Increased Risk of Crashes for Commercial Motor Vehicle Drivers with OSA

A substantial research literature supports the conclusion that severe OSA—that is, OSA with an AHI over 30—is associated with increased crash risk for the nonprofessional driver. Some of the major contributors to this literature include Aldrich (1989), Findley et al. (2000), George and Smiley (1999), Teran-Santos et al. (1999), and Wu and Yan-Go (1996). (See Smolensky et al. [2011] for a summary of studies analyzing the relationship between OSA and frequency of crashes.) The reason for this increased crash risk is that OSA causes cognitive dysfunction, due largely to sleep fragmentation (Bedard et al., 1993; Day et al., 1999; Deaconson et al., 1988; Drummond et al., 2000; Feuerstein et al., 1997; Kaneko et al., 2003; Kim et al., 1997).

Besides fatigue, the cumulative effects on the brain of chronic repeated nocturnal hypoxic episodes may cause irreversible cognitive deficits (Bedard et al., 1993; Nowak et al., 2006). Evidence indicates that some individuals with OSA are unaware of their cognitive impairment or even of being drowsy (Dement et al., 1978; Engleman et al., 1997). Several studies have shown little or no correlation between an OSA patient's perception of sleepiness and motor vehicle crash history (Barbe et al., 1998; Horstmann et al., 2000; Teran-Santos et al., 1999; Yamamoto et al., 2000; Young et al., 1997). Because of this lack of insight, these individuals are less likely to restrict themselves from driving than drivers who are more aware of being drowsy, despite being at increased risk for a crash.

Tregear and colleagues (2009b) performed a systematic review and meta-analysis of 18 studies drawn from seven databases described in those sources. Two of these 18 studies examine the association between OSA and crash risk for CMV drivers, while the rest are focused on passenger car drivers. Nine of the studies provide data on the relative incidence of crashes among comparable individuals with and without the disorder. Pooling of these data revealed that the mean crash risk ratio associated with OSA was likely to fall within the range of 1.30 to 5.72. Thus, if the underlying crash risk for a driver is 0.08 crashes per person-year, the crash risk for a driver with OSA can be expected to be in the range of 0.10 to 0.46 crashes per person-year. A series of sensitivity analyses was used to establish the robustness of these results to various assumptions. While the quality of the individual studies is not high, the data are qualitatively consistent, making it unlikely that future studies will contradict the finding that nonprofessional drivers with OSA are at increased risk for a motor vehicle crash. The strength of this conclusion results from the different environments and situations tested and methods used.

The question then is whether there also is an increased risk for crashes among CMV drivers with severe OSA (and what that increased risk is). A finding regarding CMV drivers reported by Tregear and colleagues (2009b) is that as “a group, drivers with OSA are at an increased risk for a motor vehicle crash when compared with comparable drivers who do not have the disorder even though a precise estimate of the magnitude of this increased risk could not be determined.” An argument might be made that definitive proof of an elevated crash risk among CMV drivers with OSA of a given hypopnea level can come only from a randomized controlled trial examining whether untreated OSA is associated with a higher risk of crashing. Such a study, however, is not feasible; it would also be unethical. Once professional drivers have been diagnosed with OSA, they must initiate treatment by using a PAP device, or they cannot continue driving. Therefore, professional drivers with OSA cannot legally or ethically be tested on the open road unless their OSA is addressed. (Such studies could be conducted in driving simulators with, say, a 2-week confinement, in which the drivers' sleep environment and schedule were controlled; see O'Neill et al. [1999].) The same barriers would forbid the testing of CMV drivers with OSA to assess the benefits of PAP devices in reducing crash risk.

There are reasons to believe that nonprofessional and CMV drivers would differ in their response to OSA. They certainly are different populations of drivers and faced with different tasks, so the hypothesis that CMV drivers with OSA are not at increased risk needs to be considered seriously. Professional drivers may take more naps when needed, they may orient their schedules to allow for more rest, they may ingest greater amounts of caffeine, or they may avoid congested areas when they feel tired. On the other hand, CMV drivers often drive longer hours and often are older than the average noncommercial driver studied. Evidence also suggests that, relative to noncommercial drivers, they suffer from more ailments that could increase their degree of fatigue. Evidence that could be used to reject the notion that there is a difference between the two categories of drivers would support the hypothesis that the findings for nonprofessional drivers of passenger cars can be transferred to professional drivers of trucks and buses. Conversely, evidence that professional drivers take actions such as those cited above more often or more effectively relative to nonprofessional drivers would support the notion that there is no increased risk of crashes among the former group.⁴

The panel did not find credible evidence that actions to counter fatigue are taken more often or more effectively by CMV drivers. However, an unpublished study by Schneider National found that the crash frequency among drivers using PAP devices was reduced compared with their previous crash record.⁵ In addition, literature on somewhat analogous populations, such as railroad engineers, those in charge of ships, and airline pilots, provides evidence that OSA is a factor in increased risk of mistakes and that use of PAP devices reduces that risk (see, e.g., Quan and Barger, 2015).

