6.1 Introduction

This report examines the safety of cellular telephone use while driving by reviewing available data and information, examining crash statistics, performing statistical analysis, and conducting a comprehensive review of relevant published research studies. In this concluding chapter, key findings are summarized and discussed. Recommendations are made in the areas of consumer education, cellular telephone design, future research, and legislation. It is hoped that these initiatives would help ensure that the economic, safety, and convenience benefits of cellular telephones can be maintained within a safety envelope acceptable to both the public and the stakeholders.

Chapter 6 - Table of Contents

Document Table of Contents

6.2 Synopsis of the Findings

Cellular telephone use is rapidly expanding worldwide and are increasingly being used by all age groups for personal communications, while business use continues unabated. Furthermore, the cellular telephone user population is expanding to include a broader representation of socioeconomic groups.

These trends have both positive and negative safety implications. Some new cellular telephone users will place calls while driving, which may lead to greater exposure to cellular telephone-

related distractions in the driving population, all else being equal. Driver inattention to the driving task, the key safety-relevant outcome of driver distraction, has been implicated in many traffic crashes (Sussman, Bishop, Hadnick, and Walter, 1985). The distraction potential may be reduced if drivers are aware of the hazards and use their cellular telephones carefully while on the road. Distraction potential can also be reduced by ergonomically sound cellular telephone designs and new Intelligent Transportation System (ITS) technologies that may be capable of compensating for driver distraction by alerting drivers when traffic conflicts or hazards are present.

The safety benefits of cellular telephones are well recognized as users frequently make calls to report disabled vehicles, accidents, hazardous road conditions, medical emergencies, and crimes in progress. However, the safety benefits are not without drawbacks. For example, some emergency response networks have reported in excess of one hundred "911" calls for the same incident, making the networks unavailable for reporting other emergencies. Furthermore, traffic safety itself may be degraded somewhat if more drivers are distracted while making such calls in hazardous driving situations, e.g., slowed or stop-and-go traffic, and rubbernecking.

Older drivers in general find it more difficult to perform concurrent tasks and process information quickly (Llaneras, Swezey, and Brock, 1993). A cellular telephone, if used while driving, may aggravate age-related problems by introducing a distracting, concurrent task. In addition, older drivers will often find it more challenging to operate cellular telephones that tend toward small displays and controls designed to specifications drawn from a younger population. As reported in Smith, Meshkati, and Robertson (1993), the older driver is generally known to take steps to minimize driving workload in general (e.g., by driving less, driving more familiar routes, driving more slowly, and anticipating traffic signal changes,). At this time, the cellular telephone use patterns of older drivers are not well documented or understood.

Survey results indicate that most people perceive cellular telephone use while driving as distracting, and a sizeable minority report they never use the cellular telephone while driving because it is too risky. This is encouraging because awareness of risk is necessary, though not necessarily sufficient, for prudent risk management. Thus, driver motivations as well as perceptions of the likelihood of a mishap may still promote cellular telephone use while driving. The result may be an increased likelihood of a crash when perceptions are inaccurate and motivations are misguided. In this regard, most of the industry material made available to cellular subscribers urges caution during phone use while driving.

It is reasonable to expect that highway safety crash records should provide definitive data on the role or non-role that cellular telephone use plays in traffic crashes. Unfortunately, only Oklahoma and Minnesota provide police crash report (PCR) forms with data elements that attempt to address cellular telephone use as a pre-crash variable. It is not clear whether the small number of cellular telephone-related crash reports in these and the NHTSA (FARS and NASS) data sources indicate under-reporting or reflect the inherently safe operation and use of the cellular telephone technology.

In an attempt to clarify this situation, a comprehensive analysis of crashes was executed using narrative data derived from police crash reports available in a North Carolina database. The analysis related crash incidence to the number of cellular telephones (as a surrogate for use while driving)¹ reported for each of several years. The models built from that data indicate a statistically reliable increase in crash incidence with increased numbers of cellular telephones over several years.

However, this analysis involved a small amount of data from a single state and required several assumptions that must be validated. Moreover, predictions may suffer if the future differs from the past in terms of substantial changes in product design, patterns of cellular telephone use, distribution of cellular telephone users, availability and use of other services, and so on.

Finally, several reasons are given for the potential of both under-reporting and over-reporting of cellular telephone involvement in the accident narratives that may influence the interpretation and prediction of trends. The analysis provides plausible but inconclusive evidence for a trend toward increased cellular telephone-related traffic mishaps as more and more drivers purchase such products and services.

The literature review of simulator, test track, and on-road studies of cellular telephone use while driving yielded the following findings: manual dialing can be disruptive of both vehicle control performance on the one hand and situational awareness and judgment on the other hand. The incidence and magnitude of vehicle control disruption while driving on public roads appears to be less than that encountered in driving simulators or on test tracks, but may nonetheless pose a safety concern. On-road studies indicate that if hands-free voice communications activities have any detrimental effects, they are on driver situational awareness and not on vehicle control performance.

The voice communications dialogue materials that have been used in this line of research often involve "intelligence test" type materials (e.g., mathematical computations) that may represent both extreme and atypical cognitive loads when compared to normal cellular telephone communications. In addition, all of these studies used voice communications that were free of emotional content (e.g., an argument with a spouse). Dialogues that involve substantial degrees of personal involvement may be even more disruptive than the cognitively challenging materials typically included in the human factors research (see Chapter 5).

Chapter 6 - Table of Contents

Document Table of Contents

6.3 Discussion

The impetus for this review was the relatively large number of public, media and congressional inquiries to NHTSA regarding the safety of using cellular telephones while driving. These inquiries were generally motivated by "close calls" experienced or observed by the public or by crashes involving cellular telephone users that were reported by the media.

While there are many sources of driver distraction that have been associated with increased risk of crashes, there has been a noticeable increase in attention to the safety of cellular telephone use while driving. This is not surprising, given the growing population of users and the ease with
which such use can be readily identified by other drivers. Thus, it may not be obvious to other drivers if one spills a soda or scolds a child while driving, but the novelty and position of hand-held cellular telephones can quickly attract attention, and the relatively long duration of the activity further increases the likelihood that it will be noticed by other drivers.

