Safety Recommendation h-12-048
Synopsis: This special investigation report looks at one of the most serious types of accidents that occur on our highways: these are collisions involving vehicles traveling the wrong way on high speed divided highways. The goal of this investigative project is to identify relevant safety recommendations to prevent wrong-way collisions on such highways and access ramps. The investigations included in this report take a focused view of the driver and highway issues affecting wrong-way collisions. The report is organized into four sections. Section 1, "Wrong-Way Collisions," defines the problem, examines the National Transportation Safety Board (NTSB) history with these types of collisions and generally surveys the data and research concerning wrong-way driving collisions. Section 2, "NTSB Investigations," summarizes nine NTSB wrong-way collision investigations. Section 3, "Characterization of Wrong-Way Driving," considers the components of wrong-way collisions and uses data, research, and NTSB investigative work to summarize these types of collisions. Section 4, "Countermeasures," provides recommendations to address wrong-way collisions. Those countermeasures are organized to address the following safety issues: Driver impairment, primarily from alcohol use, with consideration of older driver issues and possible drug involvement The need to establish—through traffic control devices and highway design—distinctly different views for motorists approaching entrance and exit ramps Monitoring and intervention programs for wrong-way collisions In-vehicle driver support systems
Recommendation: TO THE AUTOMOTIVE COALITION FOR TRAFFIC SAFETY: Work with the National Highway Traffic Safety Administration to accelerate widespread implementation of Driver Alcohol Detection System for Safety (DADSS) technology by (1) defining usability testing that will guide driver interface design and (2) implementing a communication program that will direct driver education and promote public acceptance.
Original recommendation transmittal letter: PDF
Overall Status: Open - Acceptable Response
Location: Washington DC, DC, United States
Is Reiterated: Yes
Is Hazmat: No
Accident Reports:
Report #: SIR-12-01
Accident Date: 12/11/2012
Issue Date: 12/26/2012
Date Closed:
Addressee(s) and Addressee Status: Automotive Coalition for Traffic Safety (Open - Acceptable Response)

Safety Recommendation History
From: NTSB
To: Automotive Coalition for Traffic Safety
Date: 5/24/2022
Response: The National Transportation Safety Board (NTSB) has reviewed the National Highway Traffic Safety Administration (NHTSA) request for comments (RFC) titled “New Car Assessment Program,” published at 87 Federal Register 46 on March 9, 2022. In its request, NHTSA proposes significant upgrades to the New Car Assessment Program (NCAP) by proposing (1) to add four new advanced driver assistance systems (ADAS) technologies (blind spot detection, blind spot intervention, lane keeping support, and pedestrian automatic emergency braking) to those NHTSA currently recommends, (2) changes to the test procedures and performance criteria of the four currently recommended ADAS, and (3) a roadmap for phased implementation of NCAP updates over the next several years. The RFC describes, “but does not propose,” how NHTSA might (1) rate vehicles with ADAS technologies, (2) include ADAS ratings on the Monroney label (vehicle’s window sticker at the point of sale), and (3) incorporate other safety technologies in the future, particularly those with “potential to help people make safe driving choices.” NHTSA proposes many meaningful changes in this RFC, but these proposals can improve safety only if implemented. The NCAP program will not fulfill its purpose of informing consumers unless NHTSA can keep pace with the emergence and development of safety technologies. NHTSA started the process of expanding NCAP with a 2013 RFC. In 2015, the FAST Act (Section 24322) required NHTSA to promulgate a rule to ensure that crash avoidance information be included on the Monroney (window) label within 1 year. Seven years later, NCAP still does not rate any vehicle safety technologies, the Monroney label has not been modified, nor has NHTSA proposed to do either of these things in the current RFC. Further, some of the technologies described as “emerging” in NHTSA’s 10-year roadmap are already available on vehicles, and ratings of those technologies are being implemented by other NCAPs around the world. In our response to this RFC, the NTSB discusses the overall state of NCAP in relation to our safety recommendations, mandates in other countries, and vehicle rating programs. We also offer comments on specific questions asked in this