IATSS Research 37 (2013) 1–7

Contents lists available at SciVerse ScienceDirect

IATSS Research

Short duration, sleep disorders, and traffic accidents

Yoko Komada, Shoichi Asaoka, Takashi Abe, Yuichi Inoue ⁎

Department of Somnology, Tokyo Medical University, Japan article info abstract

Article history: Sleepiness is an important factor in traffic accidents caused by human error. The purpose of this paper is to Received 26 March 2013 review a number of studies conducted over the years regarding the effect of the lack of sleep on the incidence Received in revised form 3 June 2013 of traffic accidents as well as the individual effects of various sleep disorders on accidents. In addition, we Accepted 4 June 2013 discuss recent advances in methods of detecting sleepiness and strategies for preventing traffic accident by using these methods. Keywords: © 2013 International Association of Traffic and Safety Sciences. Production and hosting by Elsevier Ltd. Sleepiness syndrome All rights reserved. Insufficient sleep Traffic accident Countermeasure

Contents

1. Introduction ...... 1 2. Lack of sleep and accidents ...... 2 3. Sleep disorders and accidents ...... 3 3.1. ...... 3 3.2. Sleep apnea syndrome ...... 3 3.3. Narcolepsy ...... 4 3.4. Behaviorally-induced insufficient sleep syndrome ...... 4 3.5. Shift-work disorder ...... 5 4. Sleepiness detection methods and accident countermeasures ...... 5 5. Conclusion ...... 6 References ...... 6

1. Introduction a further 39. The bus, on its way from Kanazawa to Tokyo, hit a roadside wall on the Kanetsu Expressway around dawn. No brake or skid marks In April 2012 a bus driver in Japan who fell asleep at the wheel were found at the site of the accident; the Gunma Prefectural Police caused a disastrous accident that took the lives of 7 people and injured reported that the driver had “fallen asleep.” Around the same time in Kameoka City, Kyoto, a mini-car ran into a group of children walking to school with their guardians, hitting 10 people, of whom 3 died and ⁎ Corresponding author at: Department of Somnology, Tokyo Medical University, 6-7-1 7 were severely injured. This accident reportedly involved an unlicensed Nishishinjuku, Shinjuku-ku, Tokyo 160-0023, Japan. Tel.: +81 3 3342 6111; fax: +81 3 3342 7083. 18-year-old driver who fell asleep at the wheel. Generally, accidents E-mail address: [email protected] (Y. Inoue). caused by drowsiness and falling asleep are the result of drivers being Peer review under responsibility of International Association of Traffic and Safety Sciences. unable to control the brakes or steering due to their impaired state. Furthermore, because such accidents often occur at high speeds, there is a much greater mortality rate or risk of serious injuries [1]. Sleepiness is an important factor in traffic accidents caused by human error. One study of 19,000 subjects across five European

0386-1112/$ – see front matter © 2013 International Association of Traffic and Safety Sciences. Production and hosting by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.iatssr.2013.06.001 2 Y. Komada et al. / IATSS Research 37 (2013) 1–7 countries [2], as well as another of 3000 Japanese adults [3], found 2. Lack of sleep and accidents that excessive daytime sleepiness (EDS) reached as much as 15%. The main causes of EDS are (1) the lack of sleep, (2) deterioration in In a previous study, we sought to identify a correlation between the the quality of sleep, (3) disruption of circadian rhythms, and (4) primary occurrence of traffic accidents due to falling asleep at the wheel and . In terms of lack of sleep, it has been suggested that sleep duration on the day before the accident. We analyzed 1772 of lifestyles requiring activity around the clock or late at night are the the 4871 traffic accidents that occurred in Tsukuba City from August main cause of this condition. In 1960, 90% of people reportedly went 1993 to December 2003, excluding those caused by drunk driving [9]. to sleep at 11:00 PM and slept for an average of more than 8 h, but First, accidents were categorized by type as rear-end collisions (n = the number of people active late at night has gradually risen, resulting 240), single-car accidents (n = 293), or other accidents (n = 1239). in shorter sleeping times. In 2010, half of the population was still A rear-end collision was defined as a trailing car crashing into the rear awake at 11:00 PM and the mean sleeping time was only 7 h and of a leading vehicle. A single-car accident was defined as a car crashing 14 min, suggesting a reduction of around 1 h of sleeping time compared against a structure on a road or shoulder such as a parked vehicle, wall, with 40 years ago [4] (Fig. 1). In a study conducted on the general house, or telegraph pole. Accidents involving interaction with other population (n = 12,000) in Finland, 20.4% of participants were found vehicles or persons were classified as other accidents. Our analysis of to have insufficient sleep [5], while a study of 4000 adult subjects in the time of day that each accident occurred revealed that rear-end colli- Japan revealed that 28% had fewer than 6 h of sleep each night [6]. sions most frequently occurred between 8:00 and 9:00 AM. On the other Thus, there appears to be a high rate of insufficient sleep in developed hand, single-car accidents occurred more frequently during the night countries. Of course, people with chronically short sleeping times than during the day (Fig. 2). These findings suggest that the time zone often have EDS. However, although an accumulated lack of nocturnal of accident occurrence differs by the type of accident. The risk of rear- sleep can result in serious deficits in neurobehavioral function, the end collisions grows with increased traffic density, especially during increase in subjective sleepiness under such conditions remains mild. the morning rush hour, which may lead to less space between cars. That is, people with chronic lack of adequate sleep may underestimate In contrast, it was hypothesized that single-car accidents could be attrib- their own sleepiness [7]. Unwanted sleepiness may also occur due to uted to low nighttime traffic density that reduced driver alertness and the disruption of circadian rhythms in situations such as or shift increased driving speed. work, as well as primary circadian rhythm sleep disorders such as For both rear-end collisions and single-car accidents, a higher per- delayed sleep phase. People whose quality of sleep deteriorates as a centage of subjects were found to have had less than 6 h of sleep on result of sleep disorders that cause frequent interruption of nocturnal the day before the accident than for other accidents (11.3% in rear- sleep, such as syndrome and periodic limb move- end collisions, 10.2% in single-car accidents, 4.8% in other accidents). ment disorder, can develop secondary hypersomnia. In addition, while Multivariate logistic regression analysis showed that the occurrence of primary hypersomnia such as narcolepsy and is relatively rare, it can cause serious problems with EDS [8]. a In this report, we survey a number of studies conducted over the 300 years regarding the effect of the lack of sleep on the incidence of traffic 250 accidents as well as the individual effects of various sleep disorders on accidents. In addition, we discuss recent advances in methods of 200 detecting sleepiness and strategies for preventing trafficaccidentby using these methods. 150 100

