙƞО{K_hߨZ-KٳHP̼ѾھM2P|G::x!VV  E##$P%J')&*N+d,-.. /W/0H2334:7:<8=&BCDGzHI M*Q VYUbf@jmtxx|~DdNƌŽޓ8-Ҙ(Llc/"? OF\ J8 obPeqf`3O9u  IL_     !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^6l d$$%}(+0R;AGJT^eknbq?rukd_-HݻH2P:ax  $N+W/037:7:@-G=P6anx~NNƌ0FI oI7`IL  5 !"#$%&'()*+,-./0123456789:;<=>?C@ABCDEFGHIJKLMNOPQRSTUVWXqYZ[D\]^++_ z z z z z z z z z z z z z z z z z z z z z z z z z z z z z z z z z !z "z #z $z %z &z 'z (z )z *z +z ,z -z .z /z 0z 1z 2z 3z 4z 5z 6z 7z 8z 9z :z ;z <z =z >z ?z @z Az Bz Cz Dz Ez Fz Gz Hz Iz Jz Kz Lz Mz Nz Oz Pz Qz Rz Sz Tz Uz Vz Wz Xz Yz Zz [z \z ]z ^zCK R! 1).l08OBK5O9QVN]_sciYov5y|˅[3r^[&1?xwi "}'M/.5a:<mB\HN;X=`{eRpu{v'\?g`+U-D=b[ I+ H   a   6u/aH-'g? D!a" #$%6&v'()*$+,j-Y. /V0L1213V4536l7 89:;<=>?M@AB!CDE2F0GHIUJ&KL*M@NOP5QDR| S=T`UV: W XYnZ[\]a^ !"#$%&')*+,-./0123456789:;<=>?@ABCUVeg~/z{|5^@&$| @aA g%#%a)+.k0000\48;?OBE1IKAK5OoPpPPRTW[}^bfjleptttiuv^x_x~xxy3y4y5yTyyy z z zz|gd˅̅&2>JKUZ`fmnx}Y[debr^_RSpeg1cefn- /Tb!%%&(+-/J26a:?@@AmBFOJ:MCPKUX\_bbb{ehu"/ O  C{D-rQd_ R!k#%(,,/3J68,:5::=o>??A2DDEGfHHqIIJ7KLMONVOQTTWW\I^v_b$cceggkpRtz|ÀR6^l7vs{Splׯ|eƴҵ9AeI INWf2.-{T| l   R D -7=v: !$C%(*****++++++0000x0000000x0x000000000000000000000000(0 00 0 0 0 0 0H00C0C0C0C0C0C0C0C0C0C0C0C0C0C0Cx0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0C0CxH000I0I0I0I0I0I0I0I0I0pI0 I0 I0I0I0x0(0k0(0k0U00U000U00U00xU00U008U00@U00(0k0U0KU0K(0k0H(0k0xU0pPU0pPU0pPU0pPU0pPU0pPU0pPU0pPU0pPU0pPU0pPhU0pP(0k00tp0tx0tp0tp0tp0t0t0t0t0t0tx0t0t0t0tx0tx0txU0tU0tU0txU0t0t00̅0̅0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅ 0̅0̅0̅0̅0̅0̅0̅(0̅U0U0U00(0(0U0U080U0eU0eU0eU0e80U0U0U0U0U0U00(0(0U0_U0_U0_U0_p0_(q 0U0SU0SU0SU0SU0S(q 0U0U0U0U0U00(q 0U0I0J0U0J0U0U0(q 0I0fI0fU0fU0fU0f0f(q 000U0xU0U0000U0U0U0U0U0(q 00%0% 0%0%(0%(U0%U0%0U0%xU0%0%(q 00@0@@0@@0@@0@H0@H0@p0@P0@0@0@(q 0U0bU0b`U0b0bp0b(q 0U0SpU0SphU0SppU0SpU0SpU0SpU0SppU0SpU0Sp0p(0(00000000(r 0U0000(r 0I05I05U05U05U05U05(r 0U0ƴU0ƴU0ƴU0ƴ0ƴ0ƴ(r 0U0YU0YU0YU0YU0Y0Y00(s 