Issue Navigator

Volume 11 No. 04
Earn CME
Accepted Papers

Scientific Investigations

Predictors for Progression of Sleep Disordered Breathing among Public Transport Drivers: A 3-Year Follow-Up Study

Cheng-Yu Lin, MD1,2,3; Tung-Sheng Shih, ScD4,5; Saou-Hsing Liou, MD6; Ming-Hsiu Lin, PhD5; Cheng-Ping Chang, PhD7; Tzu-Chieh Chou, PhD8,9
1Department of Otolaryngology, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan; 2Department of Environmental and Occupational Health, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan; 3Department of Otolaryngology, Tainan Hospital, Ministry of Health and Welfare, Tainan, Taiwan; 4Department of Occupational Safety and Health, College of Public Health, China Medical University, Taichung, Taiwan; 5Institute of Labor, Occupational Safety and Health, Ministry of Labor, Executive Yuan, New Taipei, Taiwan; 6Division of Environmental Health and Occupational Medicine, National Health Research Institutes, Miaoli, Taiwan; 7Department of Occupational Safety and Health, Chang Jung Christian University, Tainan, Taiwan; 8Department of Public Health, College of Public Health, China Medical University, Taichung, Taiwan; 9Department of Health Risk Management, College of Public Health, China Medical University, Taichung, Taiwan


Study Objectives:

Sleep disordered breathing (SDB) is associated with an increased risk of motor vehicle crashes. This study aimed to understand SDB progression and related factors among professional drivers.


A total of 524 professional male drivers from a transportation company were included in this study. These drivers completed overnight in-home pulse oximetry studies both in 2006 and in 2009. Participants with abnormal results (oxygen desaturation index [ODI] ≥ 10 events/h) comprised the SDB group. Data included questionnaire information on demographics, medical history, SDB symptoms, and anthropometric measurements.


A total of 318 male workers were recruited for further analysis. Fifty of these workers belonged to the SDB group. Workers with untreated SDB significantly progressed to a more severe state after three years. Baseline body mass index (BMI), baseline ODI, and change in BMI were all significant positive predictors of SDB progression (β = 0.823, 0.242, and 1.626; p = 0.047, 0.013, and 0.004, respectively). Compared with non-SDB drivers, SDB subjects showed a greater proportion of newly diagnosed cardiovascular disease (38.0%) at follow-up.


Untreated SDB was a gradually progressive disorder in professional drivers over a three-year period. Subjects with high BMI and moderate to severe SDB should be closely monitored to allow for early detection of worsening SDB. Weight control should be highlighted in the management of SDB.


A commentary on this article appears in this issue on page 409.


Lin CY, Shih TS, Liou SH, Lin MH, Chang CP, Chou TC. Predictors for progression of sleep disordered breathing among public transport drivers: a 3-year follow-up study. J Clin Sleep Med 2015;11(4):419–425.

Sleep disordered breathing (SDB) is a condition of repetitive episodes of decreased or arrested respiratory airflow during sleep.1 SDB produces diurnal and nocturnal symptoms, including excessive daytime sleepiness, sleep attacks, psychomotor deficits, and disrupted nighttime sleep caused by frequent arousal.2 Patients with clinically diagnosed SDB perform poorly in driving simulation tests3 and tend to have a two- to seven-fold increased risk of at-fault motor vehicle crashes (MVCs).46


Current Knowledge/Study Rationale: To prevent serious chronic illness and an increased risk of motor vehicle crashes (MVCs) because of sleep disordered breathing (SDB), the progression of SDB should be understood and factors associated with disease worsening should be identified. We examined the natural evolution of SDB among public transport drivers, and identified the determinants of the three-year changes in disease severity.

Study Impact: Public transport drivers, especially those who are heavy and have moderate to severe SDB, should be closely monitored by regular follow-up to allow for early detection of worsening SDB. In addition, this study provided further evidence that drivers with SDB had significantly increased cardiovascular risk, and a large prospective cohort study should be performed to further explain the SDB and cardiovascular disease association in order to develop effective preventive strategies.