Number of accidents 50

0 0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 b 100

80

60

40

20 in each time zone (%)

Proportion of each accident 0 0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 Time of day

Accidents of other types Fig. 1. Changes in Japanese sleep duration and percentage of people staying awake past 11:00 PM. This figure shows the percentage of Japanese people who stay awake past Single-car accidents 11:00 PM on weekdays and the average subjective total sleep time from 1960 to Rear-end collisions 2010. In 1960, 90% of the population were already asleep at 11:00 PM and typically slept for over 8 h. However, sleeping times gradually fell over the surveyed period. Fig. 2. (a) Time of day distribution of all accidents, and (b) proportion of rear-end In 2010, half of the population remained awake past 11:00 PM with an average collisions, single-car accidents and other accidents among all accidents in each time sleeping time of 7 h and 14 min. zone. Created by the NHK National Time Life Survey. Abe et al., Journal of Sleep Research. 2010. Y. Komada et al. / IATSS Research 37 (2013) 1–7 3 rear-end collisions was significantly associated with sleep duration of falling asleep at the wheel has been studied by numerous researchers less than 6 h and more frequent driving, while the occurrence of [28,30]. The results of a meta-analysis study by Sassani et al. [31] single-car accidents was significantly associated with younger age showed that the odds ratio of traffic accidents increased in people (≤25 years old), sleep duration of less than 6 h, the increment between with SAS by 2.52 (95% CI: 1.84–3.45). spells of driving, and nighttime (21:00–06:00) driving. Our systematic investigation of MVA risk among Japanese drivers In another study, we administered a questionnaire in October 2008 with obstructive sleep apnea syndrome (OSAS) revealed that the odds to approximately 4000 non-professional drivers and investigated ratios for MVAs and dozing off at the wheel in the OSAS group com- factors associated with motor vehicle accidents (MVAs). The results pared with the general population were 2.4 (95% CI: 1.6–3.4) and 13.2 obtained were as follows: self-reported sleep duration on weekdays of (95% CI: 10.0–17.4), respectively [32]. Multivariate logistic regression less than 6 h (OR = 8.02, 95% confidence interval [CI]: 1.08–59.60, analysis revealed that MVAs were significantly associated with either p b .05), more than 6 h and less than 7 h (OR = 6.28, 95% CI: 0.83– high scores on the Epworth Sleepiness Scale (ESS ≥ 11) or frequent 47.43, p = .07), and loud or apnea (OR = 2.07, 95% CI: 1.12– respiratory disorder events (apnea Hypopnea Index: AHI ≥ 40/h). 3.82, p b .05) were all associated with an increased risk of MVAs [10]. Further studies have found that long-term continuous positive airway This result suggests that in order to prevent MVAs caused by falling pressure (CPAP) treatment was clearly effective in reducing the number asleep, drivers should have adequate sleep and undergo screening for of MVAs [32,33]. In our study indicated above, the proportion of sleep apnea syndrome. subjects who experienced MVAs was significantly higher in the OSAS group at baseline compared with the control group and OSAS group at 3. Sleep disorders and accidents the follow-up period (OSAS group at the baseline: 16.8%; control group: 4.7%; OSAS group at follow-up period: 3.1%; p b 0.01). Further- The International Classification of Sleep Disorders [8], includes the more, there was no significant difference in the rate between the following classifications for sleep disorders: insomnia, sleep-related control group and the CPAP-treated OSAS group at follow-up. Results breathing disorders, of central origin, circadian rhythm from a study using driving simulators also demonstrated a significant sleep disorders, , sleep-related movement disorders, improvement in subjective sleepiness and driving performance in isolated symptoms, and other sleep disorders. Among them, drivers CPAP-treated OSAS patients [34]. with sleep disorders that can cause severe EDS such as sleep apnea syn- Interestingly, in our study AHI scores were significantly higher in drome or hypersomnias of central origin are reported to have 1.5–4 those who had multiple MVAs than in those who had only a single times greater risk of traffic accidents than drivers without any sleep MVA, despite the ESS scores of the former group being significantly disorder [11,12]. Undoubtedly, diagnosis and treatment of these sleep lower (Fig. 3). This finding supports the hypothesis that self-reported disorders are important to reduce traffic accidents caused by falling measures tend to underestimate the severity of sleepiness in severe asleep at the wheel. Here, we focus on several sleep disorders, all of OSAS patients [35]. In addition, in a systematic review examining MVA which are associated with an increased risk of trafficaccidents.Inaddi- rates in OSAS patients, ESS scores did not appear to be predictive of tion to introducing accident conditions for each disorder, which have crashes in 8 of the 15 studies examined [36]. Mulgrew et al. [37] also been published previously, we discuss issues that should be clarified reported no significant relationship between subjective sleepiness and to keep traffic safe from driving problems caused by the disorders. MVA rate, and suggested that patients who are less aware of their im- paired vigilance may be at much higher risk. Thus, we should consider 3.1. Insomnia that MVAs might occur when subjective measures of sleepiness are lower than expected from the AHI, especially in patients severely affected Insomnia is known to be common [13–15], affecting approximately by OSAS. Therefore, early detection and treatment of severe OSAS should 10% to 15% of the general population [2,16]. In addition, the rate of be promoted to prevent multiple MVAs, regardless of whether there is occasional sleep medication use in the general population is approxi- subjective sleepiness. mately 3% to 11% [17–19]. In people with insomnia symptoms, the In order to elucidate the current rate of falling asleep at the wheel rate is approximately 26% [20]. Nocturnal insomnia is frequently associ- among driver's license holders and to understand the correlation ated with daytime sleepiness and fatigue [21,22],psychomotordeficits between falling asleep at the wheel and SAS, approximately 5000 [23,24] and cognitive impairments [25], but no study has yet revealed a driver's license holders who went for renewal were asked to answer direct correlation between insomnia and an increased risk of traffic an anonymous and voluntary questionnaire in September 2006 (respon- accidents. The residual effects of the next morning, however, dents: 3235; response rate: 61.8%). As mentioned above, the percentage may lead to substantial impairment in psychomotor functioning, possi- of people with EDS is about 15% of the general population [2,3].The bly increasing the risk of motor vehicle accidents, although various results of this study, however, revealed that around 40.4% of the 3235 prescription hypnotics may be equally effective in promoting sleep respondents had experienced sleepiness while driving [19]. Thus, sleep- [26]. A study on the effects of after-midnight intake of zolpidem and iness can easily occur among ordinary drivers in daily life while driving a temazepam on the driving ability of 18 insomniac women using a driving vehicle. Furthermore, 20.3% of driver's license holders reported having simulator revealed that 5.5 h after administration there were no major fallen asleep at the wheel while 10.4% had experienced near-miss acci- differences in psychomotor performance in women who had taken dents or actually had accidents due to falling asleep at the wheel [19]. these two hypnotics compared with those who had taken a placebo, in- Other studies from Western countries revealed that 29% of subjects dicating an absence of significant residual effects early the next morning had experienced almost falling asleep at the wheel [38] while 18.3% [27]. However, certain subjects were more susceptible than others to the had experienced near misses due to falling asleep at the wheel [39].In drugs' effects. This underlines the necessity to strongly advocate against addition, 1.3% had experienced an actual accident [39].Otherstudies taking hypnotics late at night if patients intend to drive a car early the have reported the percentage of accidents that occur in relation to sleep- next morning. iness to be around 20% [1,40]. Looking at these studies, the occurrence of accidents in Japanese drivers is at least the same as or a little bit higher 3.2. Sleep apnea syndrome than for drivers in Western countries, and there is a clear association between sleepiness and traffic accidents. Sleep apnea syndrome (SAS), with its repeated episodes of nocturnal Factors related to falling asleep at the wheel are being male, having hypoxemia and sleep fragmentation, increases sleepiness, which in turn a driver's license for a long period, longer weekly driving distance, increases inattention while driving [28]. Following the initial report by appearance of sleepiness even after a short driving time, and self- George et al. in 1987 [29], the risk of accident in SAS patients due to awareness of having SAS or has been diagnosed with SAS [19].An 4 Y. Komada et al. / IATSS Research 37 (2013) 1–7