The consequent magnitude of public attention to cellular telephone use by drivers may therefore not truly reflect a problem of sufficient magnitude to require some form of intervention, but rather the obvious nature of the behavior and associated consequences for driving. While the information and data provided in this report have presented evidence to suggest that use of cellular telephones while driving can increase the risk of crashes from several standpoints, there is little data that would allow one to determine and characterize with precision the magnitude of the problem. The discussion below highlights some of the many issues that have been raised in this report.

Quality of Crash Data

This report highlights, on a number of occasions, the deficiencies in crash data relative to the involvement of cellular telephone use as a contributing or causal factor. The identified deficiencies have underscored not only the lack of focused and rigorous efforts in collecting relevant data, but also the care that must be exercised in interpreting the data, where it is available. Consider, for example, the North Carolina data (see Wierwille, Chapter 4).

For December of 1989, the Cellular Telecommunications Industry Association (CTIA) data indicate there were 3,508,944 cellular customers nationwide. For the same year, 12.5 ² cellular telephone related crashes were identified in the North Carolina data (13.2 adjusted, Chapter 4). In December 1994 there were 24,134,421 cellular customers, almost a seven fold increase. Extrapolating to 1994 on the basis of industry growth alone, one might expect levels of cellular related crashes to be about 87.5 crashes, all else being equal. However, North Carolina crash data indicate only 21 cellular telephone related crashes in 1994 (20.1 adjusted, Chapter 4), a much smaller, though still practically significant, increase.

This discrepancy highlights the complexity of the issues. There are many possible explanations for these findings. It may be that these crashes are being reported less frequently or are harder to identify than before, that more recent cellular telephones are safer to use, that drivers are learning to use cellular telephones more safely, or that drivers are finding use of cellular telephones so unsafe to use (a characteristic often associated with hand-held phones) that they are simply not using them as often while driving.

In addition, trends in usage, from business to personal, from fixed installations to hand-held, and an emphasis on applications to safety (e.g., reporting congestion, drunk drivers) influence the relative risks associated with using cellular telephones while driving. The trends also highlight the difficulty in defining the magnitude of the problem and in predicting the impact of future trends on the basis of incomplete or limited crash data.

This is illustrated again in Oklahoma (Chapter 3), which is the only state that includes cellular telephone presence and use on its crash reporting forms. Based on the data alone, it would appear that if a vehicle is involved in a crash and has a cellular telephone, there is about a one in ten chance the phone was in use at the time of the crash. However, based on uncertainties in reporting techniques, the data may not be reliable. For example, a vehicle may be reported to have a cellular telephone only if the investigating officer
sees it, in which case he would ask the driver or witnesses if it was in use at the time of the crash. If a hand-held cellular telephone was in the vehicle, it would not be reported unless it was visible.

Based on discussions with instructors at the Oklahoma State Police Training Academy, there are no strict guidelines for collecting this information, and it cannot be determined from the data whether a cellular telephone was being used at the time of the crash or was being used to report the crash. Hence, what appears to be an indication of a potential safety problem is likely a consequence of reporting deficiencies.

Cellular Telephone Industry Emphasis on Safety

While limitations in the available data and the fast pace of change in the industry make it difficult to establish whether a problem exists at a level requiring some form of intervention, it is clear that the nature of the tasks imposed by cellular telephone use as well as trends in technology and usage raise many legitimate safety concerns.

As discussed in Chapter 1, some states along with the cellular telephone industry itself have long recognized safety as an issue and have frequently focused their attention on enhancing the safety of cellular telephones through design enhancement and public information.

The following examples clearly illustrate these points. Durham Radio, Inc., states in one of its ads promoting safety enhancing accessories:

Likewise, the industry as a whole, through CTIA, has frequently demonstrated concern for public safety in the use of its technologies (See Table 6-1).

Similar concerns have also been raised at the state level. The State of California, for example, after an extensive review of the issue in 1987, recommended:

Cellular Telephone Safety Benefits

It is often argued that cellular telephones provide so many safety and highway travel benefits (e.g., emergency calls, reporting congestion) that to limit or restrict their use would be counterproductive. There is undoubtedly some truth to this argument. Nonetheless, it is somewhat tempered by the fact that such use is often carried out from a stopped or slowed vehicle (e.g., to report congestion from within a line of slowed traffic) and any restriction on use from a moving vehicle would have a minimal impact on safety or highway travel benefits.

A related issue, pointed out earlier in this report, is that the increasing availability of cellular telephones on the roadway has led to a dramatic increase in duplicate emergency "911" calls. In some localities this situation has resulted in a significant burden on response networks, given available resources. This situation not only may prevent other emergencies from being reported promptly, but such extensive use in these situations may also lead to a substantial increase in caller exposure under the inherently more hazardous conditions of stop-and-go traffic, abrupt changes in speed, and reduced lane availability

In view of the continued growth of the cellular industry, these problems are likely to increase significantly. Government agencies at the federal and state levels are currently exploring the means with which to deal with the multiple reporting issue.

Table 6-1: Cellular Phone Safe Driving Tips (Source: CTIA)

Safe driving is your first priority.
Always buckle up, keep your hands on the wheel and your eyes on the road.

Make sure that your phone is positioned where it is easy to see and easy to reach.
Be familiar with the operation of your phone, so that you're comfortable using it on the road.

Use a hands-free microphone while driving.
Make sure your phone is dealer-installed to get the best possible sound quality.

Use the speed dialing feature to program in frequently called numbers.
Then you can make a call by touching only two or three buttons. Most phones will store up to 99 numbers.

When dialing manually without the speed dialing feature, dial only when stopped.
If you can't stop, or pull over, dial a few digits, then survey traffic before completing the call. (Better yet, have a passenger dial.)

Never take notes while driving.
Pull off the road to jot something down; if it's a phone number, many mobile phones have an electronic scratchpad that allows you to key in a new number while having a conversation.

Let your wireless network's voice mail pick up your calls when it's inconvenient or unsafe to answer the car phone.
You can even use your voice mail to leave yourself reminders.

Be a cellular Samaritan.
Dialing 9-1-1 is a free call for cellular subscribers; use it to report crimes in progress or other potential life-threatening emergencies, accidents or drunk driving.