Finally, it can be argued that the absence of proof is not proof of the converse. Unless good studies show that CMV drivers with severe OSA are not at increased risk of crashes, it is a principle of preventive medicine and a reasonable precaution to accept that the parallel work among general drivers applies to CMV drivers as well. Increasingly, observational science is being held to an impossible standard—either it is proven that the adverse event occurs in the “exact” population of interest, or there is considered to be no proof. The panel believes many preventable losses will be suffered if a very narrow view of what constitutes an evidence base is used to argue that only evidence that is direct and not indirect is sufficient.

The panel notes that the relationship between degree of OSA and increase in crash risk is likely represented by a smooth response curve, such that those drivers with mild OSA have little or no greater risk of crashes than drivers without OSA. Therefore, it would be extremely helpful if in the future, the research literature could be more precise about the level of OSA being considered.

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.

Treatment for OSA

The only treatment for OSA besides weight loss that has shown widespread efficacy is the use of positive airway pressure (PAP) devices. For nonprofessional drivers with OSA, use of such devices is associated with a reduced risk of crashes compared with those with OSA who do not use such devices (Cassel et al., 1996; Findley et al., 2000; George and Smiley, 2001; Tregear et al., 2009b). A key question, then, is whether the use of PAP devices also reduces the risk for CMV drivers, and if so, to what extent?

As in the case of linking OSA and crash risk, while studies linking PAP use to decreased crash risk have some limitations, the panel believes the finding of reduced risk for nonprofessional drivers can be strongly supported. However, there is not a substantial literature directly showing that the same assertion can be made for CMV drivers (with a given degree of hypopnea).

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.

A major complication is that acceptance of and adherence to PAP therapy is a problem for many patients (Kribbs et al., 1993; Weaver et al., 1997). Moreover, the amount of PAP use needed to produce clinically meaningful improvements in real-world sleep, cognition, and behavior remains unclear (Gay et al., 2006).

Medical Examination Policy Regarding OSA

FMCSA's current medical examination policy for CMV drivers states, “A person is physically qualified to drive a motor vehicle if that person has no established medical history or clinical diagnosis of a respiratory dysfunction likely to interfere with his/her ability to control and drive a motor vehicle safely.” Thus if the medical examiner detects a respiratory dysfunction that in any way is likely to interfere with the driver 's ability to control and drive a commercial motor vehicle safely, the driver must be referred to a specialist for further evaluation and therapy before being certified. While no regulation specifically mentions OSA, it would be covered under this respiratory standard.

In 1998, OSA was first mentioned in the advisory criteria for the respiratory standard:

There are many conditions that interfere with oxygen exchange and may result in incapacitation, including emphysema, chronic asthma, carcinoma, tuberculosis, chronic bronchitis and sleep apnea. . . .

Medical examiners could find minimal guidance on OSA in a 1991 Conference on Pulmonary/Respiratory Disorders and Commercial Drivers (Turino et al., 1991), sponsored by the Office of Motor Carriers, a predecessor of FMCSA. The experts on that panel recommended that if there were any suspicion that a driver had OSA, the driver should be evaluated and the condition successfully treated before the driver returned to work. Medical examiners were offered no specific criteria for having concern or suspicion that a driver was at risk of having OSA. The conference participants recommended at least a 1-month waiting period after treatment for OSA had been initiated before a driver returned to commercial driving. They also advised evaluation of the effectiveness of treatment through either multiple sleep-latency testing (MSLT) or polysomnography. They recommended further that drivers with OSA who were medically qualified to drive commercial motor vehicles be reevaluated annually by means of sleep studies or MSLT.

The 1991 conference report was not widely disseminated, and the majority of medical examiners likely were unaware of its existence. In 2000,⁶ the examination form used to evaluate CMV drivers was updated and for the first time included a specific question on OSA, asking the driver to indicate “yes” or “no” with respect to “sleep disorders, pauses in breathing while asleep, daytime sleepiness, loud snoring.” When this form was first issued, many drivers responded “yes” to this question, and examiners reported that they either asked these drivers to provide documentation from their treating provider or referred the drivers for further evaluation. By the second year of use of this form, examiners indicated that few drivers checked “yes” for this question.