The prevalence of SDB in the general population ranges from 2% to 32.8%.79 In studies on transportation workers that mostly focused on professional automotive drivers, such as bus and truck drivers, the prevalence of SDB ranged from 15.8% to 78%.1014 Furthermore, some studies have shown a dose-response relationship between SDB severity and MVC risk.1517 Aldrich showed that the proportion of severe sleep apneics who have had sleep-related accidents is almost twice that of patients with mild or moderate apnea.18 Transportation drivers operate larger vehicles that may contain hazardous chemicals or a large number of passengers. The fatality risk for occupants of other vehicles involved in accidents is quite high: trucks > 10,000 pounds are seven times more likely to be fatal to other motorists than to the truck occupants themselves.19,20 Given that public safety is of extreme importance, early diagnosis and treatment of SDB among transportation workers are necessary.

In past decades, the natural history of SDB in the general population has been investigated by many researchers.2026 Abundant data from several large studies, including the Wisconsin Sleep Cohort Study and the Sleep Heart Health Study, demonstrated that SDB progressed over time.25,26 In the Wisconsin Sleep Cohort Study, 690 randomly selected, employed Wisconsin residents (mean age, 46 years; 56% male) were evaluated twice at four-year intervals for SDB.25 The mean (SD) apnea-hypopnea index (AHI) ranged from 4.1 (9.1) to 5.5 (10.8) events/h. In the Sleep Heart Health Study, a total of 2,968 US participants (mean age, 62 years; 45% male) were assessed at two polysomnography (PSG) examinations approximately five years apart.26 During the follow-up period, the mean (SD) change in AHI was +3.4 (12.4) in men and +2.2 (9.0) in women. The results demonstrated that changes in body weight were related to an increase in SDB, and this association was stronger in men than in women. As we know, these large investigations were all community-based studies. Compared to the general population, public transport drivers endure special working conditions, such as irregular shifts, monotonous and sedentary operations, and unfixed meal times. These factors may influence the natural history of their SDB progression. In the interest of public safety, it is worthwhile to evaluate whether the outcomes of community-based studies can be extrapolated to public transport drivers.

To prevent serious chronic illness and an increased risk of MVC because of SDB, the progression of SDB should be understood and factors associated with disease worsening should be identified. The progressive nature of the disease requires early and frequent follow-up visits for patients with mild cases who did not receive treatment; a potentially effective treatment plan should also be implemented at an early stage.27,28 We examined the natural evolution of SDB among public transport drivers in Taiwan, and identified the determinants of the three-year changes in disease severity.


Subject Selection

The subjects were recruited from a public transportation company. Of the total number of screened subjects, 524 employees were screened for SDB in 2006 (Figure 1). Subjects who had surgical procedures on nasal or pharyngeal tissues or underwent continuous treatment with either continuous positive airway pressure (CPAP) or oral appliances were excluded. All subjects were free of any medicines or drugs to improve sleep quality or sleep behavior. The nocturnal oxygen desaturation at home was measured in 2006 by a portable pulse oximeter. Data from questionnaires of sleep quality and physical examinations were also collected. In 2009, data from the at-home nocturnal oxygen desaturations, questionnaires, and physical examinations were obtained again from the same subjects. After analysis of the nocturnal oxygen desaturation measurements, data collected from 2 groups were compared. The 2 groups were the non-SDB group (defined as oxygen de-saturation index [ODI] dropping by ≥ 3% fewer than 10 times in both 2006 and 2009, n = 268), and the SDB group (defined as ODI dropping by ≥ 3% 10 or more times in both 2006 and 2009, n = 50). All subjects provided informed consent for the use of their data from the questionnaire and physical examination. The study protocol was approved by the medical ethics committee of the China Medical University Hospital. The study was conducted according to the principles in the Declaration of Helsinki.

Flow diagram of subject selection in the study.


Figure 1

Flow diagram of subject selection in the study.

(more ...)