Use of Hands-Free Dialing to Address Safety Concerns

With the evolution of small, hand-held cellular telephones, there has been increasing concern for the ability of a driver to operate a vehicle safely with one hand while holding/manipulating the phone with the other. The tasks of searching for the phone, extending the antenna, accessing the display, dialing or simply holding the phone, along with the potential for dropping the phone have all been associated with increased risk of a crash. In this regard, the introduction of technology that permits hands-free dialing and conversing has been touted as a potential solution to mitigating the safety problems associated with cellular telephone use while driving.

Many third party suppliers are now providing conversion kits that allow older, fixed installations and hand-held cellular systems to be modified to hands-free use. The exact nature of a hands-free capability varies considerably, from one button dialing to voice activated control of both dialing and conversation, although the driver typically must take some manual action to initiate a call.

Future systems in development even include availability of phone book information using a head-up display projected on the windshield.

It should be noted that foreign laws restricting the use of cellular telephones in vehicles often restrict only the use of hand-held phones and specifically permit hands free operation (see Chapter 1). Similar provisions have been characteristic of domestic attempts at legislation (see Chapter 1 and Appendix A).

While the hands-free approach may at first seem like an obvious solution to cellular telephone related safety problems, it presumes that crashes caused by cellular telephone use result primarily from dialing, from having only one hand on the wheel, or from reaching for, holding or dropping a phone. Although these factors certainly contribute to the crash picture, the data from North Carolina as well as the NASS case studies suggest that conversation itself is the most prevalent single behavior associated with cellular telephone related crashes in the United States.

This is not surprising for several reasons. First, because conversing may take place over minutes while dialing typically takes place over seconds, the greatest exposure occurs while conversing. To put this into perspective, using the CTIA 1995 average call duration of 2.15 minutes, at 65 mph, this would translate to about 2.3 miles of roadway traversed for the average duration of a conversation. While having only one hand on the wheel may influence the ability of the driver to turn or respond appropriately to adverse situations created by use of the cellular telephone, this is not the only factor that would influence the outcome of an evasive maneuver.

Second, cellular conversation may hold drivers' attention (cognitive capture) over a more prolonged period, transforming what is typically characterized as a closed loop activity (i.e., driving) to an open loop activity ( i.e., lost in thought) where the driver is less likely to respond

appropriately to outside events. This phenomenon, though not unique to cellular telephone use, is suggested in some of the case studies reviewed where drivers have drifted off the road or into an adjacent lane.

Third, the emotional (i.e., personal involvement) or critical nature of conversation can be particularly distracting (e.g., a domestic argument, closing a deal, etc.) and is also highlighted in case studies as a causal factor.

Finally, as pointed out earlier, the driver is not fully in control of the conversation since the party at the other end has no way of knowing the traffic situation and can't adapt the conversation accordingly (see discussion below). The Japanese (1996) findings that 42 percent of cellular telephone related crashes occurred in responding to calls, indicates that even a ringing phone can elicit inappropriate responses from some drivers (e.g., startle, or reaching/searching for a phone at an inopportune time), a finding that is consistent with some of the case studies presented in Chapter 3.

Understanding the relative contribution of behaviors associated with cellular telephone use to crashes is important in evaluating the potential for successful intervention, but this is not the whole story. In the discussion of the Prevention Magazine survey data (Chapter 2) it was pointed out that the majority of cellular telephone users do not regularly use the phone while driving and many who do, find cellular telephone use as distracting or more distracting than tuning a radio.

Drivers, however, might readily adjust their behavior when the perceived risk changes, a phenomenon sometimes referred to as risk adaptation or behavioral adaptation (OECD, 1990). Thus, drivers who believe that a system is safer to use or has added safety benefits might adjust their behavior to accomodate the improvement in perceived safety. Where such changes in behavior are not consistent with actual improvements or where the margin of risk is adjusted to accommodate the new (perceived) capabilities, the net outcome may not be as expected.

In some circumstances, then, changes in systems and associated changes in behavior result in outcomes that are inconsistent with the intended goals. Such considerations may also apply to legislative actions. For example, as was pointed out in Chapter 1, Washington State has amended a Senate Bill to permit the use of an "approved" headphone for use with "hands-free wireless communications systems." While such use may facilitate communications, it may also introduce another manual task for the driver - for example, having to reach for and put on a headphone in response to an incoming call. Such an action may itself place the driver at risk.

Within the context of cellular telephones, a transition to hands-free operation will undoubtedly improve the safety for the individual user insofar as it will address the concerns associated with dialing, holding, reaching for, and dropping the phone, as well as steering with one hand. However, if we assume that the population of drivers willing to use the phone while driving now increases substantially because of the touted safety benefits of hands-free operation, individual cellular telephone use and perhaps duration of calls may increase.

To the extent that conversation itself is associated with a higher risk of crashes (relative to manual dialing), the intended safety benefits of hands-free operation may paradoxically increase exposure to distraction-induced crash hazards. Where hands-free architectures are legislatively mandated, such an outcome would likely take place over time. As the population of users transition to hands-free operation, a reduction in cellular telephone crashes would likely take place initially, since the majority of use from vehicles is not currently hands-free. Thus, in the long term, the outcome may be a net increase in total crashes across the population of users. While improving safety for the individual driver, the overall magnitude of the problem may, therefore be increased. This also serves to underscore the need to enhance the safe use of cellular telephones by drivers in a comprehensive way, i.e., by addressing all aspects of cellular telephone use. These points are again raised in the discussion of human factors considerations that follows later in this chapter.

Cellular Telephone vs. In-Vehicle Conversation

Comparisons between talking on the cellular telephone and conversing with a passenger in the car have been made frequently. It has been suggested or inferred that cellular telephone conversation is less than or no more disruptive of driving than in-vehicle conversation. However, data does

Figure 6-1 - Number of Crashes Distributed by Sources of Attentional Distraction Please click on the thumbnail to view larger version of the document.

indicate that a passenger in a vehicle can accommodate the conversation to the driving situation (e.g., stop conversing under high demand situations) (Parkes, 1991).