p<0.01 p<0.01 p<0.05 160 p<0.05 25

20 120

15

80 AHI

ESS score 10

40 5

0 0 =0 =1 2 =0 =1 2 Number of car accidents within past 5 years

Fig. 3. Comparison of AHI (left panel) and ESS score (right panel) between groups categorized by number of MVAs (Komada et al., Tohoku Journal of Experimental Medicine. 2009).

Group without MVA: n = 496, Group with one MVA: n = 56, Group with two or more MVAs: n = 18. AHI: F2,567 = 7.2, p b 0.001; ESS: F2,567 = 15.9, p b 0.0001. AHI: Apnea hypopnea index; ESS: Epworth sleepiness scale. analysis conducted on driver's license holders who had SAS revealed orexinergic control of wakefulness, patients with BIISS are unable to that 7.5% responded that they “might suffer from SAS,” whereas 1.1% maintain normal alertness and wakefulness due to a chronic lack of responded that they “were diagnosed with SAS.” However, that this sleep. study was of driver's license holders who had come to the driver's In reports from other countries, 2% of patients who visited a sleep license examination center for license renewal may have introduced a disorder center with any sleep complaint and 6% of those with EDS were sampling bias. Despite this, because the percentage of drivers who diagnosed to have BIISS [44]. Our previous study showed that among had been diagnosed with SAS among those who thought that they the 1243 patients who visited our clinic due to EDS, the percentage of might have SAS remained low, it may be important to promote early patients diagnosed with BIISS was 7.1% [45]. This rate was higher than diagnosis and treatment of SAS in order to prevent as-yet-undiagnosed those for circadian rhythm sleep disorders and periodic limb movement drivers from falling asleep at the wheel. Furthermore, it is also important disorders, despite being lower than the rate for major hypersomnias to disseminate information about the relationship between exhaustion such as sleep apnea syndrome, idiopathic hypersomnia, and narcolepsy due to driving for long periods, falling asleep at the wheel, and accidents (Table 1). Considering this, cases with EDS caused by chronic insuffi- resulting from falling asleep. cient sleep should not be ignored clinically. Furthermore, the rate of the population with chronic EDS in epidemiological surveys is markedly 3.3. Narcolepsy higher than the rate of patients with BIISS who consult medical insti- tutes. That is, most people with BIISS are thought unlikely to seek med- Narcolepsy is a closely related to EDS. Usually, a ical assistance. We believe that it is necessary to raise awareness of the normal person does not fall asleep during driving or while having a significance of chronic sleep insufficiency in the general population. conversation, but people suffering from narcolepsy have such tenden- Symptoms of BIISS include fatigue, irritation, depression, muscle cies. In addition, triggered with certain kinds of emotions pain, concentration disturbance, and a decline in cognitive function may also occur. That is, bodily muscle tone is completely lost when a as well as daytime vigilance [46,47]. Another study confirmed that strong emotional reaction occurs. In general, sleepiness in narcolepsy patients is reportedly stronger than in SAS patients. For this reason, Table 1 the deterioration of driving performance in patients with narcolepsy Diagnoses of 1243 patients with excessive daytime sleepiness. has been systematically studied using a simulator [41,42]. According Komada et al., . 2008. to Aldrich, the motor vehicular accident rate due to sleepiness in the Diagnosis n % general population is around 5%; however, that rate increases up to Obstructive sleep apnea syndrome 431 34.7 50% in narcoleptic patients [11]. Another previous study found that Idiopathic hypersomnia 136 10.9 75% of narcolepsy patients with cataplexy and 50% of those without Narcolepsy 109 8.8 cataplexy had experienced traffic accidents or near misses in the last Behaviorally induced insufficient sleep syndrome 88 7.1 five years [43]. The results of a logistic regression analysis showed Circadian rhythm sleep disorders 76 6.1 Sleep disorders associated with mental disorders 54 4.3 that ESS scores were significantly associated with increased automobile Periodic limb movement disorder or 33 2.7 – accidents and near misses (severe; OR = 14.6, 95% CI: 2.0 108.7) [43]. Insomnia 30 2.4 Parasomnias 14 1.1 3.4. Behaviorally-induced insufficient sleep syndrome Long sleeper 12 1.0 Recurrent hypersomnia 3 0.2 fi More than two diseases causative for daytime sleepiness 69 5.6 In the 2nd edition of the ICSD, behaviorally induced insuf cient (Of which ISS + other hypersomnia) 25 2.0 sleep syndrome (BIISS) is classified as a category of “hypersomnia of Undiagnosed 188 15.1 central origin which is not due to a circadian rhythm sleep disorder, Undiagnosed: Patients who did not have a confirmed diagnosis due to interruption of sleep-related breathing disorder, or other cause of disturbed nocturnal diagnostic examination or treatment, or patients in whom two clinicians did not sleep” [8]. Unlike narcolepsy, which occurs based on decreased completely agree on the final diagnosis. Y. Komada et al. / IATSS Research 37 (2013) 1–7 5 the performance of tasks that require both concentration and vigilance [55], and 24.4% of nurses working with rapid-rotation schedules in was reduced in populations with BIISS [48]. In our study, the ESS score Japan [56]. Our previous study of 3109 subjects engaged in shift work of people with BIISS at an initial assessment was 13.6 (SD = 3.4), as public transportation drivers demonstrated that SWD was the main which was significantly lower than that of narcolepsy patients, signifi- factor of EDS among them (Table 3). Notably, the strength