Hello, and welcome to the webinar entitled Addressing Drug Impaired Driving, an Evidence-Based Approach. I'm Dr. Angela Michelide, Vice President of Programs and Education at the American College of Preventive Medicine. The American College of Preventive Medicine is a professional medical society of more than 2,000 physicians dedicated to improving the health and quality of life of individuals, families, communities, and populations through disease prevention and health promotion. Our members are licensed medical doctors or doctors of osteopathy who possess expertise in a broad range of healthcare skills, including biostatistics, epidemiology, management of healthcare organizations, research, and the practice of prevention in clinical settings. Our physicians work in a variety of settings, including hospitals, government at the local, state, and national levels, the military, laboratories, businesses, and academia. I have just a few housekeeping items to share with you today. The audience has been placed in listen-only mode, and we will be posting a recording of this webinar on the ACPM.org website in just a few days in the event that you'd like to share it with friends and colleagues. We encourage you to type questions throughout the presentation in the Q&A box, and we want to know what you think and how we can do better. So please fill out the survey that you will receive after the webinar so that we can improve our offerings. I'm delighted to introduce our two speakers today. Sarah Wendel is a PhD student in the Department of Epidemiology at McGill University, where she is affiliated with McGill's Institute for Health and Social Policy and the Public Policy and Population Health Observatory. Sarah is supported by doctoral training awards from the Canadian Institutes of Health Research and the Quebec Research Fund's help. Her research interests include policy evaluation, substance use, education, and maternal and child health. Sarah received a master's degree in public health from the University of Alabama at Birmingham in 2009. Her recent work as a clinical research associate at the Lady Davis Institute of the Jewish General Hospital in Montreal included collaboration on clinical trials and knowledge syntheses in a variety of research areas including smoking cessation, e-cigarettes, and cannabis legislation. Our second speaker is Amir Razagisan, and he is a second year Master of Science student at McGill University, where he received a Bachelor of Science with specialization in pharmacology and therapeutics. He completed his training in epidemiology and biostatistics at the Lady Davis Institute Center for Clinical Epidemiology. His primary research interests are in public health harm reduction and cardiovascular health promotion. He is actively involved in education in his program and academic initiatives to promote research trainees' involvement in knowledge translation. The American Journal of Preventive Medicine has selected his drug-impaired driving prevention article for three months of free online access. This program is part of the journal's continuing medical education program for physicians who read AGPN. So without further ado, let's get started with Sarah. Good afternoon, everyone. Just one second here. Okay, there we go. All right, thanks so much for having me. My presentation was specifically concerned recreational cannabis legalization and impaired driving. I'm going to start with just a brief overview of cannabis in a global setting. We'll review a little previous research in this area and then discuss the findings of a new research study that was funded by the Canadian Institutes for Health Research. I'll then briefly discuss some prevention and deterrent strategies that are currently being used, as well as suggest a possible role for healthcare professionals in promoting safer cannabis use. Afterwards, we'll hear from Amir, who will discuss the paper that was mentioned concerning drug-impaired driving prevention more generally. Cannabis is the most common psychoactive substance used worldwide. There's an estimated 180 million users. Over the past 25 years, we've really seen a global shift towards cannabis decriminalization and legalization, with more than 30 countries worldwide having decriminalized cannabis. Two countries in particular, Uruguay in 2013 and Canada in 2018, have legalized cannabis nationally. More than half of all U.S. states have legalized medical cannabis, and 15 states and D.C. have legalized recreational cannabis as well. With this increasing trend towards legalization worldwide, concerns have been raised regarding cannabis and impaired driving. Cannabis is known to impair psychomotor skills and cognitive functions that are needed for driving, effects including slower reaction time, increased lane weaving. However, the relative contribution of cannabis to these collisions is really unclear, particularly as cannabis is often consumed with other legal and illegal substances. We know that effects can vary by individual. However, guidelines recommend a minimum wait time of at least six hours prior to driving. And of importance, you know, in 2019, 13% of Canadian cannabis users with a driver's license reported having driven within two hours of consumption. So in terms of what literature is already out there, there are observational studies that have examined outcomes such as the proportion of drivers with detectable levels of cannabis before and after legalization, or changes in self-reported driving under the influence of cannabis. However, these types of studies have some significant limitations. Particularly, you know, for starters, detectable levels of THC don't necessarily correspond to impairment, and we'll talk about that a little bit more later. This measure is also potentially confounded by increases in enforcement testing and reporting that occur because of legalization. And as well, we know that individuals may be more likely to report their cannabis use and their associated driving when cannabis use is legal compared to illegal. So really the best evidence out there, there's a number of quasi-experimental studies which don't rely on biochemical measures or self-reported measures. And these are really the best available evidence to date. These