In preparation for the National Registry of Certified Medical Examiners (NRCME), FMCSA compiled the Medical Examiner Handbook from existing guidance drawn from various FHWA (or FMCSA)-sponsored panels, frequently asked questions (FAQs), advisory criteria, and regulatory guidance. The section with information on OSA was the last section posted—not until fall 2010—and was not significantly different from what had been in the 1991 conference report.

The Medical Examiner Handbook indicated that examiners “should not certify the driver with suspected or untreated sleep apnea until etiology is confirmed and treatment has been shown to be stable, safe, and adequate/ effective,” but still offered no specifics on how to identify drivers suspected of having OSA. Additional guidance for examiners had been provided in three FAQs on the FMCSA website addressing sleep disorders. FMCSA also had a section of its website devoted to sleep apnea, “Spotlight on Sleep Apnea,” with information for drivers and carriers, but no information on screening, diagnosis, or treatment. In early 2014, to comply with Public Law 113-45,⁷ FMCSA removed the section of the Medical Examiner Handbook on OSA, the FAQs on sleep disorders, and “Spotlight on Sleep Apnea” from the FMCSA website. In early 2015, as noted earlier, the entire Medical Examiner Handbook was removed from the FMCSA website for revision.

There have been several efforts to develop criteria that examiners could consider for identifying drivers at highest risk of having OSA. In 2006, a Joint Task Force of the American College of Occupational and Environmental Medicine, the American College of Chest Physicians, and the National Sleep Foundation recommended OSA screening criteria for drivers (Hartenbaum et al., 2006). In 2007 and 2008, based in part on an FMCSA-supported evidence report (updated in 2011) (Williams et al., 2011), FMCSA's Medical Expert Panel on Obstructive Sleep Apnea and Commercial Motor Vehicle Driver Safety (MEP) (Ancoli-Israel et al., 2008) and Medical Review Board (MRB) (Federal Motor Carrier Safety Administration Medical Review Board, 2008) developed recommendations, referenced earlier, that would have required testing a larger percentage of drivers than the Joint Task Force had proposed.⁸ In 2012, the Motor Carrier Safety Advisory Committee (MCSAC), an industry advisory group, and the FMCSA MRB offered joint recommendations (Federal Motor Carrier Safety Administration Medical Review Board, 2012).

FMCSA did not adopt the three sets of recommendations described in the previous paragraph. Examiners therefore continue to vary in how they evaluate drivers who may be at risk of having OSA. Durand and Kales (2009) found that while members of the American College of Occupational and Environmental Medicine believed that screening for OSA was important, fewer than 50 percent used any specific criteria for such screening.

Until 2014, medical examiners could be any health care providers licensed by their state to perform physical examinations, including medical doctors; doctors of osteopathy; nurse practitioners; physician assistants; and in some states, chiropractors, dentists, or even physical therapists. The NRCME was fully implemented in 2014,⁹ and examiners on the registry are now required to be trained and certified in accordance with the NRCME Core Curriculum, which focused on the regulations, guidelines, and other official information from FMCSA available at that time.

With implementation of the NRCME, it was hoped that some consistency among medical examiners would be established, as all examiners were required to complete training that at least met the FMCSA Core Curriculum. While some training programs were limiting instruction to what was in the Medical Examiner Handbook, others were introducing additional resources, such as the MRB, MEP, or MCSAC recommendations, and still others were teaching (erroneously) that there were specific required criteria for screening drivers for OSA. The NRCME Medical Examiner Sample Training Handbook (National Registry of Certified Medical Examiners, 2012) noted that training programs could teach material beyond the Medical Examiner Handbook, provided they clearly highlighted that the material was not endorsed by FMCSA, and that examiners could use more current guidance than that issued by FMCSA in making certification determinations. FMCSA had indicated that, while the NRCME certification examination would include only information that had undergone public notice and comment, examiners should consider current best practice in making certification determinations, not only for OSA but also for other conditions for which official information from FMCSA was lacking, such as Parkinson's disease or use of potentially impairing medications such as opioids.