of subjective cantly higher than that of people with SAS and circadian rhythm sleep sleepiness in drivers with SWD was higher than that of OSAS drivers disorders, and on the same level as those with idiopathic hypersomnia [57]. In workers who met the criteria for SWD, there was a high inci- (Table 2). Furthermore, among people with BIISS, 22.1% had experi- dence of accidents related with sleepiness [52], suggesting that early enced driving or work-related accidents or near misses. Considering diagnosis and treatment of SWD is also important for ensuring traffic that the traffic accident rate due to falling asleep remains below 3% in safety. However, SWD is often undiagnosed and untreated [58].Because the general population [49], the accident rate of people with BIISS shift workers are no longer a small minority, and approximately 50% of may be much higher. Compared with the group who had no such professional drivers have irregular schedules or are engaged in night- experiences, the ESS scores of the group that had experienced accidents time driving [59], it is important to perform systematic SWD screenings or near misses was significantly higher (group with accident experience: to maintain both the physical and mental health of workers and to 15.5 (SD = 3.8) points, group without accident experience: 13.0 (SD = ensure traffic safety. As for the treatment of SWD, there are several 3.1) points; p b .01). available methods such as taking a short before or during night shifts, using bright light exposure and melatonin at appropriate times, 3.5. Shift-work disorder and using drugs such as modafinil to stay awake as needed [60].These methods were reported to be effective for maintaining wakefulness The period length of circadian rhythm in humans, including sleep– during shifts or for increasing the amount of nocturnal sleep [58].How- wake pattern, melatonin hormone secretory pattern, and body temper- ever, no established study has investigated how these methods act to ature, exceeds 24 h without light–dark changes and time information, ensure the traffic safety of shift workers. both of which are crucial for maintaining constant circadian rhythm. In normal life, the circadian rhythms of the sleep–wake pattern, body 4. Sleepiness detection methods and accident countermeasures temperature, and melatonin secretion run synchronously. Body tem- perature becomes lowest in the middle of nocturnal sleep, and highest Because ocular variables (e.g., pupils, eye movement, and blinking) during afternoon hours. In addition, the secretion of melatonin starts a change with the level of sleepiness and can be measured without contact few hours before , promoting the induction of human or restraint, several researchers have proposed methods for measuring sleep. However, when a forced change in the sleep-awake pattern drivers' sleepiness using ocular variables [61–65]. In a previous study, (desynchronization) occurs for some reason, unbearable feelings of we investigated the detection capability of ocular variables for decreased sleepiness can occur during the daytime. The most important cause of vigilance or falling asleep while performing monotonous tasks that desynchronized circadian rhythms is rotating shift work. According to can easily induce sleepiness [66]. A cap-type head sensor and an eye- the Ministry of Health, Labor and Welfare, 18.6% of companies in tracking system (EMR-9; NAC Image Technology Inc., Tokyo, Japan) Japan adopted shift-work systems and 17.9% of the working population (Fig. 4) were used to record several ocular variables. Decreased blink worked at night [50], suggesting that this is not a problem only for a frequency and pupil diameter, increased percentage of eyelid closure small minority of workers. time (PERCLOS), and slow eye movement were observed as consecutive Those working on a shift schedule have different circadian rhythm missed responses increased. Among these variables, PERCLOS was the timing than those of day workers. They are not only forced to stay most suitable for detecting missed responses and three or more consec- awake and engaged in their daily work but also prone to develop utive missed responses. This result suggests that, among the ocular insufficient sleep, because sleep after night shifts is shortened [51]. variables we assessed, PERCLOS can most effectively prevent error or According to an analysis of 2570 adult subjects by Drake et al., 36%– accident caused by decreased vigilance while driving. Further study on 45% of those engaged in night work or shift work reported unbearable the development of systems to alert drivers using this variable can be daytime sleepiness [52]. In a study by Scott et al., 600 out of 895 nurses expected to prevent accidents due to sleepiness. reported that they had fallen asleep at the wheel at least once [53].Scott As for countermeasures for drowsy driving, laboratory- and et al. also found that 30 nurses reported driving in a sleepy state after questionnaire-based epidemiological studies have demonstrated the shift work. Shift-work disorder (SWD) is a sleep disorder related to effectiveness of napping and drinking coffee [67–69]. Taking a nap as shift work, one that causes excessive sleepiness and/or complaints of a measure to counter sleepiness at the wheel is common among drivers nocturnal sleep [8]. The prevalence of SWD was 10.1% in day workers between 46 and 64 years of age, those who have already experienced a [54], 32.1% among night workers [54], 28.9% of nurses working on a sleep-related traffic accident, and those who have experienced severe three-shift rotation [55], 44.3% of those working on a two-shift rotation sleepiness during driving [70]. A short nap is thought to be effective in