0U0U0U0U0U0U0U0U00(s 0U0LU0LU0LpU0L0L(s 0U0xU0xU0x(s 0U0(00[(s 0U0U0U0pU0H0p@0 p0U0p0p@0@0@0@0@0@0@0x@0@0@0@0@0@0x@0@0@0@0x@0@0@0x@0@0@0P@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0x@0@0@0@0@0@0@0@0@0@0@0@0@0x@0@0h@0@0@0@0H@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0x@0@0@0x@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0x@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0x@0@0@0@0@0@0@0@0@0@0p@0@0@0@0@0@0@0@0@0@0@0x@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0x@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0x@0@0@0@0@0@0@0@0h@0@0@0@0x@0@0@0@0@0@0@0@0@0@0@0@0@0@0p@0@0@0@0@@0p@0@0@0x@0@0@0@0@0@0@0@0x@0@0@0X@0@0@0x@0p@0@0@0@0x@0@0@0@0@0@0@0p@0@0p@0@0@0@0@0@0@0@0@0@0@0x@0@0@0@0x@0@0@0x@0@0@0@0@0p@0x@0x@0@0@0@0@0@0@0@0@0@0@0@0p@0@0@0@0x@0@0x@0@0@0p@0@0@0@0@0@0@0@0@0x@0@0@0x@0@0@0@0x@0@0@0@0@0p@0@0@0@0@0@0x@0@0@0@0x@0x@0@0@0@0@0@0@0@0@0@0@0@0@0@0p@0@0@0@0@002W@L0@L0@0L0@00lR0000123456789A g%#%a)+.k000\48;?OBE1IAK5OoPPRTW[}^bfjleptttvJKmnY ITUI*%H([)F,023:48=?=??@AGEGmHIIJxNpPUXXYY[+\]zaceeiuo"tutwyg{|~}w7aV<+>;00h6<;00>;00>;0 0`6<;0 0<;0 0<;0 0<;00<;00<;00<;00<;00<;00O900L"DM900O900>;00<;00<;00>;00>;00<;00>;00>;00>;00>;0"0<;0"0<;00<;00>;00<;00>;00>;00<;0*0V6<;0*0<;0*0>;00<;00>;00<;00>;00>;020>;0206<;020>;020>;00<;00<;070<;00>;00<;00>;00<;0<0<;00>;00<;00>;00<;0A0>;00>;00>;00(\C<;00(<;00(C<;00(<;00(C<;00(<;00(D<;00(<;0L0b<;0L0<;0L0<;00>;00>;00 >;00>;00>{00Oy00My00Oy00 0>>{00<{00`lX=H.1>{00>{00(6<{00<{00UwTXYZ>{00<{00>{00>{00>{00<{0 0KrK20>{00>{00(+<{00>{00@0<{00(x+<{00>{00@0>{00<{00(`+>{00<{00@<Y<{00<{00@0Oy0 0>{0 0<{0 0<{0 0@0>{00>{00<{0&0<{0&0@0gm>{00>{0*0(4}6<{0*0<{0*0@0my >{00>{0/0(6>{0/0>{0/0@0>{000  /Igzخ4!BY_ofH"2;GsXhvK,bYsR+<NP%Zfl}JBTQ>,q23 "$%'(+235689;=>?ACEFHIKLNOPRSTUWXYZ[]^`abdefhijkmnoqrstvwxy{|}~3 a%#XπgJfR .{m:Ƽ:<m+ k+OV63!#&)*,-./0147:<@BDGJMQV\_cglpuz3_xzx|x5yPyRy+::!!   $$$$$$&&&CDDDD_DbDHII!LLLOLMT]]]` a alllm nn*nEnHnpnnnqqqct~ttfuuuuuu9wdwgw$OR:UXˇ͇ ;>&AC %(^y|٩ݩAlnӳֳ 9=(CE37W,0?jmP{Lgk%>%@%:,U,X,-..