Unlike a caller (or answering machine) on the other end of a cellular telephone line, the passenger can see when the driver needs to focus on driving and can further serve to alert the driver to hazards. This suggests that future ITS technologies may have a potential role in not only alerting the driver to potentially hazardous situations, but also in alerting the individual at the other end of
the conversation. Nevertheless, the data presented in Figure 6.1 must be acknowledged. This figure illustrates the distribution of causal factors related to driver inattention found in the North Carolina data for 1989 (Tijerina, et al. 1995). The element "interaction with another person or animal in vehicle" when broken down further indicates that the specific acts of talking, listening, and arguing account for about 38 percent of the 210 incidents reported in the data.

A component of this may involve the act of turning towards and looking at the passenger, a behavior not characteristic of cellular telephone conversation. Hand-held cellular telephones nevertheless sometimes require the driver to change position in order to achieve better reception of the signal and ensure the connection is not lost. Thus, it would appear that the analogy between the two activities is not that straightforward.

This analysis also serves to highlight the potential risks associated with in-vehicle conversation of any kind, if pursued at inopportune times. Thus, development of means to address or mitigate the distraction potential of cellular telephone conversation, at least, appears worthwhile.

Implications of Future Trends and their Potential Impact on Safety

Over the past several years, there has been a progressive trend towards the integration and merging of function among what has typically been highly disparate technologies. Thus, the functions of voice communications, data communications, paging, automated collision notification (ACN), faxing, e-mail, navigation, vehicle position (Global Positioning System [GPS]), security, and safety have typically been identified by separate and distinct hardware and software components. Increasingly, the technological barriers have been narrowing and this has resulted in multifunctional systems that will provide a wide range of services using a single wireless device. Such an evolution has taken place within the cellular in dustry insofar as remote communications represent a critical element that can support a wide range of function.

Existing capabilities already reflect these trends. By linking cellular communications with fax machines and laptop computers, it is now possible to receive and transmit faxes, receive and send e-mail, and, in fact, "surf the net" from within a vehicle. While we do not have any indication of the extent of such usage, anecdotal information suggests that it is more common than might be expected, given the potential safety implications.

We are beginning to see crashes, such as in the North Carolina data, where drivers were using laptop computers while driving, and third party suppliers are now providing hardware for mounting laptop computers adjacent to the driver or, in some cases, right on the steering wheel (over the airbag) (see Appendix B). In the 1996 ( No. 2) issue of Inc. Technology, an article entitled "DWT (Driving While Typing)" describes how "work-on-the-road drivers" are mounting desks within their vehicles to enable them to phone, fax, e-mail, compute, and "put themselves at risk." Although manufacturers of such products warn drivers not to use the systems while the vehicle is in motion, based on observations of other "extreme" driver behavior (e.g., reading, shaving, and brushing teeth) the expectation is that some drivers will use them, regardless of the risk. [Note that some of the steering wheel mounted support brackets will not remain in place during driving when the steering wheel is rotated and thus cannot be used while the vehicle is in motion.]

The evolution of cellular technology is perhaps best dramatized by recent announcements of products now available or on the immediate horizon. The following excerpts illustrate the latest trends.

...next year will roll out Internet services for users of GSM [Global System for Mobile communications] based smart phones, offering customized travel and financial information, entertainment and electronic commerce capabilities, magazine titles, and other content ...

From PCWEEK, July 22,1996*

In addition to digital voice capabilities, the unit enables mobile users to send and receive faxes, e-mail, and short messages, as well as access the Internet and corporate and public databases ...
users can maintain a conversation while viewing documents on the screen or launching applications.

From PCWEEK, March 25, 1996*

Three Swedish companies are developing wireless data transmission technology that enables mobile users to conduct video conferences and gain high speed access to the Internet while on the road.

From PCWEEK, June 3, 1996*

Other related technologies are also evident.

GOANYWHERE [a modem] combines a packet radio modem and a conventional data/fax modem in a Type 2 PC Card.

From PCWEEK, July 22, 1996*

While it is unlikely that current drivers will use the capabilities offered by these integrated technologies to any great degree, given their relatively high projected cost, it may be only a matter of time before such capabilities are generally available at affordable pricing. Furthermore, current trends in the automotive industry, along with efforts supporting Intelligent Transportation System (ITS)

Adapted from T. Ross and G. Burnett, "The Right Road to Take," ITS International, June, 1996, prepared under Project V1037 STAMMI (CEC DGXIII DRIVE Programme)

initiatives, have highlighted the potential importance of cellular technology to various programs. These include incorporation of portable cellular telephone interfaces in vehicles (to achieve universal hands free operation), automated collision notification (ACN), in-vehicle information systems (IVIS) as well as a host of systems supporting heavy vehicle operations.

Thus, cellular capabilities may increasingly become integral to both the automobile and commercial truck fleets to support various functions other than voice communication. Such integration with in-vehicle systems and, in particular, crash avoidance technologies, may eventually lead to "intelligent" or "cooperative" systems that are responsive to lapses in driver attention and would provide appropriate warnings or control.

Comprehensive efforts at improving highway safety may thus address some of the concerns associated with the use of cellular telephones. Nevertheless, with the addition of new technologies and available services, there will likely be an associated increase in driver workload. Such an increase may itself create new safety concerns and make voice communications even more challenging. NHTSA is particularly concerned about possible synergistec effects of using multiple technologies that may increase workload beyond acceptable levels.

Human Factors (Ergonomic) Design Considerations

In surveying wireless technologies, it became apparent that there were extensive differences between the various wireless communications devices in terms of design features that could influence ease-of-use and safety. These "human factors" aspects of the systems in use go far beyond the issue of hands-free operation and how it is implemented. Rather they encompass specific design considerations related to the display, controls, size, shape, location and other aspects of the systems.

It is suggested here that industry attention to them may offer significant benefits in reducing risk associated with use of cellular telephone systems from a moving vehicle. The trends towards miniaturization (with some future systems projected to weigh as little as 3 ounces), small keypads, miniature displays and increased services, clearly have the potential to place greater demands on the driver using such systems; improvements to design may be capable of reducing such demands.

While the above considerations are very important, they must be viewed within the context of overall safety. As pointed out earlier, enhanced cellular telephone ease-of-use may promote greater frequency of use as a by-product. Current cellular telephone users and limited-use drivers may feel more secure using a hands-free phone over a hand-held unit, for instance, and consequently increase their use while driving.