RFC. 1. Roadmap for Phased Implementation of NCAP Updates The current RFC contains a section on emerging vehicle technology in which NHTSA describes its 10-year roadmap to conduct research and develop test procedures to help determine whether these technologies merit inclusion in NCAP. The emerging technologies that NHTSA discusses in the RFC are: (1) driver monitoring systems, (2) driver distraction, (3) alcohol detection, (4) seat belt interlocks, (5) intelligent speed assist, and (6) rear seat child reminder assist. Our response focuses on several of these emerging technologies. Our safety recommendation history and the current European Union mandates and Euro NCAP requirements indicate that these are far from novel technologies. In fact, most are currently available technologies. Specifically: • Driver monitoring systems o In the Mountain View, California, collision of a vehicle operating in partial automation mode with a crash attenuator, the system designed to determine driver attentiveness was ineffective. The NTSB recommended that NHTSA and SAE International develop performance standards for driver monitoring systems and require them on new vehicles with partial automation capabilities (Safety Recommendations H 20-3 and -4, currently classified “Open?Acceptable Response”). The benefits of such systems would extend to all vehicles regardless of any automation capabilities. o Starting in July 2022, driver drowsiness and attention warning and advanced driver distraction warning systems will be mandatory in passenger vehicles in the European Union. Additionally, Euro NCAP will start rating occupant status monitoring systems in 2023, based on the 2022 testing protocols. • Driver distraction o In 2012, NHTSA published a notice of proposed rulemaking (NPRM) titled “Visual-Manual NHTSA Driver Distraction Guidelines for In-Vehicle Electronic Devices.” In response to this NPRM, the NTSB stated that, as soon as the proposed Driver Distraction Guidelines are adopted, NHTSA should immediately add to its NCAP information a notice of whether a new car complies with the guidelines and also note those vehicles that do not comply. We further emphasized the need for incorporation into NCAP in our response to NHTSA’s 2013 RFC. Although these guidelines were adopted in 2013, NCAP still does not address these aspects, leaving consumers without key information. • Alcohol detection o Although still in development, alcohol detection systems are far from novel. In 2012, as part of the NTSB’s special investigation of wrong-way driving crashes, we recommended that NHTSA and the Automotive Coalition for Traffic Safety Inc. (ACTS) work together to accelerate widespread implementation of Driver Alcohol Detection System for Safety (DADSS) technology by (1) defining usability testing that will guide driver interface design and (2) implementing a communication program that will direct driver education and promote public acceptance. Wrong-Way Driving, Safety Investigation Report NTSB/SIR-12/01 (Washington, DC: NTSB). Safety Recommendation H-12-43 to NHTSA is classified “Open?Unacceptable Response,” and H-12-48 to ACTS is classified “Open?Acceptable Response.” o In 2021, recognizing the lifesaving potential of drunk- and impaired driving prevention technology, Congress stated that such technology must be standard equipment in in all new passenger vehicles and that NHTSA must issue a rule prescribing a federal motor vehicle safety standard to require such technology within 3 years. Section 24220, Infrastructure Investment and Jobs Act, enacted on November 15, 2021. • Intelligent speed assist o As noted earlier (see p. 3), in 2017, the NTSB recommended that NHTSA incentivize passenger vehicle manufacturers and consumers to adopt ISA systems by, for example, including ISA in NCAP. o Starting in July 2022, ISA systems in passenger vehicles will be mandatory in the European Union. According to 661/2009/ED, new models of light vehicles in the European Union will be required to be equipped with ISA by July 2022. This mandate extends to all passenger vehicle models by July 2024.Additionally, Euro NCAP will start rating speed assist systems in 2023, based on the 2022 testing protocols. In this RFC, NHTSA states that it is “exploring opportunities to encourage the development and deployment of these technologies.” Technologies that the European Union started to mandate this year, that Euro NCAP is starting to rate next year, and that the NTSB has been recommending be required for years, are only now starting to be considered by NHTSA. NHTSA is a decade behind the progress of these technologies, and the NTSB urges NHTSA to incorporate these technologies much sooner than the planned 5–10 years from now.