%.@.C..//0/K/N/<<<<=!=(>C>F>mAAAlCCCH!H$HGHrHuHU5U9U$W?WCW[5[9[^^^Yjujyjuuuuv vxxxy-y0y~q‰$(ӎŏדۓ &* !8cf̦ګܫ(,Gru"=?r268SW )DH ;?JN '+-HL%@D+VY8<:ei/Z]hV$?ChAELHHHHtHtHHtHtHtHtHtHtHtHtHHHHHHHtHtHHTHHtHTHHHHHHHHHtHTHHHHHHHHTHHtHHHHHHTHTHHTHHHHHHHTHHHTHHHHHHHHHHHHHTHtHTHHtHtHHHtHtHTHHTHtHHtHHHtHTHHtHHHTHHTHHHHtHHtHHTHTHtHTHHTHtHHHTHtl,b$-8ҸЊ,.@<(    S ^A> http://www.madd.ca/newlogo.gif#" `B S  ?'+ MT _Toc36985761 _Toc38694074 _Ref28663687 _Ref27041912 _Ref39412874 _Ref31621750 _Ref39410505 _Ref39279574 _Ref27040181 _Ref27141668 _Hlt30940491 _Toc36985762 _Toc38694075 _Toc36985763 _Toc38694076 _Ref31552096 _Ref31550317 _Toc36985764 _Toc38694077 _Toc36985765 _Toc38694078 _Ref27037342 _Hlt31550958 _Ref39319539 _Ref39473312 _Ref24703662 _Hlt31550861 _Hlt31551068 _Hlt31551655 _Ref24703298 _Ref31550738 _Ref24199174 _Hlt31552695 _Ref31550835 _Ref24773057 _Ref24775104 _Hlt31552036 _Ref30678626 _Hlt31552043 _Toc36985766 _Toc38694079 _Ref39319833 _Ref39319863 _Ref24774866 _Hlt31621189 _Ref27039357 _Hlt31551956 _Hlt31621231 _Ref31621217 _Toc36985767 _Toc38694080 _Toc36985768 _Toc38694081 _Toc36985769 _Toc38694082 _Toc36985770 _Toc38694083 _Toc38694084 _Toc38694085 _Ref39410271 _Ref39474944 _Toc36985771 _Toc38694086 _Ref524766047 _Ref525114817 _Ref39414949 _Ref524766210 _Toc36985772 _Toc38694087 _Ref24198695 _Hlt31553311 _Ref519666271 _Hlt31614636 _Hlt31706039 _Hlt31706059 _Ref24709415 _Hlt31706057 _Toc36985773 _Toc38694088 _Ref31621366 _Ref27095291 _Hlt31618933 _Hlt31618936 _Ref31632202 _Ref39761201 _Toc36985774 _Toc38694089 _Ref27095724 _Ref31272227 _Ref29735431 _Ref29735443 _Ref31271506 _Toc36985775 _Toc38694090 _Ref30754198 _Ref30756493 _Ref30756099 _Ref30942871 _Toc36985776 _Toc38694091 _Ref24706238 _Ref27040565 _Ref27097501 _Ref519656913 _Ref24382586 _Ref40165698 _Ref24461099 _Toc36985777 _Toc38694092 _Ref24709545 _Hlt31620604 _Ref39474587 _Ref25926048 _Hlt31632501 _Ref25926059 _Toc36985778 _Toc38694093 _Ref31618657 _Ref31618674 _Ref39473284 _Toc36985779 _Toc38694094 _Toc36985780 _Toc38694095 _Ref39317654 _Ref24461412 _Hlt29737024 _Ref27041224 _Ref39474647 _Ref31620379 _Ref39474413 _Ref25931980 _Ref31550214 _Toc36985781 _Toc38694096 _Toc36985782 _Toc38694097 _Toc36985783 _Toc38694098 _Ref26510034 _Hlt31621392 _Hlt31621402 _Ref30048273 _Hlt31622047 _Hlt31623072 _Toc36985784 _Toc38694099 _Toc36985785 _Toc38694100 _Ref27095554 _Ref39761482 _Ref31623195 _Ref27097136 _Toc36985786 _Toc38694101 _Ref31622532 _Ref31622502 _Hlt31622511 _Toc36985787 _Toc38694102 _Toc36985788 _Toc38694103 _Ref27097303 _Ref27042598 _Ref39761269 _Ref25315493 _Hlt31622763 _Ref27042471 _Hlt27042589 _Ref26492231 _Hlt31623688 _Ref39761691 _Ref25415210 _Toc36985789 _Toc38694104 _Ref26626890 _Ref31623436 _Ref26537265 _Toc36985790 _Toc38694105 _Toc36985791 _Toc38694106 _Toc36985792 _Toc38694107 _Toc36985793 _Toc38694108 _Ref39566579 _Ref26709310 _Ref27095953 _Ref25314429 _Ref26535870 _Hlt31621777 _Hlt31550297 _Hlt31550858 _Hlt31550954 _Hlt31551652 _Hlt31551953 _Hlt31552041 _Hlt31632395 _Hlt45443655 _Hlt45443656 _Hlt31621229 _Hlt31623619 _Hlt31550272 _Hlt31550322 _Hlt31623420 ..6/6/k0l0005BKKoPoPPPPPPPPPUWU\U\]cggiitt}~&&&̅̅e__?SSXXXX---Y(4_ffB4 %%334L9L99@@SH9IIbbSpSp{{{|J~34ه!!!55ƴƴw'YYUVLLs\^xx[[q8  bF(AgnLΙ*(XWXWl+ @ @@@@ @!"$#@&%@'()*,+@/-@.@0123456789:;<=>?@ABCDEF@JG@H@I@KL@MNORP@Q@STUV[WXYZ\]^_`abcdefghijklnm@oqp@rstuvwxyz{|~}@@@@@@@@@@@@@@@@@@@@@@@@ ..6/7/00005B@K@KPPPPPPPPPPUWU\V\]cggiitt}~&&'Ș@ ooXXXY--.)5`fC5,,8 &&334L9M99 AATH:IIbbpp{{{|K~55E&&  !!" x(VWrr]_  r9  cF(Agn{Y(vWvW+v n(v !&v o(v n(O]!&O]T!&v sv sO]LsO]/(v T/(v  v  v O v N v DL!v L!v !v ī!v l(v ,(v $&v L%v Qv  v 'v |W&v <'O]\Y%O]'v 6'v Uv $v Yv G v Lv Z v l[v Av 1v v v <v |v (v 4(v t(v (v (v 4(v t(v (v (v 4(v t(v Nv Nv OO]TOO]OO]OO]Pv TPv Pv Pv Qv TQv Qv Qv Rv TRv RO]RO]Sv TSv Sv Sv Tv TTv Tv Tv Uv TUv Uv Uv Vv TVv Vv Vv Wv TWv Wv Wv Xv