Others who may not be inclined to use a cellular telephone at all from a moving vehicle may now be willing to do so if they believe it is safer. The consequent increase in use among the driving public can therefore increase overall crash hazard exposure. Thus, while hands-free operation reduces or eliminates the demands of manipulation, more drivers may now be engaged in conversation, which has been shown to be distracting in itself. Facilitating use through other improvements to human factors design and implementation of wireless systems may influence exposure similarly. This type of effect has precedents in traffic safety and driver behavior which may be understood in terms of human behavior feedback or behavioral adaptation.

Evans (1991) has written a thoughtful review of driver responses to interventions that might influence traffic safety. Evans addressed such varied systems as crashworthiness enhancements, studded tires, changes in speed limits, anti-lock brakes, and so forth. The review indicated that safety may increase, remain unchanged, or decrease in sometimes perverse ways. Evans concludes that human behavior feedback or reaction to safety systems or safety-related enhancements may greatly alter safety outcomes from what is expected. A general pattern that appears is that safety change effects that noticeably improve vehicle performance will probably increase mobility by way of increased speeds, closer car following, faster cornering, and the like. Safety may also increase, Evans points out, but by less than if there had been no behavioral response.

Ergonomic enhancements to cellular telephone design and implementation may likewise induce a sense of security or safety that is not justified relative to compensatory changes in driving behavior. Thus, there is a legitimate concern that safety benefits from human factors design considerations may be less than expected. This does not mean that such human factors intervention is counterproductive, but rather that such involvement should be comprehensive and include after-market evaluations and longitudinal studies to gauge the effects increased ease-of-use has wrought.

The implications of these design issues and the need to understand their potential influence on safety are, in fact, called out in the recommendations that follow. Appendix F presents a taxonomy of human factors considerations that have been identified by the authors as having a potential influence on the ease with which these wireless communication devices can be used.

Secondary Safety Issues that May "Impact" the Driver

In preparing this report it became apparent that there were safety issues that extended beyond the primary concern of the influence of cellular telephone use on the ability of individuals to drive safely. These issues concern crashworthiness related to the position of installed equipment and the use of hand-held cellular telephones within the context of airbag deployment. This is an issue of considerable interest to NHTSA insofar as objects in the path of a deploying airbag can become injurious, potentially lethal projectiles or objects of impact.

In the most extreme cases, laptop computers (often used in conjunction with wireless technology) have been mounted on the steering column directly over the airbag, and have been configured to remain folded open while the vehicle is in motion. While the potential danger of such an installation is obvious, there may be other, less evident installations that pose a similar danger. The public and industry should be sensitized to this issue to ensure that equipment is not positioned to interfere with airbag deployment. These concerns are equally relevant to the holding of a cellular telephone while driving, where the proximity of the phone to the face and head, or placement in front of the steering wheel during use (e.g., for dialing), is also of concern. In this regard, the use of hands-free cellular telephones mounted on the console should be encouraged.

Thus far, there is no data available to suggest that cellular telephones may play a role in airbag related injuries, but this may be a consequence of the relatively small number of cellular telephone related crashes that have been evaluated in depth. The collection of such data is addressed in the recommendations that follow.

Society, New Technology and Perception of Risk

At the beginning of this report a 67-year-old quote (Nicholas Trott, 1930) was provided to illustrate a societal dilemma that has been with us since the Industrial Revolution, that is, the consequent risks to personal safety associated with the use of new technologies. The concerns about the use of the radio while driving, balanced against claimed benefits and comparisons to other in-vehicle distractions, are strikingly similar to what we are faced with today with wireless communications in vehicles.

Interestingly, as indicated in data from North Carolina (Wierwille and Tijerina, 1995), radio use or tuning is a common factor associated with crashes related to in-vehicle distraction, although the true extent of this causal factor at a national level also remains elusive. Nevertheless, while the early concerns have been borne out, at least in principle, there does not appear to be an epidemic of crashes related to operation of the radio. Indeed, drivers appear to be aware of the risks associated with distraction in general, and the survey data (see Chapter 2) clearly suggests drivers will frequently adjust or temper their use of the cellular telephone because of these concerns.

This is not to say that use of the cellular telephone is directly analogous to the radio since, as has been pointed out earlier, there are significant differences. Rather, it highlights an acceptance of some degree of risk associated with the use of technology and the willingness of most drivers to adjust their behavior accordingly. It is when perceptions are inaccurate, motivations are misguided, or the timing of coincident events are inopportune, that drivers appear to run into trouble

Chapter 6 - Table of Contents

Document Table of Contents

6.4 Conclusions

What conclusions can be drawn, given the data reviewed in this report? The cogency of a conclusion depends on the adequacy of evidence, the degree to which the conclusions drawn logically follow from the evidence, and the degree to which no relevant information has been omitted from consideration. These three points will be considered for each of the following key questions:

 
• · Does use of cellular telephone technology while driving increase the risk of a crash?
• · What is the magnitude of the traffic safety problem related to cellular telephone use while driving?
• · Will crashes likely increase with increasing numbers of users of cellular telephone technology in the fleet?
• · What are the options for enhancing the safe use of cellular telephones by drivers?

Does cellular telephone use while driving increase the risk of a crash?

The available evidence is adequate to support the conclusion that the answer to this question is "Yes," at least in isolated cases. The conclusion appears reasonably plausible, particularly in light of the trends in the data, the growing complexity of the technology, and the inherent distraction potential of using such devices from a moving vehicle. What remains unknown is the relative contribution of cellular phone use, per se, and charactertistics of the involved drivers (e.g., less capacity to time-share attention between cellular telephone use and driving tasks, greater propensity for risk taking, fatigue).

What is the magnitude of the traffic safety problem related to cellular telephone use while driving?

The data reviewed here are inconclusive as to the magnitude of the problem. Cellular telephone use while driving is currently inadequately reported in crash records. As a result, the data that could serve as a basis for determining the magnitude of the crash problem do not exist. The lack of data cannot be interpreted to mean that there is no problem of sufficient magnitude to warrant action. The trends in the available data reviewed in this report, the growing complexity of the technology and the sensitivity of political and societal considerations, only serve to reinforce the need to collect more comprehensive and accurate data. In the recommendations that follow various approaches are proposed for enhancing the availability and quality of the data to support a more accurate determination of the magnitude of the problem.