From: NTSB
To: Automotive Coalition for Traffic Safety
Date: 9/6/2013
Response: We are encouraged that ACTS has been working with NHTSA to develop DADSS technology and that funding authorized by the Moving Ahead for Progress in the 21st Century Act has allowed ACTS and NHTSA to continue this project, increase usability testing, conduct public opinion research, and begin a public education communication program. Pending completion of these actions, Safety Recommendation H-12-48 is classified OPEN—ACCEPTABLE RESPONSE.

From: Automotive Coalition for Traffic Safety
Date: 6/26/2013
Response: -From Susan A. Ferguson, Ph.D., President and DADSS Program Manager: This letter is in response to the National Transportation Safety Board’s (NTSB) recommendation (H–12–48) to the Automotive Coalition for Traffic Safety (ACTS) concerning in–vehicle technologies that are being researched in partnership with the National Highway Traffic Safety Administration (NHTSA). The NTSB is recommending that ACTS, “work with NHTSA to accelerate widespread implementation of Driver Alcohol Detection System for Safety (DADSS) technology by (1) defining usability testing that will guide driver interface design and (2) implementing a communication program that will direct driver education and promote public acceptance.” Alcohol–impaired driving is an enduring problem and typically accounts for about a third of all motor vehicle fatalities annually in the United States. In continuing efforts to reduce the adverse consequences of alcohol–impaired driving, NHTSA and ACTS entered into a research program in February 2008 to try to investigate potential in– vehicle countermeasures to the problem. This cooperative research partnership, also known as the Driver Alcohol Detection System for Safety (DADSS) Program, seeks to explore the feasibility, the potential benefits of, and the public policy challenges associated with a more widespread use of non–invasive technology to prevent alcohol– impaired driving. At the culmination of this initial effort, a research vehicle will be available in early 2014 with two different driver alcohol measurement technologies installed that will allow a first look at how such technology might work within the vehicle environment. Since the inception of the DADSS program, it has been clearly acknowledged that for in–vehicle alcohol detection technologies to be acceptable for use among drivers, many of whom do not drink and drive, they must be seamless with the driving task; they must be non–intrusive, that is, accurate, fast, reliable, durable, and require little or no maintenance. This design philosophy is reflected in the Performance Specifications1 established for DADSS and is all the more important because ACTS and NHTSA are adopting a voluntary, non–regulatory, market–driven approach to implementation. The in–vehicle technology will have to meet the needs and approval of the driving public in order for them to be sufficiently motivated to buy it. Research is underway to develop technologies that could prevent the vehicle from being driven when the device registers that the driver’s blood alcohol concentration (BAC) is at or above 0.08 g/dL. The challenge of the DADSS project is to develop and test prototypes that meet the exacting performance standards developed by ACTS. This is a pre–requisite for technologies to be considered for vehicle integration thereafter. Another, equally important, challenge is to ensure that the driving public will be accepting of in–vehicle alcohol detection technology once it meets the stringent performance criteria for in–vehicle use. To that end, a parallel effort is underway to engage the driving public in discussions about the DADSS technology so that their feedback can be incorporated into the final performance specifications. The challenges to meet these requirements are considerable, but the potential life saving benefits are significant. An analysis by the Insurance Institute for Highway Safety (IIHS) estimates that if driver BACs were no greater than 0.08 g/dL, 7,082 of the 10,228 alcohol– impaired road user fatalities occurring in 2010 would have been prevented. For fiscal year 2010, Congress’ appropriation to NHTSA included funding for, “the development of advanced alcohol detection technologies that are less intrusive and hold the most promise for being accepted by the general public.” As DADSS technology research progresses and decisions are made about various features that could be employed and how to integrate such technology into vehicles, NHTSA and ACTS have sought to gain a better understanding of public preferences with respect to in–vehicle alcohol detection devices. Responding to Congress’ direction, the initial DADSS technology research plan has been expanded to include a public opinion research