Approximately 4,000 fatalities result from crashes involving trucks and buses in the United States each year. Although estimates are wide-ranging, 10 to 20 percent of these crashes may have involved fatigued drivers. The stresses associated with work as a commercial motor vehicle (CMV) driver (e.g., irregular schedules, economic pressures) and the lifestyle many of these drivers lead put them at substantial risk for insufficient sleep and the development of short- and long-term health problems. Sleep disorders such as obstructive sleep apnea (OSA), for example, appear to be common among many CMV drivers. OSA is a major contributor to driver fatigue, which in turn raises a driver’s risk for involvement in crashes. Moreover, it has become increasingly clear that drivers who regularly obtain insufficient sleep, whether as a result of irregular work patterns, sleep disorders, or other reasons, are likely at increased risk for a number of serious long-term health problems.
To address this problem, the Federal Motor Carrier Safety Administration (FMCSA) asked the National Academies of Sciences, Engineering, and Medicine to convene the Panel on Research Methodologies and Statistical Approaches to Understanding Driver Fatigue Factors in Motor Carrier Safety and Driver Health. The panel was charged with providing advice on additional data collection and analytic techniques with the potential to support a more comprehensive understanding of the links between operator fatigue and highway safety and between fatigue and long-term health problems, such as cardiovascular diseases. Specifically, the charge to the panel was to “assess the state of knowledge about
the relationship of factors such as hours of driving, hours on duty, and periods of rest to the fatigue experienced by truck and bus drivers while driving and the implications for the safe operation of their vehicles. The panel will also assess the relationship of these factors to drivers’ health over the longer term. It will identify improvements in data and research methods that can lead to better understanding in both areas.”
FMCSA’s mission is to “reduce crashes, injuries, and fatalities involving large trucks and buses.” The agency works toward this goal in at least three ways. First, a major policy lever at FMCSA’s disposal is that it issues and administers hours-of-service (HOS) regulations for truck and bus drivers that specify the maximum number of hours they can work in a day and in a workweek. The hope is that if they drive limited hours, drivers will have enough time to obtain adequate sleep between work shifts, and therefore will be more alert and less fatigued while driving. As a result, it is believed, the risk of crashes will be lower.
Second, FMCSA is responsible for the medical certification of CMV drivers through the National Registry of Certified Medical Examiners (NRCME). Members of the NRCME examine CMV operators at least every 2 years to determine whether they meet FMCSA’s medical standards.
Third, Transport Canada, FMCSA, trucking industry trade associations, and other agencies developed the North American Fatigue Management Program (NAFMP), an Internet-based online educational program that informs drivers, their employers, and anyone involved in commercial carrier operations about the causes of driver fatigue, the increased risk of crashes due to fatigue, the long-term health consequences of CMV driving (such as regular sleep insufficiency), and, most important, suggests countermeasures that can be used to manage driver fatigue.
The specifics of these three FMCSA programs are based on the current scientific understanding of operator fatigue, its causes, and its consequences. A considerable amount of research has been conducted to clarify the relationship among hours of service, driver fatigue, and crash risk, as well as the relationship between fatigue and long-term health and wellness. As yet, knowledge of these relationships is not comprehensive, and the relationships themselves may be changing. However, the quality and quantity of the information available on these relationships is constantly increasing. In particular, new in-vehicle technologies and roadway improvements are regularly being applied to promote safety, and several of these innovations allow for the capture of important data that could be used to better inform policy and/or improve driving procedures. Addi-
tionally, recent advances in statistical methods could be applied to make better practical and research use of data that either are currently collected or could be collected if targeted for research purposes.
HOS regulations can only limit hours spent driving and working; they cannot mandate rest, so they inherently cannot ensure, by themselves, that drivers will be well rested and alert. Therefore, it is not straightforward to determine how additional modifications of the current HOS regulations would result in more or less fatigue in CMV drivers that might, respectively, raise or lower crash risk.
In addition, driver fatigue obviously is not the only cause of highway crashes. Crash risk factors can be grouped into at least four main types: driver characteristics; truck or bus characteristics; factors stemming from employment circumstances, especially with respect to scheduling and work assignments; and the physical environments encountered while driving. The variety of factors from which a crash can result, acting either solely or in combination, makes it challenging to develop an understanding of the nature of an individual risk factor—in the present case fatigue—since there are so many confounders that are difficult to control or otherwise account for in analyses. Therefore, the study designs and analytic tools used in such research are critically important.