TXw Xw XO]YO]TYw Yw Yw Zw TZw Z w Z w [ w T[ w [ w [w \w T\w \w \w ]w T]w ]w ]w ^w T^w ^w ^w _w T_w _w _w `w T` w `!w `"w a#w Ta$w a%w a&w b'w Tb(w b)w b*w c+w d/,w /-w /.w $//w d/0w /1w /2w $/3w d/4w /5w /6w $/7w d/M]|M]M]M]<M]|M]M]<M]|M]M]M]<M]|M]M]M]<M]|M]M]M]<O]|O]M]M]<M]|M]M]M]<M]|M]M]<M]|M]M]M]<M]|M]M]M]<M]|N]N]N]<N]|N]N]N]< N]| N]  N]  N]N]|/?N]/@N]/AN]<0O]|0O]0O]<1O]|1O]1GN]1HN]<2IN]|2JN]2KN]2LN]<3MN]|3NN]3ON]3PN]<4QN]|4RN]4O]4O]<5UN]|5VN]5O]5O]<6O]|6ZN]6[N]<7\N]|7]N]7^N]7_N]|8`N]8O]8O]<9cN]|9dN]9O]9O]<:O]|:O]:O]:O]<;kN]|;O];O]<<nN]|<oN]<pN]<qN]<=rN]=sN]<>tN]|>uN]>vN]>wN]O]|s?O]s@O]sAO]3>3336r7r7778 9 9<<==>AAAAqBqB-D-DRDRDDD/E/EvEvEoHoHKKKLKLQM!O!OQQRSS#T#TTTVVrVrVWWg\]]A`A`J```aaajajab)bbbggiiiOmOmTm`m"p"pwwAyAyOyOyyyE~E~ZZހހjj66rrŘŘJJQn[[ƹƹ ,,BBHH""[[}}     ))))a)a) - -4U8U8::jDjD}ELFLFsGsGHHPIPI&K2Q2QQ_U_UjUjUsUsU|U|UUUUUUU X XYYQ_Q_d_vbuqqqmmxxѕѕJJ<<FXX{{٪٪ررijijγ΁׳׳  ))66>>HHRR]]44::^^6vvUURR99::00}}^^       A A . . e e   99o!!##((+      !"#$%&'()*+,-./0123456789:;<>=?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:<;=>?A@BCDEFGHIJLKMPNOQRSTUVWXYZ[\]^_`abcdefghijklmnopqrsutvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^`_abcdefghijklmnopqrstuvwxyz{|}~' ' O O **vv77gg%%_)_)p)p)..KKLLLLLLtMtMNNoNoNPPPPP\\]hhjjyyyy``^^hh kkc &&__ii  HHJJKKXwXwex&&""00BBUUͱͱdd88``ssQQRRbbvv&&hh " """t"t" % %%%#&#&&&&&&&&'' ' '''''-'-'5'5'?'?'G'G'N'N'V'V'g'g'q'q'''''''*)*)))**** +++C,,,----.////00D3D3336x7x777899<<== ?AAAAwBwB3D3DXDXDDD5E5E|E|EsHsH&K&KRLRLaM,O,ORRRSS.T.TTT#V#VvVvVWWs\]]H`Q`Q```a a aqaqa$b1bbbggiiiSm_mpmpm(p(pwwGyGyUyUyyyM~M~bbqq==  xx˘˘PUU~aa  ιι33 %%  LL  OO11cc!!     -)-)e)e)--4]8]8::pDpDE\F\FyGyGHHWIWI4K6Q6QQhUhUqUqUzUzUUUUUUUUUXXYYY_Y_s_byqqqvvווYYDNNttܱܱ³³̳̳ճճ޳޳  ##44<<FFPPWWcc;;EE&&ffE]]\\AA@@44dd!!     " N N 6 6 m m ??~""##((+  !"#$%&'()*+,-./0123456789:;<>=?