Pulse Oximetry

Nocturnal oxygen desaturation was measured at night by a portable pulse oximeter (PulSox-3 iA, Minolta, Osaka, Japan). Subjects were instructed on the usage of the device and were provided oximeters for unattended monitoring of SpO2 and pulse rate while sleeping at home. The pulse oximeter sensor was mounted on one fingertip, and the pulse oximeter system was attached to the distal forearm. All values were continuously measured and recorded every 5 s for the entire night during testing periods. Data from the oximeter were collected and stored in the data logger, which could input data into a computer with a suitable interface. After analysis by appropriate software, the pulse rate, SpO2 values, ODI, lowest SpO2 value, and cumulative percentage of time spent at saturation below 88% were calculated. All testing times had to be > 180 min in duration for effective measurement. In this study, ODI was the diagnostic criterion for SDB. Subjects with SDB had ≥ 10 events per hour of desaturation by ≥ 3% from baseline (ODI 3% ≥ 10).


Subjects were asked to fill out a questionnaire that included personal information, smoking and drinking statuses, family health history, sleep quality, and sleep-related problems. The Pittsburg Sleep Quality Index (PSQI) was used to evaluate the quality of sleep. With respect to sleep-related problems, the Epworth Sleepiness Scale (ESS) was used to determine the daytime sleepiness status of subjects. The Snore Outcomes Survey (SOS) was used to evaluate the severity of snoring during sleep periods.

Physical Examination

We acquired characteristics of the subjects by annual physical examination. All occupational workers were required to undergo regular physical examinations yearly. The parameters related to basic characteristics and biochemical indices of health and disease status were evaluated by biochemical tests of urine and blood samples collected during the annual physical examination. Such parameters included body mass index (BMI), blood pressure, blood glucose, neck/waist circumference, and blood lipid content. These parameters were used in further data analysis. Hypertension, cardiovascular disease (CVD), diabetes mellitus, and metabolic syndrome were also identified by a physician during examination. The presence of CVD was defined as follows: documented hospital admission for myocardial infarction, stroke, and acute coronary syndromes requiring coronary artery bypass surgery or coronary angioplasty.

Data Analysis

All data are presented as frequencies for nominal variables and as mean ± standard deviation (SD) for continuous variables. For all parameters examined in this study, the Shapiro-Wilk normality test was used to evaluate whether the values of testing variables followed a normal distribution. For continuous and binary data, Student t-test and the χ2 test, respectively, were used to compare the differences between the 2 groups. The paired t-test was used to compare the physical examination results in the initial study and those after 3 years. Finally, univariate and multivariate regression models were used to determine the relationship between the difference of ODI and physical examination results in 2006 and at follow-up after three years (in 2009). Statistical significance was set at p < 0.05. Statistical analysis was performed using Jump 5.0 software (SAS Institute Inc., Cary, NC, USA).


Of the 524 subjects who participated in the SDB screening performed in 2006, 383 subjects were included in our follow-up investigation in 2009. As shown in Figure 1, 99 subjects were excluded from this study because of sleep treatments and the use of medications during the screening survey periods. An additional 42 subjects were excluded to avoid confounding effects because of the lack of information about disease history. The study questionnaire did not ask participants to report the number of crashes (or even near crashes). The SDB screening results demonstrate 316 subjects with ODIs < 10 events/h and 67 subjects with ODIs > 10 events/h. In 2009, all 383 subjects underwent a second screening for SDB. We found 268 subjects with ODIs < 10 events/h, and this group was designated as the non-SDB group. The 50 subjects with ODIs > 10 events/h were defined as the SDB group. A total of 318 male workers were analyzed during follow-up. The average age was 46.8 ± 6.1 years, and the mean BMI was 24.6 ± 2.9 kg/m2. The baseline characteristics of the non-SDB and the SDB groups are shown in Table 1. Age, BMI, and neck/waist circumference were significantly greater in the SDB group (p < 0.05) than in the non-SDB group. For sleep-related assessments, ODI and SOS scores were significantly more severe in the SDB group (p < 0.001) than in the non-SDB group. In addition, the ESS was completed by all study subjects. The mean score of all drivers was 7.8; there was no significant difference between the SDB and non-SDB groups. The percentage of participants who suffered from hypertension or metabolic syndrome up until the study period was also higher in the SDB group than in the non-SDB group (p < 0.001).

Baseline characteristics of the study population (n = 318).



table icon
Table 1

Baseline characteristics of the study population (n = 318).

(more ...)