have used, for the most part, fatal motor vehicle collisions as the most widely used outcome. This is something that is systematically collected in North America, and the data collection importantly has not changed as a result of cannabis legalization. So a couple studies have looked at data from Colorado and Washington State. Both of these studies use differences and differences approaches with data through 2017. They found a potential increase in fatal collisions and fatalities from collisions following recreational cannabis legalization. One study also found a potential increase in collisions after the opening of commercial dispensaries specifically. However, the confidence intervals from these studies were largely inconclusive, even though they were suggestive of increased risk following legalization and opening of dispensaries. An additional study looked at data from Oregon as well. This one used a slightly different methodology. It was an interrupted time series with data through 2016. This one also found an increase in fatalities only in the first year following legalization. And then they found a reduction in trend per month after the first year. And lastly, another very recent study also looked at Alaska as well. This was again a differences and differences approach, used data through 2018, and did as well find an increase in fatalities after the opening of cannabis dispensaries. So overall, the previous literature is suggestive of a potential increase in collisions following the legalization or the opening of dispensaries. However, many of these studies had findings that were inconclusive. And one study, that one study in particular, suggested that the potential increase in collisions may be temporary. These studies were also limited to the US jurisdictions that legalized the earliest. So we really set out to examine this question with all of the available data. And the findings of our study were just published this year. So our objective was to examine the association between recreational cannabis legalization and fatal collisions in the United States. We used data from the US Fatality Analysis Reporting System. This system includes motor vehicle collisions that occurred on public roadways and resulted in at least one fatality within 30 days. The study period we selected was 2007 to 2018. We picked 2007 to provide at least five years of data prior to legalization for all jurisdictions. And 2018 was the last year that data was available. We looked at both legalization and opening of dispensaries, as well as collisions and associated deaths. And these are just a slight distinction. You know, each collision could result in more than one fatality. So you can see recreational cannabis was first legalized in the United States in 2012 in Washington State and Colorado. As of the end of 2020, there were 16 US jurisdictions, 15 states and the DC, that had passed legislation legalizing recreational cannabis. You can see South Dakota is the latest of these. Its legislation is actually not in effect yet. It's predicted for July 1st of this year. So we extracted data for the jurisdictions which had legalized legislation in effect prior to 2019. Therefore, pre and post legalization data were available for 11 of these 16 jurisdictions for analysis. So we estimated the incidence rate ratio and 95% confidence interval using Poisson regression in jurisdiction specific analyses for both collisions and associated deaths and by both legalization and the opening of dispensaries, as I mentioned. We adjusted the rate ratios for calendar time to account for temporal trends within jurisdictions. And then we took the jurisdiction specific estimates and meta-analyzed them using Darcymonian and layered random effects models with inverse variance weighting. So just very briefly, in terms of the raw numbers, there were nearly 74,000 fatal collisions, which occurred between 2007 and 2018 in the 11 jurisdictions that were included in our analysis. And of these collisions, approximately 17,000 occurred after legalization. These collisions resulted in more than 80,000 deaths, nearly 19,000 of which occurred during periods when recreational cannabis was legalized. So overall, we found that there was a 15% increase in fatal collisions following legalization with a 95% confidence interval of 6% to 26%. So while there was some variability between jurisdictions, with the exception of Michigan, all had point estimates favoring an increase in collisions following legalization. Michigan was the last jurisdiction in this analysis to have legalized legislation in effect. They actually only had a single month of data available post legalization. Among the other jurisdictions, the point estimates ranged from 1.01 to 1.92. So as expected, when we look at deaths from fatal motor vehicle collisions, this is a very similar 16% increase following legalization with a 95% confidence interval of 6% to 27%. So in addition, in a secondary analysis, we examined the association between the opening of the recreational cannabis dispensaries and fatal collisions. So in most jurisdictions, legalization and the opening of dispensaries isn't simultaneous. In many cases, dispensaries didn't open until several years following legalization. Excuse me. Among jurisdictions with legalized recreational cannabis, the District of Columbia and Vermont actually don't permit the sale of recreational cannabis at present. Vermont's trying to change this. I haven't heard anything about D.C. trying to change that at present. So basically, they permit home growing for personal consumption, but you can't buy recreational cannabis. So among the remaining jurisdictions, only seven had recreational cannabis dispensaries open prior to 2019 and therefore had FARS data available for analysis. We found that the opening of recreational cannabis dispensaries was associated with an 18% increase in fatal collisions with a 95% confidence interval of 6% to 32%. And as expected, a corresponding increase of 18% in deaths associated with the opening of dispensaries. So from the previous literature, we saw that one study that suggested that the increase in fatal collisions might be temporary, for example, due to a celebratory effect or a large increase in inexperienced users in