A further complication is that fatigue is very difficult to define and therefore to measure objectively. If fatigue is loosely defined as the inability to sustain performance over time, under such a vague definition, it is not directly measurable. Therefore, it is somewhat difficult to assess fatigue, and thus to regulate how to avoid driving while fatigued. As a result, researchers and policy makers must instead assess how hours of service affect various components of fatigue, many of which are measurable. One of the most important, objectively measureable components of fatigue is “drowsiness” or lack of alertness.
To add to this complexity, the commercial truck and bus industries are highly heterogeneous, with a great variety of types of employment, methods for compensation, and so on. As a result of their specific type of employment, many CMV drivers are affected in different ways, and sometimes not at all, by changes in HOS regulations. Therefore, stratifying, or otherwise accounting for, the type of employment in some way is important when carrying out research in this area.
Adding still further to this complexity is the difficulty of assessing some of the primary inputs—sleep duration, hours of service, level of driver alertness—and the primary output—rates of crashes due to fatigue. Deriving satisfactory measures of the amount or quality of sleep obtained by a truck or bus driver during the previous night or nights is difficult because self-reports often are not highly reliable and because attempts to directly capture measures of drivers’ sleep are invasive in
nature. Also, it has been shown that paper logs recording hours of service for CMV drivers are often inaccurate. In crash reporting databases, moreover, the primary judges of the factors involved in a highway crash are police officers, who must make such assessments after the fact, with little information to go on. Partly as a result of these difficulties, the available crash reporting databases often provide a paucity of information on sleep deficiency in CMV drivers, their adherence to HOS regulations, and their crash frequency as a result of fatigued driving. Therefore, research on the linkage among hours of service, fatigue, and accident frequency is hampered by imperfect knowledge of the three most central variables.
THE RESEARCH QUESTIONS MOST ESSENTIAL TO FMCSA
Any newly proposed changes to the HOS regulations for CMV drivers, to the process for drivers’ medical certification, and to the NAFMP need to be based on research-supported understanding of the costs and benefits of such changes. The following list represents the panel’s attempt to articulate questions that, if answered satisfactorily, should assist FMCSA in understanding the costs and benefits of existing and proposed changes to its policies and regulations in these areas:
- How much sleep do typical CMV drivers need to maintain suitable sustained levels of alertness and to avoid being drowsy to the point of driving while impaired?
- To what extent would any proposed change in HOS regulations affect the amount of sleep obtained by CMV drivers in different industry sectors?
- What degree of hypopnea (severity level of OSA) results in enough sleep loss to increase the risk of crashes for CMV drivers?
- To what extent does regular use of positive airway pressure (PAP) and related OSA treatment technologies and measures mitigate that increased risk?
- To what extent are various collision avoidance and driver fatigue alert technologies (both in-vehicle technologies and infrastructure measures such as roadway rumble strips) useful for reducing the risk of crashes?
- What substances, if any, reduce impairment due to sleep insufficiency?
- To what extent is chronic sleep deprivation related to an increased risk of developing health threats or various medical conditions?
- To what extent do CMV drivers, their employers, corporate officials, fleet supervisors, safety and risk managers, and drivers’ families make use of the NAFMP materials on the Internet?
- To what extent do fatigue awareness training and fatigue management initiatives result in behavioral improvements in CMV drivers?
Considerable progress has been made toward answering many of these questions using techniques ranging from laboratory observation and driving simulators for focused comparisons; to various kinds of epidemiological techniques, especially case-control and cohort studies; to naturalistic driving studies assessing what happens in the field. Laboratory studies can enable comparison of specific treatments or interventions in highly specialized environments; however, the necessary extrapolation from laboratory observation to field implementation is not always straightforward. Epidemiological studies and naturalistic driving studies, and even summary crash data, can be highly informative about what is happening in the field; however, these studies can be subject to confounding influences when attempts are made to compare interventions or treatments.
SUMMARY OF WHAT IS KNOWN
What Causes Driver Fatigue?