The comparison of each clinical variable between 2006 and 2009 is listed in Table 2. For the entire population, BMI, neck/waist circumference, fasting blood glucose, ODI, and PSQI significantly increased between the 2 time points. Figure 2 demonstrates that the SDB group showed a significant increase in ODI from 2006 to 2009 (p = 0.004). Further comparison between the non-SDB and SDB groups revealed that the increases in BMI and ODI were significantly greater in the SDB group (p = 0.019 and p < 0.001, respectively) than in the non-SDB group (Table 2). In the SDB group, body weight increased by a mean of 4.2 kg during the 3-year follow-up period, which was significantly greater than in the non-SDB group (0.9 kg) (p = 0.008). Neither the incidence of hypertension nor that of metabolic syndrome changed significantly from 2006 to 2009. However, the SDB group had a greater proportion of newly diagnosed CVD at follow-up than the non-SDB group (p = 0.003). In other words, the incidence of CVD events per 100 people during the 3-year follow-up was significantly higher in the SDB group than in the non-SDB group (14.1 and 6.4 events per 100 person-years, respectively).

Characteristics of the change from baseline to three years of follow-up in the study population (n = 318).



table icon
Table 2

Characteristics of the change from baseline to three years of follow-up in the study population (n = 318).

(more ...)

Change in ODI of the non-SDB group and the SDB group over the 3-year follow-up period.

Error bars show standard deviation. ODI, oxygen desaturation index; SDB, sleep disordered breathing.



Figure 2

Change in ODI of the non-SDB group and the SDB group over the 3-year follow-up period. Error bars show standard deviation. ODI, oxygen desaturation index; SDB, sleep disordered breathing.

(more ...)

For workers with untreated SDB, univariate analysis showed that the baseline variables, including BMI (p = 0.007) and ODI (p = 0.003), were positively associated with the difference in ODI, whereas age, neck circumference, waist circumference, and SOS score were not associated at all (Table 3). The difference in ODI was significantly associated with the change in BMI (p = 0.011). After adjusting for confounding factors, multiple regression analysis indicated that baseline BMI, baseline ODI, and change in BMI were significant positive predictors for SDB progression (β = 0.823, 0.242, and 1.626; p = 0.047, 0.013, and 0.004, respectively) (Table 4).

Correlation between the variables and the difference of nocturnal ODI after three years among the SDB subjects (n = 50).



table icon
Table 3

Correlation between the variables and the difference of nocturnal ODI after three years among the SDB subjects (n = 50).

(more ...)

Multiple regression model of the difference of nocturnal ODI after three years among the SDB subjects (n = 50).



table icon
Table 4

Multiple regression model of the difference of nocturnal ODI after three years among the SDB subjects (n = 50).

(more ...)


In this cohort of 383 transport drivers who underwent in-home oximetry on two occasions three years apart (2006 and 2009), 50 subjects with untreated SDB progressed to a more severe state. Baseline BMI, baseline ODI, and change in BMI were all significant positive predictors of SDB progression. A greater proportion of SDB subjects who had no CVD at baseline developed this complication during the follow-up period compared with non-SDB subjects.

A significant increase in ODI from 20.8 events/h to 24.9 events/h was observed in these SDB subjects; however, as per AHI criteria on PSG, there may be concern that these values fall in the same category of disease severity and thus one could argue they do not represent major progression. A larger study population would allow for detection of smaller effects. Nonetheless, we believe that the observed 4.1 events/h difference during the three-year follow-up is real and represents a significant (19.7%) increase in the SDB severity. Similar to the results of many previous studies,21,22,2426 we found that untreated SDB in transportation workers worsened over a three-year period. However, another study obtained opposing results.23 This discrepancy could be attributed to differences in the characteristics of the study population and the definition of deterioration. One of the possible factors that contributed to the differences in results was the consistency and accuracy of measurements during follow-up. Svanborg and Larsson showed that the conditions on the initial and follow-up nights were not identical; patients underwent oximetry screening on the first occasion but had full PSG on the second occasion.21 In some other studies, PSG was performed at baseline and repeated at follow-up.2226 In our study, a portable pulse oximeter was implemented consistently at the two time points. Another possible factor that contributed to conflicting results was the degree of the severity of SDB in the populations studied. The mean baseline ODI for our SDB group was similar to that in the studies of Svanborg and Larsson and Pendlebury et al. at 20.8, 10.1, and 21.8 events/h, respectively.21,22 However, in the study by Sforza et al., no deterioration in ODI over time was observed, and the mean baseline ODI was considerably high at 52.2 events/h.23 This finding indicated that further deterioration does not occur at high levels of ODI. The indefinite progression of SDB observed by Sforza et al. was difficult for them to explain.