the first year after legalization. Therefore, in another secondary analysis, we tried to compare the first versus the subsequent years of legalization to assess whether or not the increases in rates of fatal collisions and associated deaths is temporary. So our results were inconclusive when we compared the first versus subsequent years. However, the point estimates are suggestive of actually potentially a lower incidence of fatal collisions in the first year. We think that this could potentially be due to really essentially limited cannabis supply in the first year compared to subsequent years. It takes a lot of time sometimes to get everything in effect after the legislation goes into effect. So just a summary of our findings. We found that legalization was associated with 15% increase in fatal collisions and a 16% increase in associated deaths. This is largely consistent with the previous studies that we saw. Our analysis included seven additional jurisdictions, as well as data through the latest available year, 2018. Our study has several limitations, the first of which is simply its ecological nature. So there's a number of factors which vary between jurisdictions and could potentially influence fatal collisions. That was why we decided actually to look within jurisdictions and then meta-analyze rather than comparing directly across jurisdictions. We also adjusted for calendar time to try and control for any potential temporal trends. We also looked at trends in the incidence of collisions prior to legalization and found that these were roughly log-linear. So in the last few minutes, I'll just share with you a few findings from a narrative review that we published in the Canadian Medical Association. This includes some detection and deterrent strategies that are currently being used, and we suggest a potential role for health care professionals in preventing cannabis impaired driving. So psychomotor assessments in the form of standardized field sobriety tests and clinical trials. We also looked at the impact of psychomotor assessments on health care professionals in preventing cannabis impaired driving. So psychomotor assessments in the form of standardized field sobriety tests are very common in North America. These can be performed from side to screen for drug impairment. They were originally developed actually for assessing impairment related to alcohol use. So there's some question about their validity for detecting impairment due to cannabis. Some areas may also use officers with specialized training, such as drug recognition experts. These are individuals who are trained to conduct more complex assessments of impairment. However, the accuracy with which they can detect impairment and particularly attribute it to a particular class of drug has also been questioned. Biochemical measures are presently being heavily studied for cannabis. I think everyone wishes that we could, it would be as simple as alcohol. Unfortunately, it's not. There's really substantial uncertainty concerning the validity of biochemical measures for detecting impairment. THC is fat soluble, so it passes quickly from the blood into other organs. And relative impairment at a particular THC concentration can vary by individual. In particular, chronic daily users could continue to have detectable levels of THC after weeks or even months of abstinence. That being said, despite limitations, per se limits may be useful in deterring impaired driving. There's a couple of interesting studies which suggest that alcohol limit laws, which were introduced in the 1980s and 1990s, may have reduced alcohol related collisions by 14 to 15% simply due to public awareness. And so of the 11 jurisdictions that we included in our analysis, actually only four had cannabis specific driving limits. Canada has taken a bit of a different approach. They passed Bill C-46 in 2018. This introduced national THC limit laws for cannabis. So in other deterrent strategies, they're commonly used. Zero tolerance policies for young drivers are pretty ubiquitous. Sobriety checks roadside, as well as administrative sanctions following impaired driving conviction. So in terms of a potential role for healthcare professionals, there's some evidence out there that brief counselling interventions in primary care settings can be effective for reducing problematic alcohol use and other risky behaviours. Therefore, it's possible that these could be adapted to counselling concerning cannabis use and driving as well. The Lower Risk Cannabis Use Guidelines were published by Fisher et al. in 2017. And these provide really good evidence-based recommendations for safer cannabis use, some of which could be imparted to patients. So just a few selected recommendations include choosing lower potency products, limiting to occasional use, and choosing non-smoking methods of consumption. Related to driving in particular, the guidelines suggest waiting at least six hours after cannabis use before driving. We know that peak THC concentrations typically occur within 30 minutes of consumption and taper off after two to four hours. However, impairment may persist for three to six hours or longer. In addition, the guidelines highlight avoiding combining cannabis and alcohol as this results in greater impairment than either substance used alone. And just a couple of additional notes. There's no evidence to suggest that driving risks are different for therapeutic versus non-therapeutic users of cannabis. Therefore, the same information could be imparted to both groups. It's also important to be aware of any discretionary or mandatory reporting requirements regarding fitness to drive in your jurisdiction. Although knowledge gaps certainly remain concerning determining fitness to drive, particularly for long-term regular users of cannabis. So in conclusion, we found that there's the potential for an important increase in fatal collisions following legalization. Our study identified a 15% increase. And evidence is needed concerning optimal prevention strategies. Certainly, there's uncertainty concerning psychomotor assessments and biochemical measures. However, public awareness of cannabis specific per se limits and other deterrent strategies may be effective. And in terms of a