A Horizon Scan of Technology for In-Vehicle Driver Remediation of Drug-Impaired Drivers

Acknowledgements

The authors would like to thank the representatives of the road safety team in the following jurisdictions for providing policy and program information: Australia (Australian Capital Territory, Northern Territory, Queensland, South Australia, Tasmania and Victoria); Canada (Alberta, British Columbia, Manitoba, New Brunswick, Newfoundland and Labrador, Northwest Territories, Nova Scotia, Prince Edward Island, Quebec, Saskatchewan and Yukon), Finland; France; Germany; Iceland; Ireland; Israel; Japan; Netherlands; New Zealand; Norway; Singapore; Spain; Sweden; Switzerland; United Kingdom; and the United States (Alaska, California, Colorado, Delaware, District of Columbia, Illinois, Iowa, Maine, Maryland, Minnesota, Mississippi, Nebraska, New Hampshire, New York, North Dakota, Oregon, Ohio, Rhode Island, South Carolina, Utah, Vermont, Virginia and Washington).

The authors would also like to thank all the manufacturers who provided information on their products: Alcohol Countermeasure Systems (Securetec); Alere Toxicology, Alere Incorporated; American Bio Medica Corp.; Cannabix Technologies; Express Diagnostics; Intelligent Fingerprinting; MAVAND Solutions GmbH; McDevitt Lab New York University/SensoDx; PharmChem Inc.; Sarstedt; Screenitalia; Six Safety Systems; and Verity Diagnostics Inc. (Dräger distributor).

In addition, the authors thank colleagues in the Ontario Ministry of Transportation’s Safety Policy and Education Branch and Licencing Services Branch, as well as colleagues from the Royal Canadian Mounted Police for providing policy and operations consultation. In particular, the authors recognize the contribution of Patrick Byrne, Kirk Dickson, Eamonn Gaffney, Shamaila Mian, Jessica Mahon, Francine Rubin, D’Arcy Smith and Joanna Tsilikas.

This report was prepared by Tracey Ma, Erin Dessau, Megan Svidski, and Yoassry Elzohairy. 

Introduction

Driving under the influence of drugs (DUID) has become a concerning issue due to its prevalence in both fatally-injured drivers (“FARS Encyclopedia: Query,” n.d.) and in drivers sampled from the general population traffic stream (Beirness & Beasley, 2010; Berning, Compton, & Wochinger, 2015). Based on available evidence, there is consensus that the use of drugs while driving is associated with an increased collision risk, although the exact magnitude varies according to drug type (Elvik, 2013). In certain jurisdictions, apprehension about drug-related traffic collisions and other public harms is coupled with additional concerns over its increasing availability, which is presumed to be associated with a changing regulatory landscape. In the United States and Canada, at least, the changing landscape is characterized by a move toward the legalization of cannabis, which happens to be the illicit drug most frequently detected among drivers (Schulze et al., 2012). It is against this backdrop that regulatory responses addressing DUID have emerged and continue to evolve.

Policy measures for preventing and mitigating DUID and its associated harms adopt a deterrence approach, of which there are two distinct mechanisms. General deterrence is aimed at all members of the target population (i.e., licenced drivers) with the goal of dissuading or preventing them from performing the undesirable behaviour. Specific deterrence targets the subgroup that has already performed the undesirable behaviour and/or has been apprehended for doing so. Internationally, both approaches are taken to address DUID (Watson & Mann, 2016). Although complementary, the distinction between general and specific deterrence is an important one. While general deterrence approaches typically demonstrate large population-level impact, specific deterrent approaches are necessary to reduce the burden of harm produced by specific sub-sections of the population such as those who are known or presumed to be repeat offenders, “high risk” drivers, or who have an underlying substance abuse problem. In our observations, specific deterrents tend to be those that are applied upon a conviction in court.

Common requirements for convicted impaired (both alcohol- and drug-impaired) drivers are the completion of remedial education/treatment and participation in remedial monitoring using ignition interlocks, although the latter is more commonly used for convicted alcohol-impaired rather than drug-impaired drivers. An overview of remedial education/treatment programs for DUID (and specifically cannabis) drivers is provided elsewhere (Watson & Mann, 2016). A key component of remedial programs are ignition interlocks. An ignition interlock is an in-vehicle alcohol breath screening device that prevents a vehicle from starting or from continuing to operate if the driver registers a blood alcohol content (BAC) level in excess of a pre-set limit. The device is installed inside the vehicle and is connected to the engine's starter mechanism. Although program design varies from jurisdiction to jurisdiction, generally speaking, ignition interlocks require drivers to provide breath samples before and while driving. Ignition interlocks have a long history of use in remedial road safety programs for preventing alcohol-impaired driving (Bjerre, 2005), and have shown to be effective while installed (Bjerre, 2003).