Table 2 Comparison of descriptive variables among the five groups with major hypersomnias. Komada et al., Sleep Medicine. 2008.

Mean age of Mean age ESS score of BMI Sleep length initial visit of symptoms the initial visit of weekday

Mean SD Mean SD Mean SD Mean SD Mean SD

Behaviorally induced insufficient sleep syndrome 30.2 7.3 28.6 7.5 13.6 3.4 22.2 2.8 5.5 0.8 ⁎⁎⁎ ⁎⁎ ⁎⁎⁎ ⁎⁎⁎ Sleep apnea syndrome 45.1 12.1 –– 12.5 3.6 27.4 4.9 6.1 1.0 ⁎⁎⁎ ⁎⁎⁎ Idiopathic hypersomnia 31.4 9.0 19.2 7.3 14.3 3.0 21.8 3.1 6.3 1.1 ⁎⁎⁎ ⁎⁎⁎ ⁎ ⁎⁎⁎ Narcolepsy 31.0 13.3 17.2 7.5 15.7 3.0 23.4 4.1 6.4 1.3 ⁎⁎⁎ ⁎ ⁎⁎⁎ Circadian rhythm sleep disorders 27.7 8.1 18.9 6.4 12.4 3.7 21.9 3.8 7.8 2.3

Results with post hoc test (compared with ISS).We exclude obstructive sleep apnea syndrome from the analysis of mean age of symptoms because the data for people with this syndrome in the population is unknown. ⁎⁎⁎ p b 0.0001. ⁎⁎ p b 0.01. ⁎ p b 0.05. 6 Y. Komada et al. / IATSS Research 37 (2013) 1–7