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:<;=>?A@BCDEFGHIJLKMOPNQRSTUVWXYZ[\]^_`abcdefghijklmnopqrsutvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^`_abcdefghijklmnopqrstuvwxyz{|}~ :*urn:schemas-microsoft-com:office:smarttagsStreet9o*urn:schemas-microsoft-com:office:smarttagsStateB *urn:schemas-microsoft-com:office:smarttagscountry-region={*urn:schemas-microsoft-com:office:smarttags PlaceNameV*urn:schemas-microsoft-com:office:smarttagsplacehttp://www.5iantlavalamp.com/8e*urn:schemas-microsoft-com:office:smarttagsCity=y*urn:schemas-microsoft-com:office:smarttags PlaceType8*urn:schemas-microsoft-com:office:smarttagsdate;*urn:schemas-microsoft-com:office:smarttagsaddress8a*urn:schemas-microsoft-com:office:smarttagstime  01101112181984199519982020002001200220032122282933044556789DayHourMinuteMonthYearoooooeooaaaooeeoooeeaoooooooooaoooooooooooooee oooe ooe o  oee e  oeeoo oe eooe eeoaaeoeaaoe{{yeooooooeeeooo oe oe oeoooeooooeeeoeoeo ooeo eoeoooeooooe oeoeoeeeooeoooooo oo ooooo ooo'(|\\LaUagg"h+hii…ƅȆІ؆܆fgX\hl+/y09dheiPT_cJN37W[&&''((//S;\;D DKKQQRREVFVVVYYff {{ŁOSy}¨+4,0enNU !""6#>#$$%%%%((((Q)V),!,g../ /1090!9)9::;;?!?8ABAAAHHHH MMM?MOO}QQQQRR[RRRSS=SESSSSSSS{UUUUUUUUVV!W)WWWWWWWXX=YBYNYRYZZZZl[q[}[[\\]]&]*]^^4_?_``aaaabb^cgccc"d+dHdQdddfggghhhhiij#jmmaofopsysst tt0t6tJtQtCuLuuuuuxxzzA{F{R{V{]{d{{{G|P|||||`}g}i}o}"*~(-‚Ȃق=>TWϑ֑hmy}~ ]_ST?H z{}&noqwjkIP+1hikqȯί Y[׶ض+-8C,1=AEK#(mtpwLQ]ax~}&*x"-9;x}OTy:DFKUZku _ikpejNXZ_ /4124:").Ycrzi"&      * .   Xy{|~?prsux*,125[chvxEInov#}###P%V%%%g(r(((:)?).+7+-#-g-l-----..//1 12223)424~44777777`8g8k8v8889:::::::::<<6=;=k=q===========??????AA{CCEEhFmFyF}FFGGGG GGGGGHH#H.H:H\?\C\^G^I^J^L^O^____aabbbbbbb c$c%c'c*cdde ekkooooqqqqqqqqqq|rruuuuuuPyXyeyyyyyyyy{{4=$&)*135JMPRUtuwxz~ۓݓ ™ęǙș̙AJՠڠin"'2>@gq( JLOPYlդݤFMR]ik5:ӨԨ֨רۨ}@EQU]bodj|ìȬά֬ (Vծ׮ܮݮmrڸ#)1źͺ¾ʾvi !#$&:?KO+28@VZ |s{`gkvLORSW]a?Euc/7msDKOZfhFJ!(,7CE)/!%0<>t!%,4 dhow '-]^a~3 7 I y z | }                  BDEFIs-MNPQUPX/0235x.55:FJgl 2 3 5 6 8 >*E*W*********++V\ _fr6z6OOppttiukuxxy2yyyyzL˂YouGNTXw{OJJLL/a9al&l ])ITV"^"i""##(X),!,V0Y01w2}>>?