Will crashes likely increase with increasing numbers of cellular telephones in the fleet?

Again, the answer is "Yes", if the North Carolina data and modeling results are any indication. But, the adequacy of that data and modeling results are modest at best. The logical strength of the statistical predictions depends on the representativeness of the data sample to the country as a whole and the adequacy of assumptions behind the model (e.g., national cellular telephone counts are a valid surrogate for frequency of cellular telephone use while driving). Extrapolation from statistical models assumes that the future will be like the past. It is evident that cellular telephone designs are evolving and cellular telephone usage patterns will change over time.

The ultimate impact of these changes on crashes cannot be predicted with great confidence. Thus, the answer to the question is less cogent than the answer given to the first question, and has been duly qualified in this report. Nonetheless, it logically follows that if more cellular telephones are in use, then there will be more opportunity for distraction and, hence, there will likely be an increase in related crashes - unless, of course, changes take place in the technology or its use that mitigates such a trend.

What are the options for enhancing the safe use of cellular telephones by drivers?

People in general are finding it harder and harder to keep up with all of the tasks and activities for which they are responsible. American motorists in particular spend substantial amounts of their day in automobiles, vans, trucks, and buses. It is not surprising that people will attempt to optimize their time in the vehicle by doing other things. It is unrealistic and ill-advised to suppose that drivers should have no advanced in-vehicle information systems at their disposal. Goals, then, should include making in-vehicle information systems, including cellular telephone technology, as compatible with safe driving as the state-of-the-art allows through the application of good engineering and human factors design practice, and educating drivers about potential risks associated with using this technology while driving. This must be done while addressing possible adverse safety implications for the population as a whole.

This report has highlighted a number of potential problem areas and issues. In the sections that follow, these problems and issues are identified and options are presented for responding to them.

Chapter 6 - Table of Contents

Document Table of Contents

6.5 Recommendations

The National Highway Traffic Safety Administration's mission is to save lives, prevent injuries, and reduce traffic-related health care and other economic costs. The Agency develops, promotes, and implements effective educational, engineering (including human factors engineering), and enforcement programs to prevent or mitigate motor vehicle crashes, injuries and fatalities, and reduce associated economic costs. This mission is accomplished through regulation, enforcement, economic incentives, educational programs, basic and applied research, and technology demonstration programs. It is therefore appropriate that this report conclude with a set of recommendations promoting safety when cellular telephone technology is used by drivers. Responsibility for implementing the various options should be shared jointly by various agencies of the federal government, state governments, industry and the private sector, both in the U.S. and worldwide.

It is also important to recognize the ongoing interest and research efforts of government and private institutions both in the U.S. and throughout the world. It is highly recommended that the results of these and future efforts be subject to peer review and shared within professional research forums. This will promote the development of a valid knowledge base.

Improved Data Collection and Reporting

There is clearly inadequate reporting of crashes that may be related to cellular telephone use while driving. To address this inadequacy, two proposals for improved tracking of cellular-phone-related crashes and near-misses are described below. The first would provide for better tracking of crashes. The second would provide one means of characterizing near-miss behavior and its frequency of occurrence.

Crash Data Collection and "Police Crash Reporting" - Only two States currently attempt to record the use of a cellular telephone during a crash on a Statewide and systematic basis. Oklahoma attempts to record both the presence and use of a cellular telephone in relation to a crash. Minnesota records only cellular telephone use as a crash contributor. A more aggressive approach (as described below) should be taken throughout the country to identify and systematically record cellular telephone use as a causal factor in crashes.

As a first step, States and local jurisdictions should be encouraged, as a part of their normal crash investigation process, to implement procedures that would help better identify and describe inattention or distraction-related crashes in general, and cellular telephone-related crashes in particular. It is also recommended that a model approach be developed to achieve uniformity in data collection and a nationally representative data sample.

In this regard, NHTSA and the FHWA are now in the process of developing a minimum standard crash data set. The data elements are expected to exceed those currently contained in Critical Automated Data Reporting Elements (CADRE) and could include one or more elements relating to cellular telephone use. The elements are being developed through a task force of experts from the traffic records community, who have been selected by NHTSA and the FHWA. Once the minimum standard crash data elements are developed, States and local jurisdictions would be free to adopt all or some of the elements, although they will not be required to do so.

It is further recommended that a study be carried out to supplement normal crash reporting with focused data collection in selected jurisdictions. This effort would be designed to provide in-depth information on the possible role of cellular telephone technologies in precipitating crashes. Police officers and crash investigators would be trained to conduct careful inspections of vehicles after crashes, and to probe for phone use during interviews with drivers and witnesses. When it is believed that a cellular telephone may have been in use at the time of the crash, cooperation with the cellular and PCS carriers would be sought for verification.

As a part of its own crash data collection activity, NHTSA has plans to continue to collect "pre-crash factor" information as part of its ongoing Fatal Analysis Reporting System (FARS) and National Automotive Sampling System (NASS) data collection efforts. Because the problem of crash under-reporting is not likely to diminish in the near term, every attempt should be made to supplement crash reporting using other approaches. For example, NHTSA and the FHWA could expand other ongoing research efforts, such as field studies that involve fleets, or NHTSA and the FHWA could include in the telephone surveys that they conduct, questions that address the use of cellular telephones and crash involvement. Respondents might be more candid about their use of a cellular telephone at the time of a crash in an anonymous telephone survey than to law enforcement officials at the crash scene. In this regard, efforts will also continue to address relevant cellular telephone issues in the Motor Vehicle Occupant Safety Survey carried out periodically for NHTSA (see Chapter 3).

Finally, it is recommended that the insurance industry cooperate with NHTSA in information sharing. The insurance industry is in an excellent position to monitor cellular telephone related crashes and preliminary discussions with members of the insurance industry indicate that the industry is already doing so.

Law Enforcement Observations - Members of the law enforcement community have stated that they routinely observe driver actions before issuing citations for speeding and other moving violations. There are instances cited in Chapter 2 when law enforcement officers reported they have pulled cellular telephone users over and issued them a warning because they were observed being inattentive to their driving. Within this context, law enforcement officers should be encouraged to record on the warnings or citations that they issue whether the driver was observed using a cellular telephone.