potential role for healthcare professionals, brief counseling, including evidence-based recommendations for safer use may be effective for reducing risky cannabis behaviors. So lastly, I'd just like to thank all of our collaborators for their contributions to this project. Particular thanks to Chris Fidiot and Paul Laminier for their contributions to the data analysis. As well as to acknowledge the funding from the CIHR for this study, as well as my individual support from the CIHR and FRQS. So thank you again. So, if you missed the brief introduction by Dr. McIlyde at the beginning of this talk, I'd like to just quickly mention that I'm a master's trainee at the Center for Outcomes Research and Evaluation at the Research Institute of the McGill University Health Center. And today I'll be presenting prevention research based on a systematic review we did for the prevention of drug-impaired driving. And I'd like to also apologize for this poor lighting. It looks like the sun just came out behind me. So this project is really important because motor vehicle collisions are a leading cause of death in North America, resulting in more than 40,000 deaths and almost 2.5 million emergency department visits per year. At present, as Sarah has already mentioned, non-medicinal cannabis has been legalized in Canada, Uruguay, as well as 15 U.S. states and the District of Columbia. And although alcohol has historically really been at the focus of impaired driving, a lot of recent literature has suggested that the legalization of cannabis may be associated with an increased rate of fatal motor vehicle collisions. And as Sarah's research just presented, the legalization of cannabis was associated with almost a 15% increase in the instance rate of fatal motor vehicle collisions. And I'd like to add that this literature is quite consistent with other research in the field. For example, in a case control study by Lean colleagues at Columbia University, they found that individuals who drove under the influence of cannabis had 1.8 times the odds of being involved in a fatal motor vehicle collision compared to un-intoxicated drivers. And this evidence is, again, consistent with what some researchers have found in Uruguay after the legalization of cannabis. For example, Nazif Munoz and colleagues at her institution conducted an interrupted time series analysis and found that after cannabis was legalized for recreational use in Uruguay, there's a 52% increase in the rate of light motor vehicle fatalities. And this evidence is really quite consistent with the effect that illicit drugs have on the psychomotor skills required for driving. And as you can see in this table from the World Health Organization, the negative effects of cannabis and other psychoactive drugs have really been well-established, or pardon me, their effect on the sensory and motor functions required for driving have been really well-established. And if you look at the top row of this table here, you can see that cannabis affects almost every single impairment domain, including drowsiness, cognitive functioning, motor functioning, mood, lateral vehicle control, time estimation, imbalance. And although the detrimental effects of drug-impaired driving have been really well-established in the literature, a lot of epidemiological data still shows that it's really highly prevalent, especially among youth. For example, if we look at this data from the National Survey on Drug Use and Health from 2019, we can see that almost 13.7 million people ages 16 or older drove under the influence of drugs, including cannabis, in the past year. And I'd like to highlight the data shown here are presented in thousands. Also, if we look at the subgroup of individuals ages between 16 and 25, we can see that drug-impaired driving is actually a lot more prevalent than driving under the influence of alcohol. For example, almost 4.5 million people drove under the influence of drugs, whereas this figure was only about 3.1 million for people who drove under the influence of alcohol. Because of this, we decided to conduct a systematic review to evaluate preventive interventions for drug-impaired driving based on measures of drug-driving knowledge, attitudes, behaviours, and traffic-related outcomes among adults and youth. But before providing a detailed description of the interventions we found, as well as their effects, I'd like to quickly touch on the systematic review and how we conducted it. Very briefly, the first thing we did was we developed a systematic search strategy with academic librarians and searched seven databases, which you can see here at the top of this figure. And you can see these databases included some standard ones, such as MBASE and Medline, as well as a host of others that are more specific towards transportation research. And overall, we identified over 20,000 records, of which we screened 11,000 after the removal of duplicate citations. And after this process, we identified 28 studies which had evidence relevant to the prevention of drug-impaired driving. Because we identified so many interventions among these 28 studies, we had to develop an analytical framework to help guide the synthesis of our findings. And the analytical framework is shown in this figure, and you can see the conceptual approach that we used. And what this figure illustrates is the three types of interventions, which were educational, deterrent, and rehabilitative, and the intermediate outcomes that they're associated with. And these include the improvement of knowledge about drug-impaired driving risks, the formation of negative attitudes and perceived risk of punishment when drug-impaired driving, which lead to our general more hard outcomes, such as traffic violations and vehicle collisions, and fatal and non-fatal motor vehicle injuries. And in order to evaluate the risk of bias of the studies we identified, we used two well-established tools, which are recommended by the Cochrane Collaboration. And these two tools are the Cochrane Risk of Bias Tool for Randomized Trials, or ROB2, and the Risk of Bias in Non-Randomized Studies of Interventions, also known as ROBINZI. And as you can see, these tools evaluate the risk of bias of studies based on seven potential domains. I won't