Assessment of Screening Tests for Sleep Apnea Syndrome in the Workplace

Transcription

1 J Occup Health 2010; 52: Journal of Occupational Health Assessment of Screening Tests for Sleep Apnea Syndrome in the Workplace Shigemi TANAKA 1, 2 and Masayuki SHIMA 2 1 Tanaka Internal Medicine Clinic and 2 Department of Public Health, Hyogo College of Medicine, Japan Abstract: Assessment of Screening Tests for Sleep Apnea Syndrome in the Workplace: Shigemi TANAKA, et al. Tanaka Internal Medicine Clinic Objectives: The Epworth sleepiness scale (ESS) is often used for screening of sleep apnea syndrome (SAS) in the workplace. We conducted pulse oximetry for workers of a large transportation company, who were selected based on their response to a questionnaire that included ESS and breathing pattern during sleep. Methods: Pulse oximetry was performed for 803 of 3,761 male workers. A 3% oxygen desaturation index (ODI) 15 was defined to represent sleep disordered breathing (SDB). We evaluated the frequencies of severe snoring and sleep apnea in the past three months reported by the bed partner or a family member. The relationships between 3% ODI 15 and various parameters were analyzed. Results: Of the 715 workers with valid recordings, 3% ODI was 15 in 108. The prevalence of 3% ODI 15 was high among workers with a body mass index (BMI) 25 kg/ m 2, or for whom severe snoring ( 4 days/wk) or sleep apnea ( 2 days/mo) was reported, but did not correlate with ESS scores. Multivariate analysis showed that 3% ODI 15 correlated significantly with BMI, and reported severe snoring and sleep apnea, but not ESS scores. The prevalence of 3% ODI 15 was high among obese workers. ESS scores did not correlate with 3% ODI 15, irrespective of obesity. Conclusions: Our findings suggest that severe snoring, sleep apnea and BMI, but not ESS, are useful screening tools for SAS. (J Occup Health 2010; 52: ) Key words: Body mass index (BMI), Epworth sleepiness scale (ESS), Pulse oximetry, Sleep apnea, Sleep apnea syndrome (SAS), Snoring Received Dec 22, 2008; Accepted Nov 26, 2009 Published online in J-STAGE Jan 28, 2010 Correspondence to: S. Tanaka, Department of Public Health, Hyogo College of Medicine, 1 1 Mukogawa-cho, Nishinomiya, Hyogo , Japan ( 7787may@kcc.zaq.ne.jp) Subjects with sleep apnea syndrome (SAS) develop frequent episodes of apnea and hypopnea during sleep and have daytime symptoms of functional impairment 1). These subjects are at risk of traffic accidents, and the incidence of traffic accidents is about four times higher in SAS subjects than in healthy individuals 2 4). SAS is not only a risk factor for industrial accidents, but it is also a cause of low productivity. The complications associated with SAS, including sleep-disordered breathing (SDB), include hypertension, ischemic heart disease, cerebrovascular disease and diabetes mellitus 5 8). Previous studies have indicated that treatment of SAS by continuous positive airway pressure (CPAP) can reduce the incidence of traffic accidents to levels comparable to those of healthy individuals 9, 10). Therefore, in the transportation industry, it is essential to screen and diagnose SAS in drivers and provide appropriate therapy to ensure safe operation in addition to maintaining the overall health of employees. The standard test for the diagnosis of SAS is an overnight polysomnography (PSG). However, in the workplace, it is difficult to perform PSG on all subjects due to time, labor and cost restraints. Consequently, PSG is performed only on high-risk workers identified by a screening test. The Epworth sleepiness scale (ESS) 11) is often used as an inexpensive and convenient test to screen for SAS by assessing daytime sleepiness, in the workplace 12). However, the validity of using the ESS for SAS screening as part of safety management in the workplace has not been sufficiently assessed. We conducted a primary screening survey on sleep using a questionnaire that included the ESS at a largescale passenger transportation company, together with overnight non-invasive measurement of arterial oxygen saturation by pulse oximetry as a secondary screening test in workers selected according to the results of the questionnaire. Pulse oximetry has been used for the assessment of SAS and hypoventilation based on falls in arterial oxygen saturation during respiratory events 13) and is considered a useful screening tool for SAS 14, 15). To assess the usefulness of the questionnaire for SAS