ƚ` UVIP.}âܢ.q.mF~ʬڬYc­+k!\_df{EMεֵ۶߷.389<insx}/Ҿ"B13=MESX!"%nz"GK_k!)%r{H=4!}H!u x  ACz{-qr}{}6 7 b g    Q!((((779+:::::<=??CCsE|EEnFoFtFuFxFF GRGWGiHnHHHH-I2IIIIJK%KKKuTwTVWWWXX\;\^H^abb"cf3fooqqqqttuavMy5z_zuzGJz|#%1lnvЙszOlGLMPW;LeŴv:e;<?O)AFGJPUY_ag,b7BO: ^TK 3           CD \/6]<ABE9 >***++::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333y5yzg΂Y%b (y(a:=FGOJ:MX\_a Q R  CD Z69 : >******++++::O?o?p??OA}A^^^ _ccffggnnQxx{{L{Ԉ91P ,/uΙ*Y0OgJxNlEdʪЭHf6UӶ'Xbu/iXy?p*9:??FGHHHHIIIIXWvWWWXX\8\^G^bbb cooqquueyy$Ut~ۓ™JlӨ}oVծvi Luct-]bI y     Bs-M/ 2 W*****+++ITSAdminITSAdminITSAdminITSAdminITSAdminITSAdminITSAdminITSAdmin Derek Ricci u02oiHmO*dy#NM8BLXV6uf*=?;y_uXA0Sʳc60{glhk V6G f>zq v n | }? JT= JkR8M A S4nuRdjP|I t~rc0Fav. ;0+~7ZO(xG6V~2E8@&=By%o6`Km08E?\"=Ymr~ |uu 8{!LY"  o#ކvI#H¬MF$Cr%E o &Z#'.'P^5)y"Y*yg+z9Ey-&\]J.j -0>iS?2aw4v3)4Ҏp5(/9ʨ4$:rFy:;L[<~vN)F?p0n=A▄HB"C@F)wEC2h~DXFY{DFnDj: EtoE] M(?IgJ̅^EIKykNĦ#EP`PQJO^{RyUTo UdzR0UI?@3U>>Up0n6X>buYJQ1Yfxq\5s\]Zp3T+^d_y`.sr/Ncy eT %|e,J f UPh6\1AhT&Ujp0n)lir<~f{u!vxX^vsfBwޖ\_x=0lyFr{P(8| | j؞D5}^<`iUhh^h`CJo(.0^`0o(()hh^h`o(Yhh^h`o( hh^h`OJQJo(808^8`0o(()^`.pLp^p`L.@ @ ^@ `.^`.L^`L.^`.^`.PLP^P`L.M^`Mo(()^`.pLp^p`L.@ @ ^@ `.^`.L^`L.^`.^`.PLP^P`L.]hh^h`CJo(.:hh^h`o( hh^h`OJQJo(0^`0o(()Khh^h`o(>hh^h`CJo(.Nhh^h`o(0^`05o(()'hh^h`o(/hh^h`CJo(.Bhh^h`CJo(.Chh^h`CJo(.M^`Mo(()^`.pLp^p`L.@ @ ^@ `.^`.L^`L.^`.^`.PLP^P`L.0^`0o(() ^`hH.  L ^ `LhH.   ^ `hH. xx^x`hH. HLH^H`LhH. ^`hH. ^`hH. L^`LhH.\hh^h`o(\hh^h`CJo(.>^`>o(()/hh^h`o(0^`056o(hH() ^`hH.  L ^ `LhH.   ^ `hH. xx^x`hH. HLH^H`LhH. ^`hH. ^`hH. L^`LhH.%hh^h`CJo(.`hh^h`CJo(.0^`0o(()^`. L ^ `L.  ^ `.xx^x`.HLH^H`L.^`.^`.L^`L.Nhh^h`6o(&hh^h`o(Khh^h`o(Jhh^h`CJo(.Vhh^h`o(Uhh^h`o(hh^h`CJo(. hh^h`OJQJo(Ihh^h`CJo(.Zhh^h`o([hh^h`CJo(.Ohh^h`CJo(.Whh^h`CJo(.88^8`o(()z^`zo(()hh^h`CJo(8^8`() hh^h`6CJo(3hh^h`o(Ohh^h`CJo(.hM^`Mo(()^`.pLp^p`L.@ @ ^@ `.^`.L^`L.^`.^`.PLP^P`L.0^`0o(()88^8`o(()Ehh^h`CJo(. hh^h`OJQJo( hh^h`OJQJo(Lhh^h`CJo(.hh^h`6o(M^`Mo(()^`.pLp^p`L.