We found a positive correlation between the changes in baseline BMI and ODI, and our results were in agreement with those of Svanborg and Larsson.21 With the mean baseline BMI of 27.1 kg/m2, Svanborg and Larsson found that BMI and individual changes in ODI were significantly correlated. Similarly, in the Cleveland Family Study with 486 cohort members (mean BMI of 27.6 kg/m2), the median five-year change in SDB severity varied with obesity.24 In the Wisconsin Sleep Cohort Study with 690 participants (mean BMI of 29.0 kg/m2), baseline BMI was a significant predictor of AHI change. These results indicated an expected increase of approximately 1% in AHI for each increment of 1 kg/m2 in baseline BMI.25 In contrast, several studies reported no significant correlation between BMI and SDB progression (mean BMI of 28.9 kg/m2 to 30.7 kg/m2).22,23,29,30 Some of our patients developed increased respiratory disturbance with low baseline BMI. Further studies should determine whether factors other than baseline BMI are important in the evolution of SDB.

In this study, participants with a large increase in BMI appeared to be especially at risk for rapid SDB progression. Peppard et al. also noted that a change in weight was related to changes in AHI in a dose-response manner.25 Adjusting for covariates, each percentage change in weight was associated with a mean ∼3% change in AHI. Within the range of ± 20% weight change, the association of weight change and SDB change was well described by a linear function. Furthermore, in the Sleep Heart Health Study with 2,968 participants, results revealed a tendency for acceleration of SDB worsening with more extreme weight gain (> 10 kg) during a five-year follow-up period.26 Similarly, in a longitudinal case study of 160 untreated adult male patients at the Israel Sleep Disorders Unit, snorers and patients with mild and moderate SDB showed an increase in AHI, which was correlated with an increase in BMI. However, patients with severe SDB had an insignificant decrease in AHI, suggesting a ceiling effect for SDB severity.30 The longitudinal associations are consistent with a causal association between obesity and SDB, but the direction of causality can be in either or both directions. SDB develops with weight gain but promotes further weight gain because of the high stress and lower daytime energy levels associated with chronically disrupted sleep.31 Our results emphasize the need to implement weight control early as part of SDB management among professional drivers.

The effects of SDB on cardiovascular health continue to receive widespread attention. Prospective cohort studies aiming to determine whether SDB is causally associated with cardiovascular risk have been published.3234 A previous meta-analysis reported that SDB is associated with a significantly increased risk of developing CVD (RR = 2.48, 95% CI: 1.98–3.10).34 The incidence of CVD per 100 people during a 10-year follow-up period was significantly higher in patients with untreated severe SDB than that in age-matched and BMI-matched healthy participants recruited from the general population (2.13 and 0.45, respectively).32

In the Multi-Ethnic Study of Atherosclerosis (MESA) cohort with 5,338 participants, results revealed the incidence of CVD during the 7.5-year follow-up was 0.8 events per 100 person-years.33 Compared with normal participants, those with physician-diagnosed sleep apnea had higher incident CVD rates (hazard ratio = 2.23). In another prospective cohort study of the cardiovascular consequences of SDB, the Sleep Heart Health Study, a total of 4,422 subjects free of CVD at baseline were followed for a median of 8.7 years.26 The rate of incident CVD was 1.2 events per 100 person-years. It also demonstrated that the highest risk for incident CVD existed in individuals with moderate to severe sleep apnea. Similarly, our observations show that the percentage of newly diagnosed CVD among SDB drivers was more than two-fold higher than that among non-SDB drivers. The complex mechanisms underlying the increase in consequent cardiovascular events in SDB patients are not well understood. SDB acutely leads to intermittent hypoxia, intrathoracic pressure swings, and recurrent arousals, which result in endothelial dysfunction, an increase in the development of atherosclerosis, and cardiovascular events.35