The use of in-vehicle technologies and devices to support driver safety is a topic that extends far beyond ignition interlocks; briefly, its breadth and history starts with occupant protection devices (e.g., seatbelts) in the 1950s and continues on to include telematics, driver assistance and warning systems, and automated control systems today. With advanced technologies becoming increasingly available, there is a need to examine these technologies and their applications to identify those that may be suitable for wide-scale adoption. While originally designed for drinking and driving deterrence, ignition interlocks are now used as a remedial monitoring tool for convicted drug-impaired drivers in Ontario and possibly elsewhere. The rationale is twofold: first, alcohol and drug consumption have been observed to co-occur (Subbaraman & Kerr, 2009), therefore by preventing alcohol-impaired persons from driving their vehicles, it is presumed that interlocks inadvertently do the same for drug-impaired persons (if those persons are impaired by drugs at the same time), and second, no known technology exists to prevent a person impaired by drugs from driving.

However, the increasing recognition that DUID is a priority issue is accompanied by increasing acknowledgement that drug-impaired driving is a fundamentally different phenomenon than alcohol-impaired driving (Swell, Poling, & Sofuoglu, 2009). As such, appropriate sanctioning regimes that are specific to drugs should be applied to those convicted of drug impaired driving. Of course, any review of road safety sanctions will show the same set of tools being used to sanction different behaviours (e.g., fines for speeding, fines for distracted driving). Our intent in suggesting unique approaches to drug-impaired driving sanctioning is not to advocate for a complete departure from extant approaches, but rather to examine whether known approaches can be modified to better suit the unique circumstances surrounding drug-impaired driving. For example, there are efforts underway to evaluate and validate the contents of education and treatment programs for drug-impaired drivers (Watson & Mann, 2016) . Returning to ignition interlocks, the design and delivery of the interlock-based intervention is not in question. What is in question is whether there is a more suitable technology (than ignition interlocks) that could be used for in-vehicle monitoring and detection of drivers who may be under the influence of drugs.

This report sought to systematically identify all drug monitoring technologies currently available in the marketplace that are capable of immediate drug detection and driving prevention. With ignition interlocks used as an analogy, our focus was placed on technologies that could be situated inside the vehicle, whether physically attached or not, and be self-administered or automatic in the collection and analysis of biological samples (breath or otherwise). A secondary goal was to identify the current stage of development for these technologies and derive requirements that would render these devices suitable for inclusion into a remedial program. Our final goal was to determine whether any jurisdictions currently use such technology or are considering doing so in the near future, as well as how they might implement such a remedial monitoring program.

This work firmly establishes the application of technology currently designed for drug screening and detection within an in-vehicle monitoring and remedial framework as a potential usage scenario (previously unfamiliar to both industry and governments), and to begin to explore the opportunity for such a solution. The process, recommendations and implications to continue the momentum of mitigating DUID harms and augmenting specific deterrent measures are also discussed.

Methods

Two simultaneous reviews were conducted: a technology scan and a jurisdictional scan.

In order to identify technologies that could act analogously to ignition interlocks to prevent driving while impaired by drugs, an appreciation of interlocks is required. Interlock technologies are applied in a broad range of usage scenarios, from application to unoffending commercial drivers as a preventive tool (Bjerre & Kostela, 2008) to application to offending passenger-vehicle drivers as a remediation tool (Coben & Larkin, 1999) (Beirness, Marques, Voas, & Tippetts, 2003). In some instances, the mandatory use of ignition interlocks are established as a condition of retaining a valid licence after a driving under the influence (DUI) conviction, whereby drivers are given the option to surrender their driving privileges for a specific duration or to continue driving during that time with an installed device.

Three broad functions of ignition interlocks serve as benchmarks for the capabilities of drug monitoring technologies targeted in our search: first, the device must collect a biological sample, then it must analyse the sample in real time to generate some sort of result (e.g., over or under a threshold, numeric concentration of substance), and finally, it must provide feedback to the vehicle or to the driver (i.e., disables ignition, initiates sound and light signals). While these aspects point to the functions required of drug monitoring technologies, there are additional characteristics that render a given technology suitable for an in-vehicle remediation usage scenario. Considerations previously identified as important in choosing a drug detection system (Jones, Shinar, & Walsh, 2003; Martin, 2010; Parmeter, Murray, & Hannum, 2000) were collapsed into a shortlist of characteristics, which were circulated to a team of experts and stakeholders in a consensus generating exercise that mimics the Delphi Technique (Thangaratinam & Redman, 2005). In these consensus rounds, characteristics necessary for this usage scenario were refined through addition/subtraction, description, and rank order. The final set of characteristics was used to inform data collection (see Results).

A systematic search of peer-reviewed literature, grey literature, and popular news articles was then initiated using pre-identified search terms in the following databases: ProQuest, PsychINFO, Scholar’s Portal, PubMed, and JSTOR. The search was also replicated on Google Scholar to ensure that other sources such as blog posts were identified. Authors may be contacted for the exact search strategy. All returned items were reviewed for relevance, and this process culminated in a contact database of device inventors and manufacturers. The identified contacts were invited to complete a structured questionnaire to self-disclose information relating to the characteristics and functions of their product. The questionnaire was designed to elicit information relevant to the final set of characteristics. Due to the wealth of information desired, the questionnaire was divided into two parts: a screening survey and a follow-up survey.