FMCSA still has not provided specific guidance for medical examiners—this despite recommendations to the agency from the National Transportation Safety Board (NTSB) (2009); the expert recommendations of the FMCSA MEP, MRB, and MCSAC; and requests from the American College of Occupational and Environmental Medicine (ACOEM).¹⁰ The absence of specific guidance to certified medical examiners on assessing CMV drivers for OSA presents challenges for employers who are relying on the examiners to make that determination but finding that inconsistent criteria are used, even within the same examiner group. There now exists a risk of grievances or legal action against employers or examiners that attempt to utilize current best practices and require diagnostic studies for some drivers. At the same time, lawsuits have been brought against employers and examiners when a crash has occurred; and the CMV driver subsequently has been found to have OSA but was not tested even though he or she would have been considered at high risk of the disorder based on one or more of the above recommendations. In January 2015,¹¹ FMCSA issued a bulletin to medical examiners and training associations stating that examiners should use current best practices in determining which drivers should have objective testing and offering some considerations with respect to OSA, but noting that FMCSA has no specific requirements related to the disorder. Many examiners and employers have indicated that they will limit screening drivers for OSA until FMCSA issues specific guidance.

A 2014 survey (Hartenbaum, 2015) of examiners, both physicians and nonphysicians, both ACOEM-trained and not, who were listed on the NRCME demonstrated no consistency in examination outcomes for five scenarios in which current best practice would recommend that drivers have diagnostic testing for OSA. Selections included certify for 2 years, certify for 1 year and instruct driver to discuss risk of OSA with primary care physician, certify for 3 months and request documentation from primary care physician on risk of OSA, certify for 3 months and require sleep study, or not certify at all. The appropriate approach to screening CMV drivers for OSA remains contentious, and the absence of specific guidance affects drivers, examiners, employers, and third-party administrators.

Conclusion 4: It is apparent that medical examiners who certify commercial motor vehicle drivers are not consistent in understanding and applying current best practice to identify drivers who may be at risk of moderate to severe obstructive sleep apnea. Many examiners therefore are inconsistent in making determinations as to when a driver should be referred for additional sleep testing.

Hypertension

Hypertension—elevated systolic or diastolic blood pressure—is a risk factor for cardiovascular disease, peripheral vascular disease, and chronic renal insufficiency. A person with a systolic blood pressure reading of greater than 140 or a diastolic reading of greater than 90 is regarded as having hypertension. CMV drivers are more likely to have hypertension than their peers in other professions (and there is some evidence that this gap is widening) (Thiese et al., 2015a, 2015b). The salient regulation in the Medical Examiner Handbook is as follows: “If a driver has hypertension and/or is being medicated for hypertension, he or she should be recertified more frequently. An individual diagnosed with Stage 1 hypertension (BP is 140/90-159/99) may be certified for one year. At recertification, an individual with BP equal to or less than 140/90 may be certified for one year; however if his or her BP is greater than 140/90 but less than 160/100, a one-time certificate for 3 months can be issued. . . . Once the driver has reduced his or her BP to equal to or less than 140/90, he or she may be recertified annually thereafter.”

Diabetes Mellitus

About 8 percent of the population has diabetes, and one-third of cases may be undiagnosed. By contrast, the prevalence of diabetes in CMV drivers may be as high as 30 percent (Thiese et al., 2015a, 2015b), and evidence indicates that the incidence of diabetes in CMV drivers increased between 2005 and 2012. Factors that can affect blood glucose control are fatigue, lack of sleep, poor diet, and stress. Drivers diagnosed with diabetes mellitus who require insulin to control their condition are not eligible for certification because poorly controlled blood glucose can lead to fatigue, lethargy, and sluggishness, and complications can lead to acute hyperglycemia, which in turn can cause seizures, loss of consciousness, and impaired cognitive function. The medical examination relevant to diabetes includes assessment of glycosuria, retinopathy, macular degeneration, peripheral neuropathy, coronary heart disease, cerebrovascular disease, autonomic neuropathy, and nephropathy.

Cardiovascular Problems

Thiese and colleagues (2015a, 2015b) suggest that up to 5 percent of CMV drivers suffer from cardiovascular disease. The primary manifestations include acute myocardial infarction, angina pectoris, and congestive heart failure. Drivers that have cardiovascular disease, relative to those without, are more likely to experience sudden death or incapacitation. Since other conditions may exacerbate cardiovascular problems, the decision on certification of those with cardiovascular disease depends on a comprehensive medical assessment of overall health. However, if a driver is diagnosed with myocardial infarction, angina pectoris, coronary insufficiency, thrombosis, or any other cardiovascular disease known to be accompanied by syncope, dyspnea, collapse, or congestive cardiac failure, he or she is not to be certified for driving. The diagnosis includes evaluation for heart murmurs, extra heart sounds, arrhythmias, an enlarged heart, abnormal pulse, carotid or arterial bruits, or varicose veins. Details on cardiovascular problems and the increased risk for truck drivers can be found in FMCSA (2002) and the references included therein.