name them all, but these include potential biases due to confounding and potential biases to selection of participants into the study, for example. Using the risk of bias information that we generated using ROB2 and ROBINZI, we evaluated the certainty of evidence associated with the intervention-specific effects that are reported. And again, we use the really well-established framework known as the Grading of Recommendations Assessment Development and Evaluation approach. And the way the framework works is quite simple. These are provided an initial certainty of evidence rating based on the design. For example, randomized control studies are given a score of four, which corresponds to high-quality evidence, whereas non-randomized studies of interventions are assigned a quality rating of two, which corresponds to low-quality evidence. Based on potential different domains, the quality of evidence can either be downgraded or upgraded. For example, we can downgrade the quality of evidence if we find that the results are inconsistent, if there's any presence of risk of bias, if there's publication bias, or if the results are indirect to our research question, or if the effect estimates are imprecise. Similarly, we can upgrade the evidence if we find that the effect estimates are large, if there's any confounding towards a null, or if there's any dose response gradients. And using the grade approach, final evidence ratings are provided for the effect estimates. And as you can see, these range from very low to high. And a quality rating of high corresponds to an effect estimate that we're confident is close to the true effect, whereas a very low quality rating or certainty rating corresponds to an effect estimate that is very uncertain and will often be far from the truth. Now for the remainder of the presentation, I'll be discussing the effective interventions that we identified. But before providing a detailed description of those interventions, I'd like to give a quick high-level overview of the studies that we included. And very briefly, these articles were published between 1987 and 2020. And most of the participants were adolescents and young adults ages 15 to 25. And as you can see, individuals between these ages accounted for almost 90% of all the participants. Lastly, the studies were published in various countries and different languages, and they utilized a wide range of different study designs. For example, we included several randomized control studies, cross-sectional, for example, randomized control trials, cross-sectional studies, and interrupted time series designs. Now I'll be focusing primarily on population-level interventions, which we found were to be quite effective. And these will be primarily of interest to the people here who are involved in government or public policy. Some of the highest quality evidence that we found in this category for population-level interventions came from a randomized control trial that was published at the University of Alberta, and it enrolled approximately 649 students. And what the authors of the study found was that traffic-related health warnings on cannabis packaging were significantly able to increase participants' belief that drug-impaired driving is a harmful activity by approximately 6%. And you can see this warning on the right, and it simply reads, the chance of motor vehicle accidents almost double while under the influence of this product. This led the authors to conclude that the use of health warnings such as this one are a cost-effective medium to communicate health information given their reach, frequency of exposure at the point of purchase, and during use. And overall, this is really quite accordant with literature in tobacco, which has shown that the warnings that we see on cigarette packagings are an effective way of communicating the harmful effects of smoking among smokers. As part of our review, we also found moderate quality evidence regarding the effect of roadside drug testing on individuals' intention to drive under the influence of drugs. For example, we came across theoretical evidence from a study which showed that high certainty apprehension scenarios, where police could randomly test for drugs nearby, were able to decrease cannabis users' odds of willing to drug drive by almost 80%. And while this evidence was only theoretical, as it used hypothetical scenarios, it was consistent with real-world data from Australia that showed that injection drug users who had been tested for drugs while driving more than once were 40% less likely to have drug drive in the last year compared to drug users who had not been tested for drugs in the past. In the review, we also came across evidence regarding the effect of media campaigns, and this evidence primarily came from the UK and Scotland, where two anti-drug driving advertisements were aired for approximately two months. And what the agencies which evaluated these programs found was that individuals who recognized the advertisements versus individuals who didn't generally perceived a greater likelihood of being stopped while driving impaired and also generally had a better understanding of drug-impaired driving laws in their respective jurisdictions. And overall, these marketing campaigns were found to be quite cost-effective as they widely increased knowledge and the perceived risk of punishment when it comes to drug-impaired driving. The remaining population-level interventions that we identified had to do with the criminalization of road traffic offenses in Spain, the withdrawal of licenses of drug offenders in select US states, and per se drug driving laws, which Sarah briefly mentioned in her presentation. And while the certainty of evidence associated with the effect estimates here are quite low, the effects are still worth mentioning. For example, in Spain, we found that the criminalization of road traffic offenses was associated with approximately a 7% decrease in the risk of traffic injury collisions. We also found that US states which withdrew the licenses of drug offenders had significantly less traffic fatalities per 10,000 miles compared to states which didn't implement these laws. And we found that per se drug driving laws in select US states generally had