The initial screening survey (see Table 1) focused on general questions related to the most important characteristics, in order to identify the suitability of the device relative to the requirements identified. Following the collection and analysis of results from the initial survey, a follow-up survey (see Table 2) was sent to select manufacturers based on the answers to the initial screening survey. Any manufacturer whose device had the potential to be a promising solution, based on the characteristics identified, was contacted for more information through the follow-up survey. The follow-up survey asked manufacturers more specific and in-depth questions about their technologies and invited them to share, among other things, the reliability, specificity, and sensitivity of their devices for each drug the device was able to detect.

A jurisdictional scan was conducted to understand whether and how other governments might implement an in-vehicle remedial monitoring framework to prevent drug-impaired driving. Our interest was to garner information about other potential usage scenarios and program designs. Further, by engaging with other jurisdictions, any promising technologies that were missed by the technology scan, detailed above, would presumably be identified. Jurisdictions were selected based on the heuristic that if they have a combination of a strong road safety record, advanced technological development, legalized or decriminalized marijuana, and/or strong impaired driving countermeasures, then one could reasonably expect them to use drug monitoring technologies in their driver safety programs if these technologies are available. All Canadian provinces and territories, select U.S. States and select countries across Europe and Asia were invited to participate (see Table 3).

Similar to manufacturers, jurisdictions were also invited to complete a structured two-stage questionnaire. The initial screening survey (see Table 4) asked broad questions about the priority level assigned to drug-impaired driving, the current measures in place, and whether the jurisdiction was looking to create a remedial monitoring regime specifically for drug-impaired driving. If jurisdictions identified drug-impaired driving as a priority and indicated that they have or were interested in developing a program for drug-impaired drivers that involves the use of technology for ongoing monitoring and detection of drug use, then they were invited to complete a follow-up survey (see Table 5). The follow-up survey asked more detailed and in-depth questions about the potential program policies, procedures and technologies that were being considered by the jurisdiction.

Results

Characteristics of the Technology

Nine broad characteristics were identified as those that would render a given technology suitable for the in-vehicle drug monitoring of drivers in a manner analogous to the ignition interlock for alcohol monitoring (see Table 6). Together, these characteristics represent scientific/technical considerations in addition to the practical considerations for this particular usage scenario. These characteristics formed the basis of data collection from manufacturers/ inventors.

Technology Availability in the Marketplace

Initial screening surveys went out to 43 manufacturers, 13 of whom responded. Manufacturers who did not initially respond were sent multiple (up to three) reminder emails and received one phone call over a three-month period. Based on the responses from the screening survey, it appeared that most devices were designed to detect the following drug classes: cannabis (100%), amphetamines (92%), cocaine (92%), opiates/opioids (92%) and benzodiazepines (75%). When ‘Other’ was indicated, respondents specified methamphetamines, ketamine, morphine, heroin, codeine, alcohol, barbituates, buprenorphine, cotinine, EDDP, methadone, oxycodone, hydrocodone, hydromorphone, oxymorphone, and PCP. Of the manufacturers who responded to the screening survey and provided relevant information, 75% (8 out of 12) indicated that their device can detect five or more types of drugs.

Nearly all manufacturers claimed that their device detects drug presence within 15 minutes following use, with most identifying that detection begins immediately. Two exceptions are: Express Diagnostics, which claims to detect cannabis and amphetamines 30 minutes after use, and opiates and methadone 60 minutes after use; and Verify Diagnostics Inc., whose device takes between 120 minutes and 240 minutes to detect some drugs (amphetamines, opiates, cocaine and PCP) following ingestion. Devices also varied in how long it takes to process the sample. Of those who provided information, manufacturers stated that drugs could be detected using the device in as quickly as 2 minutes and as long as 10 minutes.

When asked whether there is a copyright or patent on the technology, 77% of manfuactuers indicated that a patent exists while 2 others indicated that the patent is pending. Only one manufactuer does not currently have a pending or operational patent. The survey also asked if the technology has been deployed in an organization or jurisdictional setting. While some (30%) stated that their technology is not currently being used, the other 70% stated that their device is used in a organization (i.e. for workplace drug detection) or jurisdiction (i.e. for roadside drug-driving enforcement). The question did not allow participants to further explain the current use, so the actual deployment purpose or the magnitude of distribution cannot be pinpointed at this time.

When asked about how the device interacts with subjects, many (46%) manufacturers indicated that their product could be considered “on-body”, while 15% answered that their device would be considered “in-vehicle”. The remaining 38% answered “other” with such responses as “saliva testing” and “none of the above”. The question did not resonate with survey respondents, as saliva testing devices should be included under “in-vehicle”. Thus, caution must be used with categorizing the technologies in this manner and interpreting these results.