Table 3 suggesting that they do not use as a countermeasure until they The diagnoses of sleep disorders according to the International Classification of Sleep have experienced sleepiness-related traffic accidents. Undoubtedly, Disorders 2nd edition among drivers with subjective EDS (ESS ≥ 11). when sleepiness occurs during driving, drivers should take measure Asaoka et al., Journal of Occupational and Environmental Medicine. 2010. such as taking a break or drinking coffee, both of which have been n%reported to successfully reduce sleepiness and increase driving perfor- Shift-work disorder 48 32.7 mance [74]. However, driving schedule constraints on professional Obstructive sleep apnea syndrome 30 20.4 drivers seem to make them less likely than non-professional drivers to fi Behaviorally induced insuf cient sleep syndrome 12 8.2 take proactive measures to counter sleepiness while driving. It is there- Delayed sleep phase disorder 11 7.5 Insomnia 10 6.8 fore important to educate professional drivers and their employees Idiopathic hypersomnia without long sleep time 7 4.8 about the mechanism of sleepiness-related vehicular accidents and Periodic limb movement disorder 5 3.4 the importance of strategies for coping with sleepiness. Other 3 2.0 Error in answering questionnaire 6 4.1 5. Conclusion Lost to follow-up 15 10.2 Total 147 100.0 Sleepiness is a main factor responsible for traffic accidents caused by human error. People who sleep for fewer than 6 h have an increased reducing sleepiness and maintaining the vigilance level. However, sleep risk of traffic accidents. Thus, the most essential thing to prevent traffic inertia sometimes occurs after awakening from a short nap, and this accidents due to sleepiness is ensuring that drivers get sufficient sleep could be a risk factor for accidents [71].Itwasreportedthatashort at night. Of course, to prevent daytime sleepiness, it is important to nap with self-awakening was more effective in reducing sleepiness maintain nocturnal sleep quality as well as an adequate amount of and than one with forced-awakening [71]. Another exper- sleep. SAS is one of the most common and important sleep disorders, imental study showed that sleep inertia after taking a nap during sleep and makes nocturnal sleep shallow and fragmented. Patients affected deprivation can be prevented by hourly caffeine (0.3 mg/kg) intake with severe SAS may be especially likely to underestimate their own [72]. Our study of about 4000 drivers demonstrated that 67.6% had sleepiness. Considering that the risk of traffic accidents can be reduced experienced sleepiness while driving and 11.4% had fallen asleep at close to that of healthy people by providing proper treatment, the the wheel in the year preceding the investigation [73].Inthatstudy, early detection and treatment of SAS is desirable. analysis of two groups—professional (about 700 subjects) and non- Sleepiness-related accidents may have multiple causes. It is therefore professional drivers (about 3300 subjects)—revealed that the rate of important to address this matter from various approaches, including the subjects experiencing traffic accidents caused by falling asleep in the maintenance of good , measures to reduce sleepiness, and last five years, and falling asleep at the wheel during the past year, the prevention and treatment of sleep disorders. In addition, we want to was higher in professional drivers than in non-professional drivers emphasize the need to establish and maintain appropriate working con- (traffic accidents caused by falling asleep: 3.5% vs. 1.2%; falling asleep ditions for drivers. at the wheel: 17.8% vs. 10.1%, respectively). In addition, an analysis of measures taken to reduce sleepiness during driving revealed that the References most frequently used method, used by more than 40% of subject drivers, – “ ” [1] J.A. Horne, L.A. Reyner, Sleep related vehicle accidents, BMJ 310 (1995) 565 567. both non-professional and professional, was taking a rest. Other mea- [2] M.M. Ohayon, R.G. Priest, J. Zulley, S. Smirne, T. Paiva, Prevalence of narcolepsy sures included “chewing gum,”“opening a window,” and “drinking symptomatology and diagnosis in the European general population, Neurology coffee,” all used by 30% of drivers; “napping” and “smoking,” used by 58 (2002) 1826–1833. fi “ ” “ [3] X. Liu, M. Uchiyama, K. Kim, M. Okawa, K. Shibui, Y. Kudo, Y. Doi, M. Minowa, R. 20% of drivers; and nally, listening to the radio and talking with a Ogihara, Sleep loss and daytime sleepiness in the general adult population of passenger,” used by 10% of drivers. In that study, we also found that Japan, Psychiatry Research 93 (2000) 1–11. non-professional drivers are likely to take such measures when they [4] NHK Broadcasting Culture Research Institute, NHK Data Book 2010 National Time feel sleepy or are in situations where prone to dozing off at the wheel. Use SurveyNHK Publishing, Inc., Tokyo, 2011. [5] C. Hublin, J. Kaprio, M. Partinen, M. Koskenvuo, Insufficient sleep—a population- However, this tendency was not observed in professional drivers, based study in adults, Sleep 24 (2001) 392–400. [6] T. Ohida, A.M. Kamal, M. Uchiyama, K. Kim, S. Takemura, T. Sone, T. Ishii, The influence of lifestyle and health status factors on sleep loss among the Japanese general population, Sleep 24 (2001) 333–338. [7] H.P. Van Dongen, G. Maislin, J.M. Mullington, D.F. Dinges, The cumulative cost of additional wakefulness: dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total , Sleep 26 (2003) 117–126. [8] American Academy of Sleep Medicine, International Classification of Sleep Disorders: Diagnostic and Coding Manual, 2nd ed. American Academy of Sleep Medicine, Westchester, Illinois, 2005. [9] T. Abe, Y. Komada, Y. Nishida, K. Hayashida, K. Inoue, Short sleep duration and long spells of driving are associated with the occurrence of Japanese drivers' rear- end collisions and single-car accidents, Journal of Sleep Research 19 (2010) 310–316. [10] T. Abe, Y. Komada, S. Asaoka, A. Ozaki, Y. Inoue, Questionnaire-based evidence of association between sleepiness while driving and motor vehicle crashes that are subjectively not caused by falling asleep, Sleep and Biological Rhythms 9 (2011) 134–143. [11] M.S. Aldrich, Automobile accidents in patients with sleep disorders, Sleep 12 (1989) 487–494. [12] T. Young, J. Blustein, L. Finn, M. Palta, Sleep-disordered breathing and motor vehicle accidents in a population-based sample of employed adults, Sleep 20 (1997) 608–613. [13] K. Kim, M. Uchiyama, M. Okawa, X. Liu, R. Ogihara, An epidemiological study of insomnia among the Japanese general population, Sleep 23 (2000) 41–47. [14] J.C. Souza, L.A. Magna, R. Reimao, Insomnia and use in Campo Grande general population, Brazil, Arquivos de Neuro-Psiquiatria 60 (2002) 702–707. [15] M.M. Ohayon, Epidemiology of insomnia: what we know and what we still need Fig. 4. Recordings using an eye-tracking system (EMR-9; NAC Image Technology Inc., to learn, Sleep Medicine Reviews 6 (2002) 97–111. Tokyo Japan). [16] C.L. Drake, T. Roehrs, T. Roth, Insomnia causes, consequences, and therapeutics: Abe et al., International Journal of Psychophysiology. 2011. an overview, Depression and Anxiety 18 (2003) 163–176. Y. Komada et al. / IATSS Research 37 (2013) 1–7 7