The golden standard of SDB diagnosis is overnight PSG testing.36 However, overnight PSG is a highly technical, time-consuming, and costly procedure. In recent years, the portable pulse oximeter has been used as a simple alternative to PSG in diagnosing SDB.37,38 For early detection and treatment of severe SDB in work sites, the pulse oximeter is also a practical approach to screen for SDB.39 To evaluate the diagnostic performance of the pulse oximeter in detecting SDB, Chung et al. demonstrated that the cutoff of ODI > 10 has high sensitivity (93.3%) in detecting moderate and severe SDB; it also has reasonable specificity (74.6%).40 Therefore, in the present study, ODI ≥ 10 was used as a criterion for diagnosing SDB in a professional driver population.

In 2006, only 99 workers (99/524, 18.9%) received treatment for SDB; similarly, in 2009, there were 17 patients (17/67, 25.4%) with ODI > 10 who received further management. The possible reasons for these lower treatment percentages include: (1) a lack of insight into SDB and difficulty in convincing these patients to receive appropriate treatments; and (2) although CPAP is the most common and effective nonsurgical treatment for moderate and severe SDB, most patients who were recommended CPAP simply rejected treatment before even trying or soon after trialing CPAP. The associated complaints related to expense, inconvenience, poor mask fit and discomfort, skin irritation, mask leaks and sore eyes, airway drying, complaints of noise and frequent awakening, and dislike of CPAP treatment. Therefore, appropriate education, mask and pressure selection, early interventions to address problems and side effects, and follow-up support would be helpful for achieving good treatment outcomes.

This study had several limitations. We used results from a portable pulse oximeter to detect SDB. The oximeter only measured the changes in oxygen saturation, and did not monitor nasal flow, respiratory effort, and total sleep time. The oximeter could not distinguish SDB from other nonbreathing-related sleep disorders. In a meta-analysis study, data demonstrated that AHI values derived from portable sleep studies are 10% lower on average compared with those from hospital-based PSG studies.41 However, the lower cost of the portable oximeter makes it a viable screening tool for suspected SDB patients in workplaces. Given that our results were not confirmed by type I or type III diagnostic tools, future assessment of workers by using hospital-based PSG is warranted. Another limitation in this study was that it involved two visits across a relatively short follow-up period. Forming conclusions regarding the natural history of SDB was difficult. However, considering that 318 subjects participated in the study, we believe in the importance of these findings. In addition, all patients were eligible, and no patient dropped out because of missing data or technical problems. The third limitation of the study was that patients'/family physicians' reports used as the source of diagnosis of CVD possibly resulted in under-diagnosis because probably only symptomatic patients were diagnosed. This under-diagnosis may represent the lower limit of cardiovascular complications in patients with SDB.42 Including a cardiologic evaluation in the protocol may improve the accuracy of evaluations of cardiovascular complications. However, a new onset diagnosis of these complications can be made in all patients, so the results have not been skewed in any particular direction.29

In conclusion, SDB gradually progressed from 2006 to 2009 among the professional male drivers included in this study. Public transport drivers, especially those who are heavy and have moderate to severe SDB, should be closely monitored by regular follow-up to allow for early detection of worsening SDB. Weight control could be an important strategy for reducing the severity and progression of SDB. In addition, this study provided further evidence that drivers with SDB had significantly increased cardiovascular risk. A large prospective cohort study should be performed to further explain the SDB-CVD association in order to develop effective preventive strategies.


This was not an industry supported study. This work was supported by grants from the China Medical University and Hospital [grant number: CMU 100-SR-66 and CMU 100-TS-09] and the Institute of Labor, Occupational Safety and Health, Ministry of Labor [grant number: IOSH99-M321, IOSH100-M321]. The authors have indicated no financial conflicts of interest.


The authors thank Professor Yueliang Leon Guo for his comments and Ms. Chia-Li Ye and Ms. Yu-Jie Huang for their assistance in the preparation of this manuscript.



Iber C, Ancoli-Israel S, Chesson A, Quan SF, authors; for the American Academy of Sleep Medicine. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications. 1st ed. Westchester, IL: American Academy of Sleep Medicine, 2007.