All manufacturers indicated that their devices are easy to use, with all recommending either minimal training (87%) or some training (13%) to use their devices properly. Most manufacturers did not provide a clear user cost for their product. This result suggests that this has not yet been developed or that it is dependent on the quantity ordered. Those that did provide cost estimates mostly indicated that their products have single-use components ($15-40 each) for sample collection and a second, more expensive component ($4,500 - $8,700) for sample analysis. Some manufacturers indicated that price is dependent on the volumes purchased. For others, the price point is still being developed.

In-depth follow-up surveys were sent to 12 out of 13 manufacturers. The one who did not receive a follow-up survey indicated that their product was not suitable for the purpose identified. Three manufacturers responded to the follow-up survey. Manufacturers who did not initially respond were sent one reminder email.

Of the manufacturers who were sent a follow-up survey, 3 responded. Two of these manufacturers are based in the United Kingdom and one in the United States. One of the 3 devices analyzes the user’s fingerprint on a collection cartridge to identify recent drug use, while the other 2 technologies are oral screening devices that analyze saliva samples to test for the presence of drugs.

When asked if the product had been validated, 2 manufacturers indicated that their device had been. For all 3 who responded, product reliability for all drugs detected was over 95%. For the 2 manufacturers that provided information on sensitivity and specificity, across the drug panel sensitivity was above 83 and specificity was above 98. Cocaine, opioids and amphetamines were the most accurately detected drugs across products. For all 3 manufacturers who responded to the follow-up survey, their products do not expose individuals to risk of injury or have intrusive sample collection. Currently, these products are used by a third party (i.e. an officer at roadside), so there are no ethical or privacy concerns, or any opportunity for cheating.

All manufacturers indicated that their devices were designed for roadside enforcement and general drug detection, with one mentioning additional situations including the workplace, prisons or drug rehabilitation settings. No manufacturer provided an indication that their device could be fitted to specific vehicles types (i.e. passenger car, motorcycle, truck, electric vehicle or other) or that it could interact with the vehicle. When asked about the requirements for operational requirements, 2 manufacturers’ devices required a power source and both oral fluid screening devices required single-use components. The size and weight of the devices varied dramatically, with the smallest device weighing just 58 grams and measuring 3.8cm X 3.8cm X 3.8cm, and the largest device weighing 2 kilograms. Manufacturers could not directly answer how they could mitigate driver distraction while using the device, providing answers such as “not applicable” and “not sure what you are specifically looking for”.

A series of questions investigated how the device collects a sample with the following results. When asked about exposure to toxicity, manufacturers noted that either there was not any or that precautions had been considered (i.e. sample enclosed inside the device). All manufacturers indicated that their device had a single use component, with only one detailing a plan for proper disposal by placing the used component back into the bag in which it came. When asked to describe the sample collection process, manufacturers of both oral fluid devices indicated that the individual would place the swab in their mouth and actively swab. The other manufacturer indicated that the user would press their finger on the collection cartridge for five seconds. No manufacturer indicated that sample collection was intrusive.

Several questions focused on the data collection and transmission capabilities of the devices. Manufacturers indicated that devices would have a specified threshold for detection for each drug, which varied from 10ng to 50ng. When asked about user authentication, manufacturers of oral fluid devices said that they would rely on information collected by police (assuming that the test is conducted by a second individual), and the manufacturer of the fingerprint device said they could identify individuals by the fingerprint, in addition to collecting a sample. All manufacturers indicated that the data can be transmitted to a server, and two manufacturers indicated that their device can conduct random or intermittent testing. Only the manufacturer of the fingerprint device responded to questions about tampering and customized data extraction, and indicated that data could be encrypted and that reports could be customized.

When asked about how the device is currently being used, both oral fluid device manufacturers indicated that their device is currently in use for roadside law enforcement efforts, and the manufacturer of the fingerprint device answered that the device “can be used for all circumstances where conventional drug screening is currently used”.

Jurisdictional Use of Such Technology

Initial screening surveys went out to 68 jurisdictions including all Canadian provinces and territories, select U.S. states, and select countries across Europe and Asia (see Methods for selection criteria). Of the jurisdictions contacted, 87% (59 out of 68) responded to the initial screening survey. Jurisdictions who did not initially respond were sent multiple (up to 3) reminder emails over a 3-month period.

Half of jurisdictions surveyed identified drug-impaired driving as a priority in their jurisdiction, while most other respondents (41%) told us there was insufficient data to determine if a problem existed. Only 5 (8%) jurisdictions indicated that drug-impaired driving is not a priority at this time. When asked about current programs for drug-impaired drivers that use technology for ongoing monitoring and detection of drug use (analogous to an ignition interlock device for alcohol-impaired drivers), no jurisdiction indicated they had initiated such as program. While most jurisdictions (61%) indicated that they have no plans to institute such a program, 36% of jurisdictions indicated that they would be interested in developing such a program in the next 5 years.

Almost one quarter (24%) of jurisdictions identified DUID as a priority and were not interested in implementing a remedial monitoring program for DUID in the next five years. Some of these jurisdictions (30%) indicated that they are currently focusing their efforts on other components of their impaired driving enforcement including roadside detection and alcohol-impaired policies. A couple jurisdictions stated that they might be interested in implementing a remedial monitoring program for DUID when the technology becomes available. A few jurisdictions, including Ontario and Colorado, are currently imposing the ignition interlock condition on drivers convicted of DUID.