inconsistent and null findings across the identified studies, meaning that their effect is at the moment pretty unclear. Now I'd like to focus primarily on interventions which can be initiated, recommended, or provided by healthcare providers. Some of the highest quality evidence that we identified in this category came from a program called Safe Drivers Wanted, which was conducted in the US. And what the individuals who evaluated this program found was that it was able to reduce drug driving by almost 40% among participants of the program. And some of the key components of this program included reviewing risk-taking behaviors related to driving and helping parents and teens develop a contract for driving, which included potential negative consequences for breaking any of the agreements that they had. And the authors of the study actually provide a simple driving contract which can be accessed in their article. In regards to drug abuse prevention, we found that the certainty of evidence was generally very low for the effect estimates. However, we did find that a program called Life Skills Wanted, which taught students different cognitive behavioral skills to build self-esteem, resist peer pressure, and manage emotions and communicate effectively, were able to significantly reduce traffic offenses after six years by about 20%. The next intervention that we evaluated was motivational interviewing. And we found this to be one of the most effective strategies at the healthcare provider level. And what we found was two RCTs, which evaluated the effect of motivational interviewing, found that it was able to significantly decrease cannabis-impaired driving by between 13% to 50% among youth who regularly use cannabis. And one of these RCTs, which was actually by Fisher and colleagues, which Sarah mentioned earlier in her presentation, found that the effect of motivational interviewing actually persisted in the long term. After 12 months, they found that participants who were enrolled in the program reported 20% less drug-impaired driving with cannabis 12 months after initial enrollment. And motivational interviewing is defined here as a client-centered counseling style for eliciting behavior change by helping clients to explore and resolve ambivalence. Finally, the last provider-level interventions, which I would like to touch on, are substance abuse treatment programs and rehabilitation programs. And as you can see, the certainty of evidence associated with the effect estimates is quite low, but it's still important to mention the effects. For example, we saw that substance abuse treatment programs were able to reduce drug-impaired driving with cocaine and cannabis by between 10% to 26% after initial enrollment in the substance abuse treatment program. In terms of rehabilitation, we found that the evidence primarily came from Germany and Sweden, which somewhat limits their generalizability to the U.S., but we did find that a program called Drugs was able to cut recidivism by almost a third after three years. And this program consisted simply of four four-hour group sessions where previous offenders discussed the role of drugs in their lives to develop behavioral coping strategies to implement for approximately five weeks. Now, I'd like to provide a general discussion of our findings, and I'd like to touch on some of the unique challenges associated with drug-impaired driving in potential areas for future research. So one of the main challenges associated with the prevention of drug-impaired driving is the fact that a lot of previous research in the field has implicitly relied on drunk driving. And some of the main challenges that we see with drug-impaired driving, particularly with cannabis, is the fact that the perceptions of risk are usually substantially lower than for driving under the influence of alcohol. And research participants commonly report feeling more cautious when driving high, and they also report little fear of police involvement when driving impaired, likely due to the fact that a lot of the enforcement techniques haven't been correctly validated at the moment. As a result, for areas for future research, we believe that addressing social norms and perceived riskiness are likely to be important for prevention, especially for cannabis-impaired driving, as cannabis-impaired driving is currently strongly associated with having more friends who use the substance. In conclusion, we did find effective interventions for improving intermediate outcomes related to drug-impaired driving, such as knowledge and attitudes, but data were generally more limited concerning hard outcomes specific to drug-impaired driving, such as morbidity and mortality. Consequently, we believe that future research should focus on public education campaigns that emphasize social norms and perceived risks regarding drug driving, especially among youth. Finally, I'd like to take the last couple of moments of this presentation to thank everyone who was involved, and also the Canadian Institute of Health Research for funding this study. Thank you very much. Well, thank you very much, Sarah and Amir, for these excellent presentations. Those of you participating remotely, please begin to populate questions into the chat. Here's one. Is there any research addressing the impact of drug driving on road range? Yes, there is. On road range. Do we know at all if people are mellower and therefore are less likely to drive quickly or with aggressiveness? It's a really interesting question. I can't say that I have seen any literature particularly related to road rage. I don't know if Amir has seen anything, but I think people generally have a perception that they are more mellow when they're driving and that they're driving slower. Although I believe that the literature hasn't necessarily borne that out either. People tend to have a perception that they're driving safer when they're high on cannabis, which is not great. I actually completely agree with what Amir said about the need for changing social norms and the perceived riskiness around driving. I mean, we all know alcohol-impaired driving is really dangerous, but people need to be aware that driving high is not safe driving either. But yeah, unfortunately road rage in