component and includes three elements: Focus groups and a national survey to assess public interest in the technology options and to understand whether the technology options would be acceptable for use by the public; and discussions with key stakeholder groups about the potential impact of the technology options. The findings from these efforts will be assessed to ascertain whether refinement of the Performance Specifications is needed to help to ensure that the DADSS subsystems functions as consumers expect. In addition to the public opinion research, from inception the DADSS program has included limited human subject testing to assess the technology prototypes’ ability to provide accurate measurements of blood alcohol concentrations over a wide range of blood alcohol concentrations, both during the absorption and elimination stages. Once the research vehicle is available with the technologies installed, additional human subject research will be conducted to examine alcohol measurement in the vehicle compartment. In addition the vehicle will be used both to garner driver’s reactions’ to the alcohol measurement systems and to investigate the usability of the driver interface design. As the DADSS research effort proceeds and if it has been established that the DADSS technology is feasible for in–vehicle deployment, the results of the public research will help guide a communications strategy and appropriate messaging directed at driver education and public awareness and acceptance. To that end, discussions are underway with market research firms to define the next steps and scope of such an effort.

From: NTSB
To: Automotive Coalition for Traffic Safety
Date: 6/3/2013
Response: In its 2012 report on wrong-way driving, the NTSB concluded that driver alcohol impairment is the primary cause of wrong-way driving collisions and that the installation of interlocks on the vehicles of all convicted DWI offenders would reduce crashes caused by alcohol-impaired drivers (NTSB 2012c, wrong-way report). Numerous studies have shown that interlocks are effective in reducing recidivism among DWI offenders while the device is installed (Coben and Larkin 1999, 81–87; Tippetts and Voas 1998, 19–24; Willis, Lybrand, and Bellamy 2004; and Vezina 2002, 97–104). According to one estimate from the Insurance Institute for Highway Safety, if all drivers with at least one alcohol-impaired driving conviction within the previous 3 years had used zero-BAC interlocks, approximately 1,100 deaths, or about 10 percent of fatalities associated with alcohol-impaired drivers, could have been prevented in 1 year (Lund, McCartt, and Farmer 2007). As noted earlier, in MAP-21, Congress acknowledged the effectiveness of interlocks by providing funding incentives for states to implement laws requiring interlocks for all convicted DWI offenders. However, relatively few driver DWI arrests ultimately result in the installation of interlocks. In 2011, more than 1.2 million arrests were made for DWI (FBI 2013); yet, as of 2012, fewer than 280,000 interlocks were in use in the United States (Roth 2012). Given the effectiveness of interlocks in reducing the likelihood that offenders will drive impaired, in its 2012 report on wrong-way driving, the NTSB made the following recommendation to the 33 states that do not mandate interlocks for all DWI offenders, the Commonwealth of Puerto Rico, and the District of Columbia: H-12-45 Enact laws to require the use of alcohol ignition interlock devices for all individuals convicted of DWI offenses. Aside from the lack of a universal mandate for interlock use among all convicted DWI offenders, one of the most significant challenges to the potential success of interlocks is the low rate of compliance38 among those offenders required to use them (DeYoung, Tashima, and Masten 2004). For example, one recent study found that within the subset of offenders who were ordered by judges to install interlocks, only about 24 percent ultimately did so (McCartt and others 2013, 215–29). In many states, offenders may avoid installing interlocks by stating that they will not drive during their license suspension period or by claiming that they do not own a vehicle. Two NHTSA-sponsored reports provide suggested best practices for establishing and improving interlock programs (Sprattler 2009; Marques and Voas 2010). To improve offender compliance and program success, the reports advocated the following practices: • Present the interlock as an alternative to a more restrictive penalty, such as house arrest or transdermal monitoring;39 • Provide financial assistance to individuals who cannot afford interlocks using fees from other offenders, arrangements with interlock providers, or alcohol tax revenues; • Document interlock status on driver’s licenses so the information will