For example, officers could include a notation that there was a "cellular telephone in use" on police warnings and citations for inattention or reckless operation of a vehicle. Specific details regarding such data collection would have to be developed within the constraints of the various jurisdictions involved. This data could be useful to states in assessing the magnitude and nature of the problem.

Information regarding "near misses" or "other distraction indicators" would also be useful to provide a broader perspective on specific behaviors exhibited by drivers using cellular telephones. This work would help determine how drivers compensate when using a cellular telephone (e.g., by slowing down, or increasing headways), and could promote a better understanding of the safety implications of such behaviors.

Improved Consumer Education

Educational materials should be developed and promoted that focus on the various ways that distraction in general, and cellular telephones in particular, can increase the risk of crashes. The intention would be to make these materials available in driver education, licensing and cellular telephone sales facilities, or through companies that provide services or products to cellular telephone users. Such programs could inform drivers of the subtle influences of cellular telephone use while driving (e.g., loss of situational awareness even though lanekeeping is good). The programs could illustrate situations in which cellular telephone use should be avoided or minimized (e.g., high-traffic density, or negotiating intersections or turns).

Appropriate education could even address cellular telephone etiquette that provides coaching on how to politely refuse, postpone, or abruptly halt a conversation when driving conditions demand it. There may be a place in such educational materials to advocate the view that cellular telephone use while driving should be reserved for short calls that are urgent. Lengthier calls should be made while the vehicle is stationary and safely off the roadway. Such materials may also educate the public on product design and implementation considerations when purchasing wireless technologies so as to sensitize the user to features of these devices that will minimize their distraction potential.

Technology Evaluation and Monitoring

Concern has been expressed within the highway safety community regarding the potential safety implications of drivers attempting to use computers, faxes, and multifunction cellular telephones while driving. Such uses have the potential to increase driver workload far beyond acceptable levels and greatly increase the risk of crashes. The exploding market for add-on equipment, such as cellular compatible fax machines and portable personal computers with modem capabilities, along with new multifunction cellular systems, demand periodic, careful, human factors evaluation to better understand their potential for having an adverse influence on driving. Such research should seek to identify design solutions that minimize driver distraction as well.

Improved Cellular Telephone Research and Development

There are several areas of research and development that could be pursued to improve the safety of cellular telephone use. Human factors studies, for example, could be conducted to provide a better understanding of the risks associated with use of cellular telephones while driving. In addition, research and development efforts could result in cellular telephone technology or designs that are more ergonomically sound and safety-conscious.

Behavioral Research - "Naturalistic" Cellular Telephone Conversations - It was mentioned in earlier chapters that human factors studies published to date suffer from a lack of information with which
to structure realistic conversational materials. What is needed is research into the duration, content, and placement of personal and business calls while driving. Such information would provide an empirical basis with which to replace "intelligence test" type voice communications test materials of arbitrary length with materials drawn from the real world. Given the indications that voice communications may pose a greater safety hazard in real driving than manual dialing or other manual tasks, it is highly recommended that this research effort be pursued at the earliest opportunity.

Behavioral Research - "Naturalistic" Cellular Telephone Behavior and Performance - One of the major criticisms of existing cellular telephone research is the artificiality of the experimental situations created by research methodologies and hardware constraints. The use of simulators, directed tasks, and presence of experimenters, to name a few, may greatly influence the outcomes of the research.

One way to improve the validity of the data is to instrument cellular telephone users' own vehicles and monitor their behavior over an extended period of time. The data collected would not only help identify the specific behavioral and performance effects of accessing the phone, dialing, conversing, and responding to calls, but would also allow identification of "incidents" or "near misses" involving the use of the cellular telephone as well as allow an evaluation of different designs and configurations in a realistic, in-situ setting. Because of the importance of such research to all parties, it is highly recommended that it be pursued in the immediate future.

Design Research - Intelligent Answerphone - Parkes (1993) introduced the concept of an "intelligent answerphone" into the literature of cellular telephone ergonomics. Such a system would divert, record, and interrupt messages appropriately based on sensed driving conditions. The development of such a system goes far beyond anything the cellular telephone industry has marketed or reported on to date. It nonetheless is a laudable design goal and is compatible with evolving technologies and concepts that are part of the Intelligent Transportation System (ITS) initiative.

Design Research - Workload Reducing Features for Cellular Telephone Design - Some types of cellular telephones appear easier or harder to use while driving than others based upon size, shape, configuration, visual display quality, location in the vehicle, and optional features such as speed dialing. Continuing movement towards miniaturization has the potential to place an input-output burden on drivers through the use of small keypads, multi-line displays (i.e., small text) and various presentation formats (e.g., text scrolling). (see Appendix F)

Cellular telephones have attributes very similar to many emerging Intelligent Transportation System (ITS) technologies. As all these technologies are implemented, it will be necessary to ensure that designs and implementation strategies are optimized to minimize driver workload and distraction. It is thus recommended that design features and operational characteristics be identified that promote safe use of cellular telephones when used alone or along with ITS and other in-vehicle technologies.

It is recommended that these studies employ instrumented vehicles (e.g., DASCAR, Micro-DAS - NHTSA's suite of in-vehicle instrumentation for crash avoidance research) and high fidelity simulators (e.g., NADs) to collect measures such as those described in NHTSA's workload evaluation protocol document (Tijerina, et al., 1995) to identify safer design features, methods of implementation, and strategies for use. This information would be very useful for designers of cellular telephone systems and purchasers of such systems and could be used to support development of educational programs and literature.

Design Research - Cellular Telephone Technology and Intelligent Transportation Systems: Integrated Systems Research - In addition to the Intelligent Answerphone concept, it is recommended that cellular telephone technology be more fully and explicitly integrated with other aspects of Intelligent Transportation Systems (ITS). For example, cellular telephone technology, when used in conjunction with crash avoidance system (CAS) technology may promote greater levels of driver comfort, satisfaction, and safety. On the other hand, CAS technology, route guidance systems, and cellular telephone technology all together may create unacceptably high levels of driver stress or distraction unless integrated to minimize such effects. The vehicles of the 21st century will likely be substantially more sophisticated than those of the 20th century. A focus on driver-centered and safety-conscious design should promote the best possible technological evolution.