particular, no, I haven't seen anything. Could you speak to the epidemiology of cannabis use, both recreational and medicinal in the United States and Canada? How prevalent is it? So I don't have the numbers for Canada, sorry, for the US off the top of my head. Canada, there seems to be a bit of an increase after the legalization in 2018. By the end of 2019, I believe use in the past 30 days was around 16 or 17% overall, although that does vary across the country. And that's about a 2% increase, 2% absolute increase, I should say. I believe it was about 15% prior to legalization. And I think I've seen similar numbers for the US, somewhere around perhaps a third of individuals have used cannabis in the past year. So that's one in six Canadians. That's quite high. It is quite high, yes. Is there the sort of an analysis by age? For example- Yeah, so, yeah, use is always highest among the young adults, with a bit of a dip downward. Although I believe there's also an increase after legalization amongst the older age groups, which was a little bit unexpected. I wonder if a medicinal is consistent across the age groups, or maybe higher among older adults. Yeah, that's a good question. Yeah, maybe Amir can speak to this one. Yes, no, certainly, that's a really good question. But I'm not very sure. I imagine it might be a little bit evenly distributed. Certainly. That would make for an interesting research project for a member of your team. Here's another question. Given the difficulty of quick feedback testing, what are the most important messaging that can be conveyed regarding cannabis use and driving? Definitely. So what we found was, we found that the most important messaging definitely. So what we found was that, so for example, the paper that presented about the warnings we see on cannabis packaging, inherent limitation of that study was the fact that they evaluated individuals' perceptions immediately after they presented these warnings. So I think the message that they send out is really important, that it increases the risk of accidents, as well as potentially those that are fatal. So I think they should continue to send out that message, but potential future evaluations should see if it actually affects individuals' behaviors in the long term. That would be very interesting to see. And I would just add in terms of messaging, I mentioned there's all this uncertainty regarding the biochemical testing. And here in Canada, the police are actually permitted to use roadside saliva tests for cannabis. And I believe there's at least two devices that have been approved. But I mean, I think people are aware that that number doesn't necessarily correspond to impairment, but we're not articulating this well. What am I trying to say? I'm trying to say that police are trained to detect impairment. So whether they can attribute it to cannabis or something else, it doesn't really matter. If you are driving impaired, they can detect impairment per se. And that's outside of the whole THC testing, any of that. The per se limit for THC just means if you test above X amount, we will fine you. They don't have to prove impairment. So that's almost a completely separate thing. If you're driving impaired, that's what they're there to detect. If you have a per se limit, that might open you up to other fines or penalties regardless. If you were to develop educational programming for high school students, what would be three key messages that you'd like to impart within that educational programming? Certainly. So I can definitely tackle this one. So I think one of the really important things is addressing the fact that it is still harmful. Currently, a lot of youth think because it's perhaps a safer alternative to driving under the influence of alcohol, it means that it's completely safe. But it's important to essentially mention that while it may be safer, it still has inherent risks and can increase your chance of collisions. Second is the fact that right now, a lot of youth have no idea that police can use these safety measures use these saliva tests. And it would be really important to just highlight some of these newer methods that they're using because it's potentially very possible that they simply don't know that they can be caught. And that inherent perception of risk could be a really important tool towards preventing drug impaired driving. And I think those are the two main important ones. I'm not sure about the third one. If Sarah, you'd like to add anything. I think those two are definitely the big ones. And what happens when an individual drives from across state lines from one in which either recreational or medicinal cannabis is legal and goes to a state where neither is legal? What are the implications for law enforcement? That's a great question. And we don't run into that because it's pretty consistent between provinces. My understanding is, I mean, you wouldn't want to be caught transporting cannabis from a legal state to an illegal state for certain. So I think, I mean, even legal users in their home state need to be conscious of the fact that, you know, if you live near a border and you're making a quick trip to make sure that, you know, you're abiding by all the laws that you need to abide by when you get there. I wouldn't want to be caught transporting cannabis and I would hazard that, you know, in terms of impaired driving, it's probably very similar across states, you know, in terms of the methods that police are using to detect impaired driving. But yeah, just to be aware, you know, you shouldn't be transporting your cannabis across state lines. In the United States in particular, marijuana has not been legal for very long, only since 2012. And therefore the research has been impeded because funding isn't available to research an illegal substance. It hasn't traditionally been in this country. So we have about 10 years of literature. In another 10 years, what questions exist that you'd like to see answered by researchers and then published in peer reviewed literature? Or don't we know that you'd like to think we should know? So one of the things that I mentioned at the end of my presentation was the fact that there aren't a lot of hard outcomes being reported specifically to drug impaired driving. As