A follow-up survey was sent to 23 jurisdictions that answered in the screening survey that they are interested in instituting a remedial monitoring program specific to drug-impaired drivers. Nine of these jurisdictions (39%) responded to the follow-up survey. Jurisdictions that did not initially respond were sent one reminder email. No jurisdiction indicated that they have implemented a drug-specific remedial monitoring program and all indicated that they are in the early stages of program development, with many (46%) currently working to identify a technology. Consistent with this finding, nearly all respondents (12 out of 13) indicated that they have not selected a specific device for program use. Almost all (85%) jurisdictions intend to build a remedial monitoring program for drugs on their existing framework for alcohol impaired drivers, and most (78%) plan to target either administrative, post-criminal conviction or both scenarios, with the others being undecided at this point.

Discussion

Driving under the influence of drugs (DUID) and the resulting injuries and fatalities due to these type of collision is a concern for transportation officials, road safety policy makers, injury prevention stakeholders, and health system administrators. In fact, 44% of all global jurisdictions surveyed here indicate DUID to be a priority issue. Deterrent approaches need be broad. Measures to detect (e.g., standardized field sobriety test), apprehend (e.g., criminal statutes), and sanction (e.g., licence suspensions) DUID drivers at roadside are in effect, albeit to varying degrees across the globe (Watson & Mann, 2016). Yet, efforts to monitor drivers for DUID are scarce. However, with a proven monitoring model for alcohol-impaired driving in place (ignition interlocks) and with advanced technologies emerging on the marketplace, it is an opportune time to examine new and emerging technologies for their application to in-vehicle remediation.

Research into capabilities of all 43 products identified, including the self-reported responses received point to an important conclusion: that most, if not all, of these technologies are in early research and development (R&D). The implications of this finding are further outlined below. The current state of these technologies, recognizing that technical specifications and product characteristics, including reliability and validity, may have changed from the time of writing. At the present time, it is noted that certain characteristics, such as single-use components and long (3-10 minute) processing times, would not make many devices immediately amenable to in-vehicle use. Returning to the 3 functions identified at the outset (sample collection, analysis and feedback), all devices had a mechanism for sample collection and analysis (although duration and accuracy of analysis remains to be seen for some), but none were identified as currently capable of providing feedback to the vehicle or driver.

Without a mode of feedback (for ignition interlocks, the ignition is disabled while stationary and there are auditory and visual signals while the vehicle is in motion), the impaired driving behaviour could be detected but not prevented nor reprimanded. To effectively change behaviour, the stimulus needs to be paired with a response, thereby introducing a consequence to the undesirable behaviour or actively preventing the undesirable behaviour. Feedback is especially important for this usage scenario, where monitoring devices are attached to the vehicle and where there is not necessarily a third party present to take immediate action based on the result provided by the device. It is worth noting that the feedback does not necessarily need to be from the device to the vehicle or to the driver. Rather, the feedback could be an automatic alert to authorities with location coordinates embedded in the message. What is essential is that there is a viable means to enforce safe driving behaviour when undesirable behaviour (drug-impaired driving) is detected.

From our results, it appears that all devices currently available or in development were designed for a specific usage (i.e. screening at roadside or in an employment setting), but not the one identified here. Moreover, many surveyed jurisdictions (36%) indicated interest in an in-vehicle driver remediation program. This indicates that there is a market for a product that could be commercially valuable and policy relevant. It may be in the interest of governments across jurisdictions to support or guide such developments, either individually or as a collective group of end users. Such collaboration or information sharing could capitalize on existing mechanisms such as technology transfer offices, scale-up funding, procurement, and vendor information sessions that communicate intended usages and practical value, and could contribute to the development of standards and validation protocols for such technology.

Before any government program is to use such a device, external validation must occur. Many of the technologies identified here have not gone through a thorough validation process to ensure the device performs well, and fewer have had this done by an unbiased, recognized research institute. One example of such a process is the work to test the accuracy and reliability of oral fluid screening devices (Beirness & Smith, 2016). The project tested three commercially available devices (Alere, Dräger and Securetec) and found that all three were able to correctly detect the presence of drugs when the drug was present and to correctly indicate the absence of a drug when no drug was present.

Strengths and Limitations

There are several strengths to our research. The systematic approach that identified all potential technologies provided a comprehensive and exhaustive list of manufacturers with devices who could meet this niche need. The identification of selection criteria was decided upon following consultation with a panel of subject matter experts, including internal government staff from a variety of areas and those who work intimately with the ignition interlock program in Ontario as well as the assessment of oral fluid screening devices across Canada (Keeping & Huggins, 2017). Finally, identifying a clear usage scenario allowed us to build a list of characteristics and specific questions to see if programs and technologies met the identified needs.