Flemons WW, author. Clinical practice. Obstructive sleep apnea. N Engl J Med. 2002;347:498–504. [PubMed]


George CF, Boudreau AC, Smiley A, authors. Simulated driving performance in patients with obstructive sleep apnea. Am J Respir Crit Care Med. 1996;154:175–81. [PubMed]


George CF, Nickerson PW, Hanly PJ, Millar TW, Kryger MH, authors. Sleep-apnea patients have more automobile accidents. Lancet. 1987;2:447.


Stoohs RA, Guilleminault C, Itoi A, Dement WC, authors. Traffic accidents in commercial long-haul truck drivers: the influence of sleep-disordered breathing and obesity. Sleep. 1994;17:619–23. [PubMed]


Young T, Blustein J, Finn L, Palta M, authors. Sleep-disordered breathing and motor vehicle accidents in a population-based sample of employed adults. Sleep. 1997;20:608–13. [PubMed]


Ancoli-Israel S, Kripke DF, Klauber MR, Mason WJ, Fell R, Kaplan O, authors. Sleep-disordered breathing in community-dwelling elderly. Sleep. 1991;14:486–95. [PubMed Central][PubMed]


Ip MS, Lam B, Lauder IJ, et al., authors. A community study of sleep-disordered breathing in middle-aged Chinese men in Hong Kong. Chest. 2001;119:62–9. [PubMed]


Young T, Peppard PE, Gottlieb DJ, authors. Epidemiology of obstructive sleep apnea: a population health perspective. Am J Respir Crit Care Med. 2002;165:1217–39. [PubMed]


Pack AI, Dinges D, Maislin G, authors. A study of prevalence of sleep apnea among commercial truck drivers. Washington DC: Federal Motor Carrier Safety Administration, 2002.


Howard ME, Desai AV, Grunstein RR, et al., authors. Sleepiness, sleep-disordered breathing, and accident risk factors in commercial vehicle drivers. Am J Respir Crit Care Med. 2004;170:1014–21. [PubMed]


Santos EH, de Mello MT, Pradella-Hallinan M, Luchesi L, Pires ML, Tufik S, authors. Sleep and sleepiness among Brazilian shift-working bus drivers. Chronobiol Int. 2004;21:881–8. [PubMed]


Tregear S, Reston J, Schoelles K, Phillips B, authors. Obstructive sleep apnea and risk of motor vehicle crash: systematic review and meta-analysis. J Clin Sleep Med. 2009;5:573–81. [PubMed Central][PubMed]


Koyama RG, Esteves AM, Silva LOE, et al., authors. Prevalence of and risk factors for obstructive sleep apnea syndrome in Brazilian railroad workers. Sleep Med. 2012;13:1028–32. [PubMed]


Findley LJ, Fabrizio M, Thommi G, Suratt PM, authors. Severity of sleep apnea and automobile crashes. N Engl J Med. 1989;320:868–9.


George CF, Smiley A, authors. Sleep apnea & automobile crashes. Sleep. 1999;22:790–5. [PubMed]


Haraldsson PO, Carenfelt C, Diderichsen F, Nygren A, Tingvall C, authors. Clinical symptoms of sleep apnea syndrome and automobile accidents. ORL J Otorhinolaryngol Relat Spec. 1990;52:57–62. [PubMed]


Aldrich MS, author. Automobile accidents in patients with sleep disorders. Sleep. 1989;12:487–94. [PubMed]


National Highway Traffic Safety Administration. Traffic safety facts 2001. Washington, DC: U.S. Department of Transportation, 2002.


Collop N, Hartenbaum N, Rosen I, Phillips B, authors. Paying attention to at-risk commercial vehicle operators. Chest. 2006;130:637–9. [PubMed]


Svanborg E, Larsson H, authors. Development of nocturnal respiratory disturbance in untreated patients with obstructive sleep apnea syndrome. Chest. 1993;104:340–3. [PubMed]


Pendlebury ST, Pepin JL, Veale D, Levy P, authors. Natural evolution of moderate sleep apnoea syndrome: significant progression over a mean of 17 months. Thorax. 1997;52:872–8. [PubMed Central][PubMed]