Fatigue-related performance decrements are a complex function of several factors, including lengthy time on task (i.e., long drives), extended wakefulness or acute sleep deprivation, chronic insufficient sleep, and poor-quality sleep. Some of these factors are attributable to irregular work schedules, nighttime work, and misalignment of circadian phase. Other possible contributors to driver fatigue include the presence of sleep disorders, such as OSA, and other medical disorders, and even use of sedating medications. However, the extent to which each of the above individual factors, or the interaction among them, adversely affects performance is unknown.
Why Do Drivers Continue to Drive When They Are at Risk of Fatigue?
Drivers get inadequate sleep for several reasons. CMV driving is an essential part of the just-in-time delivery economy, so the demand for such workers is increasing. To understand why CMV drivers may continue to drive when sufficiently drowsy to be impaired, one must consider the form of their work compensation structure and many other confound-
ing influences, including family pressures and commuting patterns. The factors involved in decision making are relatively well known but not their specific individual contributions.
What Is the Relationship Between Sleep Deficits and Decreased Driver Alertness?
Research has shown that insufficient sleep leads to decreased alertness and eventually to performance decrements. Performance decrements can lead in turn to driver errors or inappropriate driving practices, which then can lead to crashes. These errors and inappropriate practices usually are a function of slowed reaction times, attention failures, and poor decision making.
What Are Practicable Ways of Reducing or Eliminating Driver Fatigue?
Various fatigue countermeasures are commonly attempted, some that have been tried and proven to work, and others that yield only fleeting effects and are not as helpful. Among those that work best are adhering to work-rest scheduling that permits sufficient sleep, driving primarily during the daytime rather than at night, being cognizant of the two anticipated circadian lulls of the 24-hour day, obtaining sleep immediately prior to a long trip, planning to take and taking periodic breaks from driving during trips, and inserting planned naps into a trip plan. While consuming caffeine can provide temporary relief, and rumble strips can serve to alert drivers that they are likely falling asleep, these measures really provide only temporary assistance. The only way to reduce the need for sleep before going on duty and thus to alleviate or prevent operator fatigue, aside from short and temporary postponements, is to obtain an adequate quality and quantity of sleep.
What Is the Relationship Between Acute or Chronic Sleep Loss and Increased Crash Risk for CMV Drivers?
Acute fatigue is thought to develop when drivers shortchange themselves of several hours of sleep per 24-hour day over a few days in succession or even over a workweek. Such acute fatigue can usually be ameliorated by gaining sufficient recovery sleep (e.g., during a driver’s nominal 2 days off or his/her weekend). On the other hand, chronic fatigue may be more long-lasting and, in addition to repeated sleep shortages, may involve elements of continuous pressure or stresses from
other sources (e.g., job or family and home circumstances). Knowledge currently is lacking about how to identify chronically fatigued drivers and how to determine ways in which drivers and employers can deal with the problem.
What Is the Relationship Among OSA, PAP Use, and Crash Risk?
OSA is associated with increased crash risk for CMV drivers. An important need is to determine how to identify those drivers most at risk of crashes due to OSA. The extent to which treatment of severe OSA with PAP devices mitigates the risks due to OSA is unknown, but there is some evidence that PAP use for 4 or more hours a night brings the risk for crashes close to that for drivers without OSA. Also unknown is whether drivers diagnosed with mild or moderate OSA are at significant crash risk or whether they, too, might benefit from PAP treatment to make them more alert while driving.
How Do Health and Wellness Programs and Fatigue Management Initiatives Modify the Behavior of CMV Drivers?
A few large truck carriers have derived positive results from their almost 10 years of experience in integrating health and wellness and fatigue management programs, and they have shared those experiences, including the return on their investment in such initiatives. However, most studies of these programs have not sufficiently and reliably validated their efficacy for achieving the goal of reducing crash risk or their scalability. Also, little is known about the use of health and wellness programs by independent owner-operators. Additional development, research, and demonstration of validity and reliability are needed in this area.
What Are the Likely Safety Effects of Improvements in and Deployment of Collision Avoidance and Fatigue Alert Technologies?
Studies of the potential of improved collision avoidance and fatigue alert technologies to ameliorate the impact of fatigue on crash risk in a variety of situations have been conducted in laboratories, in driving simulators, in the field, and in demonstrations of newer technological innovations. In general, however, such studies have not sufficiently and reliably validated the efficacy of these technologies in reducing crash risk or their readiness for deployment. Additional development, research, and demonstration of validity and reliability are needed in this area.