There are some limitations to this paper that must be discussed. First, the research collected relies solely on self-report data. It is possible that answers provided by manufacturers and jurisdictions may be inaccurate; potentially to sell or exaggerate features of devices or to protect early stage developments or copyright concerns. It is also possible that the manufacturer or jurisdiction contact that was reached was not aware of all developments or future plans and their survey answers could be incomplete. The response rate for all four surveys was also lower than anticipated, especially on the manufacturer side. However, we hypothesize that manufacturers and jurisdictions who do not engage in this type of work declined to provide survey information. On the manufacturer side, the low response rate could be a reflection the fact that this type of technology is not yet available and that there is a lack of awareness of this potential usage scenario. Moreover, the manufacturer survey missed an opportunity to directly ask how the sample was collected (i.e. saliva, blood, sweat). For the most part, this information came out through the answers to other questions, but could have been more clearly communicated with a direct question. Finally, it is possible that targeting our search to such a specific usage could have eliminated potential manufacturers from our search. While we did attempt to mitigate this through the use of broad search terms and a low threshold for contacting manufacturers, it is possible that our focus biased the search strategy.

Conclusion

This research sought to systematically identify all drug monitoring technologies currently available in the marketplace that are capable of immediate drug detection and driving prevention, as well as to identify the current stage of development for these technologies and derive program requirements. We used ignition interlocks as our analogy, which are used in a remedial monitoring regime for alcohol impairment. Thus our focus was placed on technologies that could be situated inside the vehicle and be self-administered or automatic in sample collection and analysis. We also sought to determine whether any jurisdictions currently use such technology or may do so in the near future, as well as how they might implement such a program. Through this work, we have firmly established the application of technology currently designed for drug screening and detection within an in-vehicle monitoring and remedial framework as a potential usage scenario, one which was previously unfamiliar to both industry and governments.

Future implementation of such monitoring technologies into the regulatory fabric will undoubtedly involve operational, policy, and practice changes. For example, in Ontario, the use of such technologies would require changes to existing provincial and federal legislation before they could be employed in the same manner as alcohol ignition interlock devices are currently used. However, against the backdrop of a constantly changing landscape, both in terms of drug-impaired driving and in terms of technological development, it is not too soon to take the first steps in broadening the tools for detection and monitoring of drug-impaired drivers.

Future work in this area could be harmonized with manufacturers and governments across the globe. This will ensure all facets of development are efficiently operating; both simultaneously and in alignment with one another. With an understanding of the context and a global view of the technological solutions and jurisdictional intentions, an opportunity may exist for partnerships to be developed across governments and industry.

Bibliography

Beirness, D. J., & Smith, D. R. (2016). An assessment of oral fluid drug screening devices. Canadian Society of Forensic Science Journal, 50(2), 55-63.

Beirness, D., Marques, P., Voas, R., & Tippetts, A. (2003). The impact of mandatory versus voluntary participation in the alberta ignition interlock program. Traffic Injury Prevention, 24-27.

Bjerre. (2003). An evaluation of the Swedish ignition interlock program. Traffic Injury Prevention, 4(2), 98-104.

Bjerre. (2005). Primary and secondary prevention of drink driving by the use of alcolock device an program: Swedish experiences. Accident; Analysis and Prevention, 1145-1152.

Bjerre, B., & Kostela, J. (2008). Primary prevention of drink driving by the large-scale use of alcolocks in commercial vehicles. Accident Analysis and Prevention, 40, 1294-1299.

Coben, J. H., & Larkin, G. L. (1999). Effectiveness of ignition interlock devices in reducing drunk driving. Am. J. Prev. Med, 16(1), 81-87.

Keeping, Z., & Huggins, R. (2017). Final Report on the Oral Fluid Drug Screening Device Pilot Project . Public Safety Canada.

Pehrsson, A., Blencowe, T., Vimpari, K., Impinen, A., Gunnar, T., & Lillsunde, P. (2011). Performance evaluation of the DrugWipe® 5/5+ on-site oral. International Journal of Legal Medicine, 675-684.

Subbaraman, M., & Kerr, W. (2009). Simultaneous versus concurrent use of alcohol and cannabis in the national alcohol survey. Alcoholism: Clinical and Experimental Research, 39(5), 872-879.

Swell, R. A., Poling, J., & Sofuoglu, M. (2009). The effect of cannabis compared with alcohol on driving. The American Journal on Addictions, 18, 185-193.

Thangaratinam, S., & Redman, C. W. (2005). The Delphi technique. The Obstetrician & Gynaecologist, 7, 120-125.

Watson, T. M., & Mann, R. E. (2016). International approaches to driving under the influence of cannabis: A review of evidence on impact. Drug and Alcohol Dependence, 169, 148-155.

Table 1: Manufacturer Screening Survey Questions

Table 1: Manufacturer Screening Survey Questions

A.

The Device

1 Which drugs does your technology detect? (check all that apply)
  1. Cannabis
  2. Cocaine
  3. Opiates/opioids
  4. Amphetamines
  5. Benzodiazepines
  6. Other (please specify)
2 Is the device meant for in-vehicle or on-body use?
  1. In-vehicle
  2. On-body
  3. Other (please specify)
3 How much training is needed to properly use the device?
  1. A great deal
  2. Some
  3. Minimal
4 How easy is it to use the device correctly?
  1. Very easy to use correctly
  2. Fairly easy to use correctly
  3. Somewhat difficult to use correctly
  4. Hard to use correctly
  5. Other (comments)

B.