[17] M.M. Ohayon, Epidemiological study on insomnia in the general population, [47] J.E. Broman, L.G. Lundh, J. Hetta, Insufficient sleep in general population, Sleep 19 (1996) S7–S15. Neurophysiologie Clinique 26 (1996) 30–39. [18] S. Englert, M. Linden, Differences in self-reported sleep complaints in elderly persons [48] T. Roehrs, V. Timms, A. Zwyghuizen-Doorenbos, R. Buzenski, T. Roth, Polysomno- living in the community who do or do not take sleep medication, The Journal of graphic, performance, and personality differences of sleepy and alert normals, Clinical Psychiatry 59 (1998) 137–144. Sleep 13 (1990) 395–402. [19] Y. Komada, T. Shiomi, K. Mishima, Y. Inoue, Associated factors for drowsy driving [49] P.M. Turkington, M. Sircar, V. Allgar, M.W. Elliot, Relationship between obstructive among licenced driver, Japanese Journal of Public Health 57 (2010) 1066–1074. sleep apnoea, driving simulator performance, and risk of road trafficaccidents, [20] K. Komada, T. Nomura, M. Kusumi, K. Nakashima, I. Okajima, T. Sasai, Y. Inoue, Thorax 56 (2001) 800–805. Correlations among insomnia symptoms, sleep medication use and depressive [50] Ministry of Health, Labour and Welfare, Annual Health, Labour and Welfare symptoms, Psychiatry and Clinical Neurosciences 65 (2011) 20–29. Report 2007, 2007. [21] D.E. Moul, E.A. Nofzinger, P.A. Pilkonis, P.R. Houck, J.M. Miewald, D.J. Buysse, [51] T. Åkerstedt, Shift work and disturbed sleep/wakefulness, Occupational Medicine Symptom reports in severe chronic insomnia, Sleep 25 (2002) 553–563. 53 (2003) 89–94. [22] E. Fortier-Brochu, S. Beaulieu-Bonneau, H. Ivers, C.M. Morin, Relations between [52] C.L. Drake, T. Roehrs, G. Richardson, J.K. Walsh, T. Roth, Shift work sleep disorder: sleep, fatigue, and health-related quality of life in individuals with insomnia, Journal prevalence and consequences beyond that of symptomatic day workers, Sleep 27 of Psychosomatic Research 69 (2010) 475–483. (2004) 1453–1462. [23] J.D. Edinger, M.K. Means, C.E. Carney, A.D. Krystal, Psychomotor performance [53] L.D. Scott, W.T. Hwang, A.E. Rogers, T. Nysse, G.E. Dean, D.F. Dinges, The relationship deficits and their relation to prior nights' sleep among individuals with primary between nurse work schedules, sleep duration, and drowsy driving, Sleep 30 (2007) insomnia, Sleep 31 (2008) 599–607. 1801–1807. [24] J.A. Shekleton, N.L. Rogers, S.M. Rajaratnam, Searching for the daytime impairments [54] L. Di Milia, S. Waage, S. Pallesen, B. Bjorvatn, Shift work disorder in a random of primary insomnia, Sleep Medicine Reviews 14 (2010) 47–60. population sample-prevalence and comorbidities, PloS One 8 (2013) e55306. [25] E. Altena, Y.D. Van Der Werf, R.L. Strijers, E.J. Van Someren, Sleep loss affects [55] E. Flo, S. Pallesen, N. Magerøy, B.E. Moen, J. Grønli, I. Hilde Nordhus, B. Bjorvatn, vigilance: effects of chronic insomnia and sleep therapy, Journal of Sleep Research Shift work disorder in nurses-assessment, prevalence and related health problems, 17 (2008) 335–343. PloS One 7 (2012) e33981. [26] J. Menzin, K.M. Lang, P. Levy, E. Levy, A general model of the effects of sleep [56] S. Asaoka, S. Aritake, Y. Komada, A. Ozaki, Y. Odagiri, S. Inoue, T. Shimomitsu, Y. medications on the risk and cost of motor vehicle accidents and its application Inoue, Factors associated with shift work disorder in nurses working with to France, PharmacoEconomics 19 (2001) 69–78. rapid-rotation schedules in Japan: the nurses' sleep health project, Chronobiology [27] M. Partinen, K. Hirvonen, C. Hublin, M. Halavaara, H. Hiltunen, Effects of after- International 30 (2013) 628–636. midnight intake of zolpidem and temazepam on driving ability in women with [57] S. Asaoka, K. Namba, S. Tsuiki, Y. Komada, Y. Inoue, Excessive daytime sleepiness non-organic insomnia, Sleep Medicine 4 (2003) 553–561. among Japanese public transportation drivers engaged in shiftwork, Journal of [28] C.F. George, Sleep apnea, alertness, and motor vehicle crashes, American Journal Occupational and Environmental Medicine 52 (2010) 813–818. of Respiratory and Critical Care Medicine 176 (2007) 954–956. [58] J.R. Schwartz, T. Roth, Shift work sleep disorder: burden of illness and approaches [29] C.F. George, P.W. Nickerson, P.J. Hanly, T.W. Millar, M.H. Kryger, Sleep apnoea to management, Drugs 66 (2006) 2357–2370. patients have more automobile accidents, Lancet 330 (1987) 447. [59] A.I. Pack, D.F. Dinges, G. Maislin, A study of prevalence of sleep apnea among [30] S. Tregear, J. Reston, K. Schoelles, B. Phillips, Obstructive sleep apnea and risk of commercial truck drivers, Federal Motor Carrier Safety Administration Publication motor vehicle crash: systematic review and meta-analysis, Journal of Clinical No DOT-RT-02-030 Washington, DC, 2002. Sleep Medicine 5 (2009) 573–581. [60] R.L. Sack, D. Auckley, R.R. Auger, M.A. Carskadon, K.P. Jr, M.V. Vitiello Wright, I.V. [31] A. Sassani, L.J. Findley, M. Kryger, E. Goldlust, C. George, T.M. Davidson, Reducing Zhdanova, Circadian rhythm sleep disorders: part I, basic principles, shift work motor-vehicle collisions, costs, and fatalities by treating obstructive sleep apnea and jet lag disorders: An American Academy of Sleep Medicine review, Sleep 30 syndrome, Sleep 27 (2004) 453–458. (2007) 1460–1483. [32] Y. Komada, Y. Nishida, K. Namba, T. Abe, S. Tsuiki, Y. Inoue, Elevated risk of motor [61] T. Åkerstedt, M. Ingre, G. Kecklund, A. Anund, D. Sandberg, M. Wahde, P. Philip, P. vehicle accident for male drivers with obstructive sleep apnea syndrome in the Kronberg, Reaction of sleepiness indicators to partial sleep