DIRECTIONS FOR FUTURE RESEARCH
Inputs and Outcomes Essential for Further Research Progress
Further progress in research on the relationship among CMV driver fatigue, drivers’ long-term health, and highway safety will require quality information on the inputs and outputs corresponding to the essential research questions listed above. Specifically, information is clearly needed on the following:
- driver loss of alertness in near real time (possibly obtained by a system that measures degree of eyelid closure);
- amount of quality sleep a driver received in the past 24 hours;
- the number of sleep-related crashes in which a driver was involved per vehicle-miles traveled over a long period of time (e.g., several years of driving);
- a driver’s development of various health conditions that can affect alertness;
- changes in components of a driver’s lifestyle, especially diet and health; and
- a driver’s number of lane deviations, unusual speed changes, and unusual brake applications.
Since roadway crashes are rare events, one measurement outcome often used as a surrogate for crashes in naturalistic driving studies is safety-critical events (SCEs). SCEs often are defined as kinematic events, such as hard braking or swerving, that can be viewed as actions that nearly resulted in crashes and are therefore reasonable surrogates for those events. SCEs include incidents that are and are not fatigue related. For some research purposes, then, only a subset of SCEs is relevant, so methods for identifying the most relevant subset of SCEs for research on CMV driver fatigue need to be determined.
To learn more about the research questions outlined above, it will be necessary to better understand the links between the relevant inputs and the relevant outcomes. Achieving this understanding will in turn require having these input and outcome variables at the level of the individual driver. And obtaining that information will likely require the ability to merge data across separate sources. At present, however, while the inputs or the outputs for many of the above research questions may be available to some parties (e.g., confidentially held by carriers/employers or accessibly electronically recorded by equipment manufacturers), many of these measures are not available to researchers.
Alternative Sources of Data
A large proportion, but not all, of the data needed to answer important research questions either is currently collected or could be collected in the near future. These data might be collected by various government agencies, by individual trucking or busing companies, by companies that help monitor the functioning of drivers and vehicles, by industrial collectives, and by various electronic recording devices on the vehicles being driven. In particular, much of these data are currently held by individual companies or industrial collectives with a promise of confidentiality that precludes their sharing for any purpose, including research. This information might include, for example, data on crashes per vehicle-mile traveled, video data of the driver that could be used to assess PERCLOS (percentage of eye closure),1 evidence of lane weaving and hard braking, information on the driving environment and the condition of the vehicle, maneuvers made immediately prior to crashes and in advance of noncrashes, demographics, driver height and weight, other driver health measures, and the type of employment. The collection of these data by private companies and their collectives demonstrates the feasibility of gathering such information and may suggest alternative data collection approaches for FMCSA to consider.
Furthermore, several data-handling procedures and statistical techniques have been used over the past 25 years to provide protection against the release of individually identifiable information but may not be well known outside of the various national statistical agencies. Use of such techniques might make it possible to revise some confidentiality agreements to allow for sharing of selected information with researchers.
For many of the research questions posed above, one potential drawback is that even after the various inputs and outcomes had been gathered, one would have to link those variables for individual drivers and vehicles (properly anonymized). Development of such linkages might require the assignment of a code that would identify a driver across data sets.
Finally, since some of the linkages and information on long-term driver health outcomes will be very difficult to acquire, research progress in this area would be advanced if the National Institute for Occupational Safety and Health (NIOSH), in conjunction with the Bureau of Labor Statistics and FMCSA, were to design and carry out an ongoing survey on the health and wellness of CMV drivers. A survey that would allow comparisons over time is needed to better understand the dynamics underlying changes in the health and wellness of truck and bus drivers.
1 A measure of the percentage of eyelid closure over a period of time (one of the most accepted measures of drowsiness).
Use of Causal Inference Techniques
Crashes often have many causes, only one of which may be driver fatigue due to sleep deficiency, and these other causes must be accounted for in research aimed at understanding how fatigue is linked to crashes. Further, since identifying the causes of a serious crash after the fact is difficult, one must be aware of the potential for substantial measurement error. Research therefore is needed that can disentangle various individual inputs, such as fatigue, from other confounding factors with respect to their impact on crash risk. Such research will require the use of recently developed data collection and statistical techniques useful for drawing causal inferences. It would be advisable for FMCSA to promote the use of such techniques in the research it supports.