Detection Capabilities

1 Can your technology detect more than one drug?
  1. Yes
  2. No
2 Name of the drug:
3 How quickly can the drug be detected after use?
4 For how long following use can it be detected by your technology?
5 What is the total processing time, from start to finish, for your technology to provide a result?
  1. 1 second
  2. 10 seconds
  3. 1 minute
  4. Other (please specify):
6 Are there any other drugs your technology can detect?
  1. Yes
  2. No
    (If YES, then answer same list of questions for each drug as applicable.)

C.

Service Provider

1 What is the user cost to obtain and install the device, and then to maintain the device (e.g. single-use components, recalibration or service, etc.)?
2 Is there a copyright or patent on your technology?
  1. Yes
  2. No
  3. Pending
3 Has your technology been deployed in an organization or jurisdiction setting?
  1. Yes
  2. No

D.

Additional Comments

1 Please leave your comments here:

E.

Contact Information

1 What is the name of your company?
2 What is the name of the person filling out this survey?
3 What is the person's email address?

Table 2: Manufacturer Follow up Survey Questions

Table 2: Manufacturer Follow up Survey Questions

A.

The Device

1 What is the name of this product?
2 Has the technology been validated?
  1. Yes
  2. No
3 How has the technology been validated? Please keep your answer to 500 characters.
4 What are the dimensions of this product (length by width by height)? Please specify your unit of measurement.
5 What is the product’s weight? Please specify your unit of measurement.
6 What is the intended use of the device? Please select all that apply.
  1. Roadside enforcement
  2. In-vehicle monitoring
  3. General drug detection
  4. Other
7 What are the requirements for operation? Please select all that apply.
  1. Power source
  2. Specific placement of device
  3. Single-use component
  4. Other
8 Could your technology be fitted to any vehicle? Please select all that apply.
  1. Passenger car
  2. Motorcycle
  3. Truck
  4. Electric Vehicle
  5. Other
9 Please discuss your efforts to mitigate driver distraction while using this product.
Please keep your answer to 500 characters.
10 Please discuss your efforts to mitigate exposure to any toxicity associated with using this product.
Please keep your answer to 500 characters.
11 If there are single-use components, is there a disposal mechanism or plan built in to the technology?
  1. There are no single-use components
  2. Yes
  3. No
  4. Other
12 How intrusive is the collection method? (Choose from scale of 1 to 5)
13 Describe how the sample will be collected. Please keep your answer to 500 characters.
14 What are the types of data collected (e.g. which drug(s) are detected, absence/presence of drug(s), concentration levels)? Please keep your answer to 500 characters.
15 Is there a way of verifying the user’s identity?
  1. Yes
  2. No
16 How is the user's identity verified? Please keep your answer to 500 characters.

B.

The Device - Per Substance

1 Can your device detect more than one type of drug?
  1. Yes
  2. No
2 Name of the drug:
3 How much of the sample is required for the test to be conducted?
4 What is the sensitivity of drug detection (percentage)?
5 What is the specificity of drug detection (percentage)?
6 If a sample were to be tested repeatedly 100 times, how many of them would yield the same result?
7 Are there any other drugs your technology can detect?
  1. Yes
  2. No

(If YES, then answer same list of questions for each drug as applicable.)

C.

Service Provider

1 Can this data be transmitted to a server?
  1. Yes
  2. No
  3. Other
2 Please discuss your efforts to mitigate the tampering of data transmitted from the product. Please keep your answer to 500 characters.
3 Once the data has been transmitted to a server, can the data then be customized for specific reporting requirements?
  1. Yes
  2. No
  3. Other
4 Do you have the following data collection capacities?
  1. Random intermittent tests
  2. Ongoing “real time” monitoring
  3. Other
5 What is your ability to interact with the vehicle (e.g. turning off engine, turning on alarm or lights) or your ability to send an alert to enforcement for action? Please keep your answer to 500 characters.
6 What are some examples of how your device can or has been used? Please keep your answer to 500 characters.

D.

Additional Comments

1 Please leave your comments here:

E.

Contact Information

1 What is the name of your company?
2 What is the name of the person filling out this survey?
3 What is the person's email address?

Table 3: Jurisdictions Surveyed

Table 3: Jurisdictions Surveyed

Canada

Alberta Newfoundland and Labrador Prince Edward Island
British Columbia Northwest Territories Quebec
Manitoba Nova Scotia Saskatchewan
New Brunswick Nunavut Yukon Territory

United States

Alabama Maine North Carolina
Alaska Maryland North Dakota
California Massachusetts Ohio
Colorado Michigan Oregon
Connecticut Minnesota Rhode Island
Delaware Mississippi South Carolina
District of Columbia Nebraska Utah
Hawaii New Hampshire Vermont
Illinois New Jersey Virginia
Iowa New York Washington
Louisiana

International

Australia Ireland Singapore
Denmark Israel Spain
Finland Japan Sweden
France Netherlands Switzerland
Germany New Zealand United Kingdom
Iceland Norway

Table 4: Jurisdictional Screening Survey Questions

Table 4: Jurisdictional Screening Survey Questions

1

To what extent is drug-impaired driving a problem in your jurisdiction?