Health and Wellness Programs of Large Carriers

While the panel did not undertake a systematic review of fatigue management or health and wellness programs instituted by large carriers, it was briefed on the experiences of three large carriers at its third meeting in September 2014. While the panel knows anecdotally that some small carriers have instituted health and wellness management programs, these appear to be few in number. This is to an extent understandable given the fixed costs of implementing such programs.

The panel heard about the employee health and wellness programs of Schneider National, J.B. Hunt, and Con-Way Motor Freight. Table 8-1 summarizes selected information from these presentations. These three carriers have integrated driver fatigue management into their health and wellness programs. Returns on investment in successful health and wellness programs include such benefits as improved employee morale and driver retention rates; enhanced corporate culture and reputation; decreased clinic visits and costs; improved insurance premium rates; and, importantly, decreased accidents and injuries, and therefore also improved safety records (Krueger, 2012).

TABLE 8-1

Health and Wellness Programs of Three Large Carriers as of September 2014.

Current Fatigue and Health and Wellness Programs in the Bus Industry

Hundreds of bus companies operate between 2 and 10 buses. Very little is known about the fatigue or health and wellness programs of these carriers. One example of a fatigue management program of a large interstate bus company was brought to the panel's attention by the chair of the Bus Industry Safety Council for the American Bus Association, Michael McDonal. In his presentation to the panel, he stated that Greyhound, a large bus company, has a fatigue management initiative based on input received from focus groups comprising drivers, operations managers, and safety directors throughout the industry. On the basis of input from focus groups, Greyhound developed countermeasures to address driver stress and fatigue. Data were collected on the effectiveness of these measures from a small set of drivers during actual night runs. Based on these data, extra board operations were modified.¹³

Fatigue Outreach and Health and Wellness Initiatives of FMSCA

Since the mid-1990s, FMCSA has encouraged carriers to engage in employer-sponsored driver fatigue management programs. To this end, the agency has conducted extensive education and outreach; held forums on the topics; and carried out extensive research on driver fatigue and performance in conjunction with cooperative, enlightened truck and bus carriers. In terms of continuously addressing driver fatigue management, FMCSA developed and sponsored (with industry) the Mastering Alertness and Managing Driver Fatigue train-the-trainer 4-hour course, which was offered to more than 4,500 trucking safety and risk managers from 1996 until 2006. FMCSA also arranged to develop (with industry) an employee health, wellness, and fitness program entitled “Gettin'-InGear,” a train-the-trainer program offered to more than 2,500 trucking safety and risk managers from 2001 to 2006.

The North American Fatigue Management Program (NAFMP), sponsored by FMCSA, Transport Canada, and others, has been available on its own dedicated website since summer 2013. This extensive set of more than 800 slides is offered in 10 training modules aimed at employers, managers, drivers, shippers, and receivers. The 10 modules are (1) introduction and overview, (2) safety culture and management practices, (3) driver education, (4) family education, (5) train the trainer, (6) truck driver safety and compliance: the role of shippers and receivers, (7) sleep disorders management for motor carriers, (8) driver sleep disorders management, (9) driver scheduling and tools, and (10) fatigue monitoring and management technologies.

NAFMP serves as an information source for the entire ground shipping industry. It offers numerous health and wellness lessons; fitness principles; and coverage of additional selected topics, especially those related to sleep hygiene, including sleep disorders such as OSA. For example, the program covers the effects of being continuously jostled, consuming a bad diet, not exercising, poor weight management, high blood pressure, using various medications, using alcohol, using tobacco, disruption of circadian rhythms, and resultant conditions such as diabetes.

As noted by Terri Hallquist of FMCSA in her presentation at the panel's third meeting, there is currently no effort under way to evaluate who visits the NAFMP website or whether those who do visit are engaging in any behavior modification, such as sleeping longer.¹⁴ This lack of evaluation to date is likely the result of the program's recent implementation. As discussed in Chapter 11, some evaluation of the program is feasible using web analytics. Given the attention paid to fatigue and health and wellness management by the large carriers, one key concern is whether NAFMP is adequate to educate and motivate behavior change among drivers for many midsize carriers that lack such programs or for small carriers with only a handful of trucks and drivers, or among the hundreds of thousands of independent owner-operator drivers nationwide.

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.