be available to law enforcement officers during traffic stops; • Establish a protocol for interlock-equipped vehicle usage; for example, track odometer readings or the number of BAC tests per month to ensure that the equipped vehicle is being used; • Penalize drivers who are caught using non-interlocked vehicles with sanctions that are equal to or greater than those associated with driving-after-suspension/revocation charges; • Establish an offender-monitoring program, with preestablished consequences for skipped or failed tests; and • Set criteria for interlock removal based on a period of alcohol-free driving. Some states have adopted some of these strategies to varying degrees. For example, Sprattler (2009) reported that 13 states and the District of Columbia had established indigent funds for offenders who could not afford interlocks. Marques and Voas (2010) reported that five states have provisions for extending interlock periods in response to repeated BAC lockouts, and since 2011, DWI offenders in Washington State may not reinstate their driver’s licenses until their installed interlocks are violation-free for not less than the last 4 months of their suspension period. Additionally, studies that have tracked the implementation of some of the above-listed practices have documented improvements in interlock compliance. For example, a pilot program in one New Mexico county found that mandating house arrest as an alternate sanction to interlock installation led to 70 percent installation rates, compared to 17 percent rates in counties that did not adopt the house-arrest alternative (Roth, Marques, and Voas 2009, 437–41). Another study, conducted in Maryland, found that close monitoring of DWI offenders with installed interlocks resulted in significantly fewer attempts to start the vehicle by drivers with a positive BAC compared to offenders in a traditionally monitored control group (Zador and others 2011, 1960–67). Existing NHTSA publications provide numerous potentially useful suggestions for states that are developing interlock programs; however, the NTSB is concerned that in the absence of more explicit information about which program elements lead to increased compliance, state interlock programs may suffer from low compliance rates and will not achieve their potential. NHTSA’s Uniform Procedures for State Highway Safety Grant Programs (78 Federal Register [FR] 4986, 2013) have established few criteria for states to meet to obtain grant funding for their interlock programs. For example, states with a minimum interlock period as short as 30 days’ duration may still be eligible for grant funding, and no compliance goals or program elements are required to obtain funding. Because NHTSA has the duty to distribute federal grant funds in such a way as to support the establishment of state interlock programs, it has the opportunity to foster the development of highly successful state programs. The NTSB concludes that states would increase the effectiveness of interlock programs by employing those practices that have been shown to increase interlock compliance. Therefore, the NTSB recommends that NHTSA develop and disseminate to the states best practices for increasing interlock installation and compliance that are based on recent NHTSA research. To encourage states to implement these best practices, the NTSB further recommends that NHTSA create incentives for states to adopt the interlock best practices developed in response to Safety Recommendation H-13-02 (above). This recommendation is consistent with the comments the NTSB made to NHTSA concerning state interlock programs on April 24, 2013, in response to the Interim Final Rule on Uniform Procedures for State Highway Safety Grant Programs. (See 78 FR 15, published January 23, 2013.) From the safety report Reaching Zero: Actions to Eliminate Alcohol Impaired Driving (NTSB/SR-13/01, adopted May 13, 2013, notation 8482): Additionally, because it continues to believe in the necessity of all-offender interlock laws, the NTSB reiterates Safety Recommendation H-12-4540 from its wrong-way driving report (NTSB 2012c), which calls on the 33 states that do not mandate the use of interlocks for all DWI offenders, the Commonwealth of Puerto Rico, and the District of Columbia to enact laws to require the use of interlock devices for all individuals convicted of DWI offenses. Although interlocks traditionally have been used as a means of sanction for DWI offenders, they are increasingly being employed by others who recognize their benefits. For example, in Finland, Sweden, and France, interlocks are required on school buses, and in some European countries, commercial transport operators have installed them voluntarily (Daoud 2012). Several highway vehicle manufacturers have developed interlock system