you may have noticed when I was mentioning traffic injuries and traffic fatalities, it was never drug fatalities or it was never validated in people who were impaired or even these injuries or collisions. We never know if these people were actually under the influence of substances besides alcohol. So I think one of the main things that should really be done in the 10 years is working on improving reporting systems and potentially post-mortem analysis of individual's blood to see if they're impaired. And using those data to evaluate whether or not these preventive activities are actually specifically reducing drug impaired driving. Because right now we essentially have no idea. And once that reporting improves, it'd be really terrific to see more interrupted time series analyses or cluster randomized control trials where perhaps some counties in the US have these media campaigns that are aired for a couple of weeks, a couple of months and seeing how that affects law enforcement or these hard outcomes that I had mentioned. Yeah, I certainly agree with the hard outcomes issue how there's actually been a big issue with using those types of data. Like the FARS database actually includes some of that in terms of what drugs were potentially involved. Because a lot of times after a fatal collision, they will actually, they will test for all of these substances. However, the issue being that around the time of legalization, a lot of times there've been changes in how things are measured and the implications being that, an increase in detected cannabis might actually just be an increase in testing. So there's challenges there, but once it becomes more consistent following legalization, it might be possible to do what Amir is suggesting where you're actually, if you're looking at a prevention effort to kind of, as long as the data collection has remained the same, then theoretically you should be able to use that information. I think it's gonna be really interesting to try and tease out really what specific aspects of policies are important. Something that's been really debated is the legal age for cannabis use, for example. We know that cannabis is not good for developing brains. And unfortunately, brain development actually continues into the 20s, but we have, for the most part, made the age of legal use of cannabis the same as the age of legal use of alcohol. So in the States, that's 21, but in Canada, it is 18 or 19 years old. So that's created a few difficulties, especially if you wanna make them not the same. And as well, Amir mentioned the PERS-A data is actually really not there for cannabis use, which surprises me a bit, but I think it might have to do with whether there is public awareness surrounding the use of a PERS-A limit. So again, looking at these kind of large-scale prevention strategies and seeing if they actually work or don't work. And here's one final question. Can you speak more to the fact that impaired driving is increasing the longer cannabis use is legalized? Any thought on why that is? So in, can you, I'm sorry, can you repeat the question? Sure, can you speak more to the fact that impaired driving is increasing the longer cannabis use is legalized? Any thoughts on why that is? So I haven't seen, I haven't seen any evidence that actually looks sort of year to year about increase. I'm not sure, maybe this question is in reference to our analysis where we compared the first versus subsequent years. So yeah, our data actually suggested that perhaps there were fewer collisions in that first year. And I really think that's probably a supply issue. In the very first year when cannabis is legalized, typically, you know, there's a whole new system of bureaucracy that has to get worked out where, you know, the state decides who can be licensed and how they can be licensed and who can grow and how those growers can supply the licensed, you know, retail facilities, et cetera. So it's a very difficult sort of mechanism to get started and there's lots of bumps in the road. So I really think it's probably related to supply that we found that first year had fewer collisions. Yeah, there was the one other study which suggested it could increase after, and I really don't have an explanation for it. I'm trying to look and see if the different study methodologies might have an impact on that, but we haven't quite gotten there yet. Do we know if potency is increasing over time? Potency has increased dramatically. Yeah, maybe Amir has some specifics on this, but I know that, yeah, there's good evidence that the cannabis today is a much, much higher potency than it was, you know, 20 years ago. Yeah, certainly. And the cannabis that was around in the 80s definitely is not the same cannabis that's around today. And there's a lot of cannabis-derived products that can be a lot stronger, for example, edibles or concentrated extracts. So if people are using this and deciding to drive, that's a lot more problematic than what we may have seen 30, 40 years ago. So that may be a contributing factor to that question as well, yeah. Okay, well, I wanna thank you both so much for sharing your expertise and your time with the American College of Preventive Medicine today. We have a few upcoming events on June the 8th at 2 p.m. We have another webinar on cannabis legalization, public health and preventive medicine considerations on June 22nd at 2 p.m. We'll hear from Dr. Boris Lushniak, who's a member of our Board of Regents and the Dean of the University of Maryland School of Public Health. And he'll be addressing the 10 essential public health services in a conversational style format. These were issued several years ago, but were recently reissued here in the United States. So please look out for the event survey from today's webinar and for more education from the American College of Preventive Medicine, we encourage you to log on to our learning management system and look at more than 200 different offerings for you to continue your education in preventive medicine and population health. Thank you. That concludes our webinar today.

cannabis use

impaired driving

cannabis legalization

fatal collisions

prevention strategies

public awareness

warning labels

roadside drug testing

long-term effects