Validation of New Technologies
A number of technological countermeasures have been and are being introduced (e.g., lane-deviation tracking systems, newer-generation PERCLOS tracking systems) with the promise of helping to mitigate driver fatigue. The National Highway Traffic Safety Administration could fill an important need by outlining a procedure that could be followed to validate the effectiveness of such new technologies.
Validation of Educational Programs
To date, the effectiveness of the NAFMP has not been properly assessed. The panel believes strongly that the longitudinal survey on CMV drivers’ health and welfare recommended below (in Recommendation 10) needs to include questions on interaction with the NAFMP.
CMV Drivers’ Health and Wellness
Conclusion 1: Insufficient sleep can increase the risk of developing various health problems, including obesity, diabetes, hypertension, and cardiovascular disease, all of which can impact an operator’s level of alertness while driving and potentially impact crash risk.
Conclusion 2: Based on the evidence on drivers who are not commercial motor vehicle drivers, obstructive sleep apnea is known to increase crash risk, and there is no evidence base or compelling reason for thinking
that the same would not also be true among commercial motor vehicle drivers.
Conclusion 3: Better understanding is needed of the effects of treating obstructive sleep apnea in commercial motor vehicle drivers with positive airway pressure (PAP) therapy with respect to the amount and quality of sleep they obtain and their cognition and driver performance following PAP treatment sessions.
Conclusion 5: Substantial data gaps limit understanding of factors that impact the health and wellness of commercial motor vehicle drivers. Closing these gaps would aid greatly in developing a better understanding of drivers’ current status and long-term prospects with respect to health and wellness.
Conclusion 8: Insufficient information exists on (1) how the variety of fatigue management and health and wellness management programs available have been designed, (2) whether drivers/employers actually adhere to these programs, and (3) whether these programs are effective in achieving their goals.
Sleep Insufficiency and Its Impact on Highway Safety
Conclusion 10: There is no biological substitute for sufficient sleep.
Conclusion 12: Despite almost three decades of research on the topic, technological innovations for detecting driver fatigue in near real time and operational strategies for their use are still in the early phases of understanding and application.
Conclusion 13: Biomathematical models can be useful for the development of general work-rest schedules. However, existing models do not account for individual variation, so care must be taken in applying them to address likely impacts of irregular work schedules.
RECOMMENDATION 1: The National Institute for Occupational Safety and Health should be enlisted to design and conduct a regularly scheduled survey every 5 to 10 years to gather information
needed to better understand the demographics and employment circumstances of all commercial motor vehicle drivers in various industry segments.
RECOMMENDATION 2: The Federal Motor Carrier Safety Administration should conduct an evaluation to determine whether commercial motor vehicle drivers’ use of electronic on-board recorders correlates with reduced frequency of hours-of-service violations and reduced frequency of crashes compared with those drivers who do not use such instruments.
RECOMMENDATION 3: Given the potential research benefits of the use of data from electronic logging devices, Congress should consider modifying Title 49 of the U.S. Code to permit the use of such data for research purposes in a manner that protects individualized confidential data from disclosure, and if such a change is made, the Federal Motor Carrier Safety Administration should make parallel provisions in its regulations.2
RECOMMENDATION 4: When commercial trucks and buses containing electronic data recorders that record data on the functioning of the driver and the truck or bus are involved in serious crashes, the relevant data should be made available to investigators and to safety researchers.
RECOMMENDATION 5: The Federal Motor Carrier Safety Administration should incentivize those that capture driver performance data (e.g., large fleets, independent trucking associations, companies that collect telematics data, insurance companies, researchers) to increase the availability of those data relevant to research issues of operator fatigue, hours of service, and highway safety. Any such efforts should ensure that data confidentiality is maintained, perhaps through restricted access arrangements or use of statistical techniques for disclosure protection.
RECOMMENDATION 6: The Federal Motor Carrier Safety Administration should work to improve the collection of and/or access to baseline data on driving exposure by including in its data collection efforts greater detail on the driving environment and by providing these data at low levels of geographic aggregation—even for indi-
2 A change has been made from the prepublication copy to update language to make it clear that FMCSA cannot change the law but it can modify its regulations.
vidual highway segments. Comparisons enabled by the availability of these baseline data would benefit several proposed lines of new research.