Crashworthiness Considerations in Cellular Telephone Placement

Although installed "mobile" or "car phones" generally are placed in the center console, next to the driver, the majority of cellular telephones are hand-held units that are sometimes placed in mounting brackets. These brackets may be located on the instrument panel, floorboard, or windshield. In addition, manufacturers have developed floor-, steering wheel-, and console-mounted desk-type devices that are used for mounting or supporting computers, fax machines, and other in-vehicle equipment that can be interfaced with a cellular telephone.

The size and placement of these support devices can sometimes interfere with vehicle safety equipment such as airbags. The devices may also contribute to driver injury by becoming projectiles during a crash. Therefore, it is recommended that educational campaigns be initiated for users, installers, and manufacturers to point out the possible hazards of inappropriate placement. Furthermore, educational programs should underscore the fact that placement of such equipment in a manner that interferes with the operation of federally required safety equipment is not only ill-advised, but may be illegal.

Emergency Medical Services

Emergency service facilities can be overwhelmed when multiple calls are received for the same incident (sometimes exceeding 100 calls). This can delay timely notification regarding other emergencies. It is thus recommended that appropriate state and federal agencies, representatives of the Cellular Telecommunications Industry Association (CTIA) and other wireless associations, as well as other national organizations representing EMS, examine and evaluate potential solutions to this problem. It is highly likely that this problem will be exacerbated by the predicted increase in cellular telephone subscriptions into the next century. Discussions with the cellular industry regarding this issue already have been initiated and cooperative efforts currently are under way.

A number of states have developed specific emergency phone numbers to be used exclusively by cellular subscribers. Examples include "#77" and "*FTP", which are intended to be used to report highway emergencies. It is recommended that a nationwide standard emergency number be created so that travelers would always have access to a unique cellular telephone emergency number regardless of their location. The Federal Communications Commission (FCC) has taken a first step toward this goal by the adoption of rulemaking to link cellular emergency calls to the existing 911 landline responders. With the cooperation of the states, and the cellular telecommunications industry, it should be possible to have new cellular telephones preprogrammed so that the driver would only have to press a designated button to summon help anywhere in the U.S.

Enforcement and Legislative Options

Laws limiting the use of cellular telephones while driving have been enacted in a number of countries throughout the world. It is unlawful to drive recklessly in every State in the United States and, in a number of states, laws specifically prohibit careless or inattentive driving. While attempts have been made to enact laws limiting the use of cellular telephones while driving in some States in the U.S., none thus far has been successful. (In Washington state, however, the motor vehicle code (Title 46, Chapter 46.37) was amended to specifically permit use of "approved" headphones by motorists "using hands-free, wireless communications systems," which may be viewed as promoting the use of hands-free cellular telephones and prohibiting the use of hand held or other unapproved systems.)

States are encouraged to actively enforce their reckless and inattentive driving laws and states without inattentive driving laws should consider enacting such provisions. When law enforcement officers observe reckless or inattentive driving that is associated with the use of cellular telephones, this should be noted in the officer's report or on the citation. Similarly, where a crash occurs, it should be noted on the police crash report whether a cellular telephone was in use during or prior to the crash. This information could be useful in allowing states to assess the magnitude and nature of the problem of cellular telephone use by drivers.

Legislative proposals that have been introduced in the States have focused primarily on prohibiting the use of cellular telephones that require drivers to use their hands to operate or hold the phone. These proposals generally permit the use of hands-free models. The outcome of any restrictions or limitations, however, may not be as clear-cut as initially believed. For example, these legislative initiatives seem to be based on the assumption that hands-free cellular telephones phones are acceptable while driving, but hand-held phones are not. Properly implemented, hands-free designs should reduce the distractions associated with dialing, holding, reaching for, or picking up a dropped handset and allow the driver to keep both hands on the wheel. However, hands-free phones do nothing to mitigate the distraction potential of cellular telephone conversation.

Proposed legislation that prohibits only the use of hand held cellular telephones may, in fact, promote cellular telephone use (e.g., drivers may use their phone more frequently and for a longer duration) among limited users and non-users by suggesting that hands free use is safe. This could potentially increase "exposure" to any safety hazards that may exist. Thus, paradoxically, the outcome of legislation specifying hands free only usage, may be an increase in cellular telephone related crashes to the extent that conversation itself is a causal factor in crashes, a finding supported by this study ¹.

Considering the inconclusive nature of empirical evidence reviewed in this report on the magnitude of the cellular telephone-related highway safety problem, existing legislative initiatives may thus be inappropriate on technical grounds alone. For this reason, it is important to supplement any legislation with adequate data collection to monitor the impact of the legislation on relevant crashes.

Given the widespread implementation of foreign laws restricting the use of cellular telephones in moving motor vehicles, every effort should be made to examine the effectiveness of these laws, not only in terms of crashes, but also in terms of the influence such laws have had on the behavior of drivers in their choice and use of wireless technology. While the extent to which these laws have been or are being evaluated is unknown, it would be beneficial to identify any such efforts.

An effort should be initiated to examine the cost-benefit tradeoffs of legislative actions related to cellular telephone use while driving. Potential costs of unrestricted cellular telephone use may include those associated with distraction-induced crashes and degraded driving performance. Benefits of unrestricted cellular telephone use include more efficient use of commuting time, emergency service notification capability, and the conveniences attendant to closer communications with family, business, and community.

Costs of legislative restrictions may include more expensive sophisticated cellular equipment, restricted access while driving to otherwise desirable features, unforeseen secondary consequences (e.g., increased exposure to other safety hazards), and enforcement costs. Potential benefits of empirically grounded legislation would include savings in personal injury, property damage, and crash-caused congestion (delay) costs. An effort to codify and represent the costs and benefits of alternative legislative actions would support more informed decision making.

In view of the complexity of the issues discussed above, it is recommended that in considering legislation, states be encouraged to base their deliberations on all available research studies, empirical observations and data that are available to them, particularly with regard to the dynamic nature of the technology and the manner in which it is used. Only when such considerations are carefully evaluated can we be assured that the outcomes will be as intended.

Chapter 6 - Table of Contents

Document Table of Contents