Sforza E, Addati G, Cirignotta F, Lugaresi E, authors. Natural evolution of sleep apnoea syndrome: a five year longitudinal study. Eur Respir J. 1994;7:1765–70. [PubMed]


Redline S, Schluchter MD, Larkin EK, Tishler PV, authors. Predictors of longitudinal change in sleep-disordered breathing in a nonclinic population. Sleep. 2003;26:703–9. [PubMed]


Peppard PE, Young T, Palta M, Dempsey J, Skatrud J, authors. Longitudinal study of moderate weight change and sleep-disordered breathing. JAMA. 2000;284:3015–21. [PubMed]


Newman AB, Foster G, Givelber R, Nieto FJ, Redline S, Young T, authors. Progression and regression of sleep-disordered breathing with changes in weight: the Sleep Heart Health Study. Arch Intern Med. 2005;165:2408–13. [PubMed]


Lavie P, author. Sleep medicine--time for a change. J Clin Sleep Med. 2006;2:207–11. [PubMed]


Li AM, Au CT, Ng SK, et al., authors. Natural history and predictors for progression of mild childhood obstructive sleep apnoea. Thorax. 2010;65:27–31. [PubMed]


Fisher D, Pillar G, Malhotra A, Peled N, Lavie P, authors. Long-term follow-up of untreated patients with sleep apnoea syndrome. Respir Med. 2002;96:337–43. [PubMed]


Berger G, Berger R, Oksenberg A, authors. Progression of snoring and obstructive sleep apnoea: the role of increasing weight and time. Eur Respir J. 2009;33:338–45. [PubMed]


Spiegel K, Leproult R, Van Cauter E, authors. Impact of sleep debt on metabolic and endocrine function. Lancet. 1999;354:1435–9. [PubMed]


Marin JM, Carrizo SJ, Vicente E, Agusti AG, authors. Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: an observational study. Lancet. 2005;365:1046–53. [PubMed]


Yeboah J, Redline S, Johnson C, et al., authors. Association between sleep apnea, snoring, incident cardiovascular events and all-cause mortality in an adult population: MESA. Atherosclerosis. 2011;219:963–8. [PubMed Central][PubMed]


Dong JY, Zhang YH, Qin LQ, authors. Obstructive sleep apnea and cardiovascular risk: meta-analysis of prospective cohort studies. Atherosclerosis. 2013;229:489–95. [PubMed]


Kohler M, Stradling JR, authors. Mechanisms of vascular damage in obstructive sleep apnea. Nat Rev Cardiol. 2010;7:677–85. [PubMed]


Berry RB, Budhiraja R, Gottlieb DJ, et al., authors. Rules for scoring respiratory events in sleep: update of the 2007 AASM Manual for the Scoring of Sleep and Associated Events. Deliberations of the Sleep Apnea Definitions Task Force of the American Academy of Sleep Medicine. J Clin Sleep Med. 2012;8:597–619. [PubMed Central][PubMed]


Tanigawa T, Horie S, Sakurai S, Iso H, authors. Screening for sleep-disordered breathing at workplaces. Ind Health. 2005;43:53–7. [PubMed]


Tanigawa T, author. Obstructive sleep apnea: its prevention and screening may contribute to the prevention of hypertension, diabetes and cardiovascular diseases. EPMA J. 2011;2:83–9. [PubMed Central][PubMed]


Niijima K, Mizoue T, Kawashima M, Enta K, Sashihara S, Morimoto Y, authors. Predisposing factors of sleep-disordered breathing in Japanese male workers. J Occup Health. 2007;49:461–6. [PubMed]


Chung F, Liao P, Elsaid H, Islam S, Shapiro CM, Sun YM, authors. Oxygen desaturation index from nocturnal oximetry: a sensitive and specific tool to detect sleep-disordered breathing in surgical patients. Anesth Analg. 2012;114:993–1000. [PubMed]


Ghegan MD, Angelos PC, Stonebraker AC, Gillespie MB, authors. Laboratory versus portable sleep studies: a meta-analysis. Laryngoscope. 2006;116:859–64. [PubMed]


Trilling JS, Froom J, authors. The urgent need to improve hypertension care. Arch Fam Med. 2000;9:794–801. [PubMed]