How CMV Drivers React to Feelings of Drowsiness
RECOMMENDATION 7: The Federal Motor Carrier Safety Administration should support research aimed at better understanding the factors associated with driver behavior related to fatigue and sleep deficiency, including what motivates drivers’ decisions about whether to continue driving when they feel fatigued.
Testing of New Technologies
RECOMMENDATION 8: Using a human-systems integration framework, the Federal Motor Carrier Safety Administration and the National Highway Traffic Safety Administration, in consultation with the Centers for Disease Control and Prevention and the National Institutes of Health, should develop evaluation guidelines and protocols for third-party testing, including field testing, conducted to evaluate new technologies that purport to reduce the impact of fatigue on driver safety.
How FMCSA Can Improve Its Research Projects Through Peer Review
RECOMMENDATION 9: The Federal Motor Carrier Safety Administration should make greater use of independent peer review in crafting requests for proposals, assisting in decisions regarding awards, and monitoring the progress of projects (including in the study design and analysis stages). Peer review should include expertise from all relevant fields, including epidemiology and statistics—especially causal inference—to address appropriate design and analysis methods.
Understanding of What Impacts the Long-Term Health of CMV Drivers
RECOMMENDATION 10: The U.S. Department of Health and Human Services and/or the U.S. Department of Transportation should fund, design, and conduct an ongoing survey that will allow longitudinal comparisons of commercial motor vehicle drivers to enable tracking of changes in their health status, and the factors
likely to be associated with those changes, over time. In addition, it would be highly desirable for the survey data thus collected to include sufficient information to enable linking of the data to relevant electronic health records, with a particular focus on conditions that may threaten drivers’ health and safety.
RECOMMENDATION 11: The Federal Motor Carrier Safety Administration should continue to encourage all individuals included in the National Registry of Certified Medical Examiners to utilize current best practices in identifying drivers who should be referred for additional sleep malady testing and in making determinations about commercial driver’s license renewal extensions. It would be highly preferable, as soon as possible, to supply the examiners with clear criteria or guidance on when it is appropriate to refer presenting drivers for sleep malady testing.
RECOMMENDATION 12: The Federal Motor Carrier Safety Administration should support peer-reviewed research on obstructive sleep apnea (OSA) and commercial motor vehicle drivers throughout all the research stages, from the drafting of requests for proposals through analysis of data. The supported research should be focused on a better understanding of the incidence of OSA in commercial motor vehicle drivers; its impact on driver fatigue, safety, and health; and the benefits of treatments. Specific research topics might include
- determining the number of commercial motor vehicle drivers whose quantity/quality of sleep and driving performance are likely affected at various levels of apnea-hypopnea (index of OSA severity);
- determining what rules for sleep-screening referrals are effective in discriminating between those commercial motor vehicle drivers with and without OSA;
- delineating the causal chain from diagnosis of OSA (preferably as a function of severity) to increased likelihood of crash frequency among commercial motor vehicle drivers;
- determining the impact of treatment with positive airway pressure (PAP) and similar devices on long-term health and crash rates among commercial motor vehicle drivers with varying degrees of apnea severity; and
- identifying the required/recommended duration of initial PAP treatment (e.g., suggested number of hours of treatment per day/week) before a driver can be certified to return to driving.
RECOMMENDATION 13: The Federal Motor Carrier Safety Administration (FCMSA) should carry out a research program on driver fatigue management and training. This research program should include
- evaluating the effectiveness of the North American Fatigue Management Program (NAFMP) for educating truck and bus drivers in how to modify their behavior to remedy various potential sources of fatigue;
- determining how effective the NAFMP training modules are in meeting the needs of drivers’ employers, including fleet managers, safety and risk managers, dispatchers, driver trainers and other corporate officials (e.g., those conducting carrier-sponsored employee health and wellness programs);
- evaluating any new education programs regarding sleep apnea that FMCSA has or plans to develop; and
- examining possibilities for the development and evaluation of incentive-based programs for improving health and fitness, including regular coaching, assessment, and support.
This page intentionally left blank.