  1. We do not have enough data to understand the magnitude of the problem
  2. We know it is a problem from our data, but it is not a priority for policy/programming
  3. We know it is a problem from our data, and it is a priority for policy/programming

Comments:

2 Does your jurisdiction currently have a program in place for drug-impaired drivers that involves the use of technology for ongoing monitoring and detection of drug use (analogous to an ignition interlock device for alcohol-impaired drivers)?
  1. No, and we have no plans to introduce any within the next 5 years
  2. No, but we are interested in implementing such a program within the next 5 years (provided that the technology is available)
  3. Yes, and we plan to continue with the program
  4. Yes, and we plan to expand the program
  5. Yes, but we plan to discontinue the program
3 If yes to Q2, which technologies are currently in use or are being considered for use?
  1. Marijuana breathalyzer
  2. Breathalyzer for hard drugs
  3. Lifeloc drug screening devices
  4. SensAbues breath testing system
  5. Intelligent fingerprinting
  6. Oral screening devices
  7. Blood testing
  8. “Lab-on-a-chip” using a saliva swab
  9. Automated colorimetrics
  10. Other
4 If you answered “Other” in Q3, please list the technologies you use or are considering using.
5 If yes to Q2, we may wish to contact you for follow-up. Please provide your contact information below.

Name:

Title:

Email:

Table 5: Jurisdictional Follow up Survey Questions

Table 5: Jurisdictional Follow up Survey Questions

A.

The Technology

1 Have you selected a specific device or type of device for in-vehicle or on-body drug detection for program use?
  1. Yes
  2. No (Skip to A3)
2 What is the name of the product and manufacturer? If you there are multiple, please list them all.
3 What are the criteria you have considered or are considering in choosing a technology?
4 (If A.1 = A) How does the technology or technologies you have selected meet these criteria?

B.

Implementation

1 What stage of policy development are you currently in?
  1. Identifying a technology
  2. Developing standards and guidelines and service provider contracts
  3. Have implemented a program (if C = question: what month and year?)
  4. Other (please specify):
2 Which scenarios does the program target, or are you considering targeting? (E.g. post-criminal conviction, administrative sanctions etc.)
3 Do you intend on building upon your existing framework in place for alcohol impaired drivers (e.g. ignition interlock program framework)? If so, what are some of the key elements of your drinking and driving program that you wish to follow?
4 What is the service delivery model you have currently or are considering? Please discuss the following:
  1. Mandatory vs. opt-in
  2. Additional concurrent, preceding, or sequential components to the program (e.g. attending remedial programs)
  3. Who would administer the program? (Service provider? Government?)
  4. How is the driver monitored through the program?
  5. What are the penalties for non-compliance?
  6. Cost of the program (e.g. user pay?)
5 Are legislative/regulatory amendments required in order to implement the program?
If so, please provide link to relevant statute and list the applicable sections.

Table 6: Characteristics of the Technology

Table 6: Characteristics of the Technology

Characteristics

Description

User Cost

  • The cost to obtain and maintain the device (single-use components, recalibration or service, etc.)

Simplicity of Use

  • How easy it is to use the device properly and how much training or materials are needed?

Detection Capabilities

  • How quickly the drug can be detected after use and for how long following use detected?
  • The total time from start to end required for a result (processing time).
  • How much of the sample is required is required for the test to be conducted (size of sample required)?
  • Which drugs does the device detect?

Reliability

  • The dependability of the test to reproduce accurate and precise results consistently.

Accuracy

  • The consistency of the method to reproduce results that are consistent with actual values.
  • The ability to detect drug presence when the drug is actually present (sensitivity).
  • The ability to show a negative reading when the drug is not there (specificity).

I & IT Infrastructure (of the service provider)

  • Does the device record data (e.g., performance failures, program violations) that can securely be transmitted to a server?
  • Ability to do random intermittent test or ability for ongoing “in real time” monitoring.
  • Ability to interact with starter mechanism of the vehicle or to send alerts to enforcement.

Safety and Environmental Issues

  • The degree to which use exposes a person to risk of injury or toxicity associated with collection.
  • Potential for driver distraction.
  • Disposal of single-use components.
  • How intrusive the extraction of the sample is?

Portability

  • How easily the device can be moved?
  • The size and weight of the device.
  • Appropriateness of use for EVs, motorcycles, trucks, etc.
  • The requirements for operation (power source, placement of device, etc.)
  • Is the device meant for in-vehicle or on-body use?

Ethical and Privacy Concerns

  • Privacy for person using device.
  • User verification and authenticity or other security measure to prevent cheating.
Back to Top