@ @ ^@ `.^`.L^`L.^`.^`.PLP^P`L.Zhh^h`o(&hh^h`CJo(.>hh^h`o(hh^h`CJo(.Ihh^h`o(hh^h`o( 0 ^ `0o(()^`.pLp^p`L.@ @ ^@ `.^`.L^`L.^`.^`.PLP^P`L.Ahh^h`CJo(.Thh^h`o(P^`PB*CJOJQJo(Hhh^h`o(Hhh^h`CJo(.8^8`()[hh^h`CJo(.Whh^h`o(Lhh^h`o(808^8`0o(() ^`hH. pLp^p`LhH. @ @ ^@ `hH. ^`hH. L^`LhH. ^`hH. ^`hH. PLP^P`LhH.Lhh^h`CJo(.hh^h`CJo(.[hh^h`o(Dhh^h`o( hh^h`OJQJo(Ehh^h`CJo(.0^`0o(()^`.pLp^p`L.@ @ ^@ `.^`.L^`L.^`.^`.PLP^P`L.0^`0o(() ^`hH.  L ^ `LhH.   ^ `hH. xx^x`hH. HLH^H`LhH. ^`hH. ^`hH. L^`LhH. hh^h`OJQJo(:hh^h`CJo(.WW^W`o(.0^`05o(()Yhh^h`o(P^`PB*CJOJQJo(Shh^h`CJo(.0^`0o(()Hhh^h`o(M^`Mo(()^`.pLp^p`L.@ @ ^@ `.^`.L^`L.^`.^`.PLP^P`L.Ihh^h`CJo(.Yhh^h`CJo(.@hh^h`o(_hh^h`CJo(.0^`0o(() hh^h`OJQJo(Lhh^h`6o(Ghh^h`o( hh^h`OJQJo(Jhh^h`CJo(.hh^h`o(P^`PB*CJOJQJo(0^`0o(()^`.pLp^p`L.@ @ ^@ `.^`.L^`L.^`.^`.PLP^P`L.^hh^h`CJo(.Rhh^h`CJo(.;hh^h`o(88^8`(.)Mhh^h`CJo(.0^`0o(()^`. L ^ `L.  ^ `.xx^x`.HLH^H`L.^`.^`.L^`L.808^8`0o(()^`OJPJQJ^Jo(-pLp^p`L.@ @ ^@ `.^`.L^`L.^`.^`.PLP^P`L. hh^h`OJQJo(Vhh^h`o(hh^h`o(Mhh^h`o(xCr%J._ZHmEy-G I#^v%|e7ZO|w4 U=A eY{DD`iOgJp52E-0Ah;0u]A0Sav{gl[<?@3U(?I~mJ(/9D5}uQ1Y{! o#3)4S?2q UTc'q\uu g+MF$02=YEBLXy:HBEG6!vs\f{u8MjPn kN%o\_x6XwEC3T+^I  UPhtoE? CZ#'BIK5)Nc8|?;"Y*d{RUj>U)F? fC "  SbuYk R0UfBwfBw R0U y`oo &_ur{#N04$:)lir#EPlyR= nDPQuu        h%                +         fG&        D        x        Ƙ}        x         Ƙ}        $P        ^        Ƙ}        Oj?WlR       { Al n^pI - Q 6,- LuQ#Av~$N2{A S!!O#W#o$&(T)5)t)V*&5+u+w/V 0"12 3 H3/4lk5k6 8@)8:i;s;P>@vzAEkmC(+~mx 2R(2/i/><|z"e(*,xFN5F kN5Z' VWHu}9=&|-yp!#k[[ErSM>LG7V7;\I:mEv,NZd^@H:\˅̅&2>JKUZ`fmnx}.m&pc*++++!D@V+@UnknownGz Times New Roman5Symbol3& z Arial5& zaTahoma"1hKwfePE_ePE_Q#Z4d 3QH?> Erika Chamberlain u                           ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 5 6 7 8 9 : ; < = > ? @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z [ \ ] ^ _ ` a b c d e f g h i j k l m n o p q r s t rdDocWord.Document.89q