deprivation, time of Tokyo metropolitan area, The Tohoku Journal of Experimental Medicine 219 day and time on task in a driving simulator: the DROWSI project, Journal of (2009) 11–16. Sleep Research 19 (2010) 298–309. [33] S. Tregear, J. Reston, K. Schoelles, B. Phillips, Continuous positive airway pressure [62] D.F. Dinges, M. Mallis, G. Maislin, J.W. Powell, Evaluation of Techniques for Ocular reduces risk of motor vehicle crash among drivers with obstructive sleep apnea: Measurement as an Index of Fatigue and as the Basis for Alertness Management, U.S. systematic review and meta-analysis, Sleep 33 (2010) 1373–1380. Department of Transportation, National Highway Traffic Safety Administration, [34] M. Orth, H.W. Duchna, M. Leidag, W. Widdig, K. Rasche, T.T. Bauer, J.W. Walther, J. 1998. de Zeeuw, J.P. Malin, G. Schultze-Werninghaus, S. Kotterba, Driving simulator and [63] M.W. Johns, A. Tucker, R. Chapman, K. Crowley, N. Michael, Monitoring eye and neuropsychological testing in OSAS before and under CPAP therapy, The European eyelid movements by infrared reflectance oculography to measure drowsiness Respiratory Journal 26 (2005) 898–903. in drivers, Somnologie 11 (2007) 234–242. [35] T. Young, P.E. Peppard, D.J. Gottlieb, Epidemiology of obstructive sleep apnea: a [64] D. Shin, H. Sakai, Y. Uchiyama, Slow eye movement detection can prevent population health perspective, American Journal of Respiratory and Critical Care sleep-related accidents effectively in a simulated driving task, Journal of Sleep Medicine 165 (2002) 1217–1239. Research 20 (2011) 416–424. [36] R.L. Ellen, S.C. Marshall, M. Palayew, F.J. Molnar, K.G. Wilson, M. Man-Son-Hing, [65] A. Ueno, S. Kokubun, Y. Uchikawa, A prototype real-time system for assessing Systematic review of motor vehicle crash risk in persons with sleep apnea, Journal vigilance levels and for alerting the subject with sound stimulation, International of Clinical Sleep Medicine 2 (2006) 193–200. Journal of Assistive Robotics and Mechatronics 8 (2007) 19–27. [37] A.T. Mulgrew, G. Nasvadi, A. Butt, R. Cheema, N. Fox, J.A. Fleetham, C.F. Ryan, P. [66] T. Abe, T. Nonomura, Y. Komada, S. Asaoka, T. Sasai, A. Ueno, Y. Inoue, Detecting Cooper, N.T. Ayas, Risk and severity of motor vehicle crashes in patients with deteriorated vigilance using percentage of eyelid closure time during behavioral obstructive sleep apnoea/hypopnoea, Thorax 63 (2008) 536–541. maintenance of wakefulness tests, International Journal of Psychophysiology 82 [38] G. Maycock, Sleepiness and driving: the experience of UK car drivers, Journal of (2011) 269–274. Sleep Research 5 (1997) 229–237. [67] J. Horne, L. Reyner, Vehicle accidents related to sleep: a review, Occupational and [39] N.B. Powell, K.B. Schechtman, R.W. Riley, C. Guilleminault, R.P. Chiang, E.M. Environmental Medicine 56 (1999) 289–294. Weaver, Sleepy driver near-misses may predict accident risks, Sleep 30 (2007) [68] P. Philip, J. Taillard, N. Moore, S. Delord, C. Valtat, P. Sagaspe, B. Bioulac, The effects 331–342. of coffee and napping on nighttime highway driving: a randomized trial, Annals [40] S. Garbarino, L. Nobili, M. Beelke, F. De Carli, F. Ferrillo, The contributing role of of Internal Medicine 144 (2006) 785–791. sleepiness in highway vehicle accidents, Sleep 24 (2001) 203–206. [69] P. Sagaspe, J. Taillard, G. Chaumet, N. Moore, B. Bioulac, P. Philip, Aging and [41] L. Findley, M. Unverzagt, R. Guchu, M. Fabrizio, J. Buckner, P. Suratt, Vigilance and nocturnal driving: better with coffee or a nap? A randomized study, Sleep 30 automobile accidents in patients with sleep apnea or narcolepsy, Chest 108 (2007) 1808–1813. (1995) 619–624. [70] A. Anund, G. Kecklund, B. Peters, T. Åkerstedt, Driver sleepiness and individual [42] C.F. George, A.C. Boudreau, A. Smiley, Comparison of simulated driving performance differences in preferences for countermeasures, Journal of Sleep Research 17 (2008) in narcolepsy and sleep apnea patients, Sleep 19 (1996) 711–717. 16–22. [43] A. Ozaki, Y. Inoue, T. Nakajima, K. Hayashida, M. Honda, Y. Komada, K. Takahashi, [71] K. Kaida, H. Nittono, M. Hayashi, T. Hori, Effects of self-awakening on sleep structure Health-related quality of life among drug-naïve patients with narcolepsy with of a daytime short nap and on subsequent arousal levels, Perceptual and Motor Skills cataplexy, narcolepsy without cataplexy, and idiopathic hypersomnia without 97 (2003) 1073–1084. long sleep time, Journal of Clinical Sleep Medicine 4 (2008) 572–578. [72] H.P. Van Dongen, N.J. Price, J.M. Mullington, M.P. Szuba, S.C. Kapoor, D.F. Dinges, [44] T. Roehrs, F. Zorick, J. Sicklesteel, R. Wittig, T. Roth, Excessive daytime sleepiness Caffeine eliminates psychomotor vigilance deficits from sleep inertia, Sleep 24 associated with insufficient sleep, Sleep 6 (1983) 319–325. (2001) 813–819. [45] Y. Komada, Y. Inoue, K. Hayashida, T. Nakajima, M. Honda, K. Takahashi, Clinical [73] S. Asaoka, T. Abe, Y. Komada, Y. Inoue, The factors associated with preferences for significance and correlates of behaviorally induced insufficient sleep syndrome, napping and drinking coffee as countermeasures for sleepiness at the wheel Sleep Medicine 9 (2008) 851–856. among Japanese drivers, Sleep Medicine 13 (2012) 354–361. [46] M.S. Aldrich, Insufficient sleep syndrome, in: M. Billiard (Ed.), Sleep: Physiology, [74] S.V. Heatherle, Caffeine withdrawal, sleepiness, and driving performance: what Investigations, and Medicine, Kluwer Academic/Plenum Publishers, New York, does the research really tell us? Nutritional Neuroscience 14 (2011) 89–95. 2003, pp. 341–346.