2 100 J Occup Health, Vol. 52, 2010 Table 1. The items listed in questionnaire for the primary screening 1. Average sleep time in the last three months: ( ) h/day 2. Frequency of severe snoring, based on information obtained from a bed partner a 1 day per mo or less b 2 3 days/mo c 1 3 days/wk d 4 5 days/wk e approximately every day f unknown 3. Frequency of sleep apnea, based on information obtained from a bed partner a 1 day/mo or less b 2 3 days/mo c 1 3 days/wk d 4 5 days/wk e approximately every day f unknown 4. Frequency of dozing off or falling asleep in the following situations (The score from 0 to 3 was chosen for each situation: 0=never doze; 1=slight chance of dozing; 2=moderate chance of dozing; 3=high chance of dozing) a Sitting and reading b Watching TV c Sitting, inactive in a public place (e.g. a theater or a meeting) d As a passenger in a car for an hour without a break e Lying down to rest in the afternoon when circumstances permit f Sitting and talking to someone g Sitting quietly after a lunch without alcohol h In a car, while stopped for a few minutes in the traffic 5. Smoking status: Nonsmoker; Ex-smoker; Smoker screening, we evaluated the relationship between the results of pulse oximetry and questionnaire data including the ESS. Subjects and Methods Study population The study subjects were male workers (N=3,917, age: yr) involved in attentive operations who were employed by a passenger transportation company in Osaka, Japan. The subjects were selected from the entire work force (N=7,692) of the company, in a preferential screening for SAS to ensure safe operation 12). The subjects consisted of train drivers, rail conductors, bus drivers, and mechanics. As a primary screening test for SAS, a questionnaire survey was conducted during regular medical check-ups in September and October Then, as a secondary screening test for the workers selected according to the results of the questionnaire survey, overnight pulse oximetry was performed between April 2005 and March Primary screening test The questionnaire used in the primary screening contained the items listed in the ESS, as well as the following items: average sleep time in the last three months and smoking status (Table 1). The ESS consists of 8 questions related to daytime sleepiness, each of which allocates a score from 0 to 3. The maximum score of the ESS is 24 points, and a score of 11 indicates excessive daytime sleepiness, and suggests possible SAS 11). Workers with an ESS score of 11 were encouraged to undergo overnight pulse oximetry. In addition, frequency of snoring and sleep apnea in the last three months was requested from sleeping partners or family members. The response to the question on the frequency of severe snoring and sleep apnea included 1 day per month or less, 2 3 days per month, 1 3 days per week, 4 5 days per wk, and approximately every day, based on information obtained from subjects bed partner or family members. Workers with at least four days a wk of severe snoring or at least two days a month of sleep apnea were also advised to undergo overnight pulse oximetry. Secondary screening test Overnight pulse oximetry was employed as the secondary screening test for SAS, using a pulse oximeter

3 Shigemi TANAKA, et al.: Screening Tests for Sleep Apnea Syndrome 101 (Pulsox-3, Konica Minolta, Tokyo, Japan) with the SR- 5C finger clip probe. Subjects were instructed on its usage and then provided with a pulse oximeter for use at home (wear the sensor on the finger before going to sleep and remove it the next morning). Pulse oximetry was conducted during the night of a weekday, and valid data consisted of at least five hours of recording during sleep. The number of oxygen desaturations per hour (oxygen desaturation index, ODI) was used as an indicator of SDB. A 3% ODI was selected as an index of oxygen desaturation, representing the number of per hour recording times in which the oxygen saturation fell by 3% 6). In previous studies 16 18), a 3% ODI 15 was considered equivalent to an apnea-hypopnea index (AHI/ h) 20, as determined by PSG, with a recommendation of treatment with CPAP for subjects showing AHI/h 20. Therefore, in the present study, a 3% ODI 15 was defined as positive for SDB. Statistical analysis and ethical considerations Valid pulse oximetry data were analyzed for the relationship between 3% ODI 15 and work conditions (shift work), smoking status, sleep time, severe snoring, sleep apnea and ESS scores by the χ 2 test. The relationships of age and body mass index (BMI) with 3% ODI 15 were evaluated by Fisher s exact test and the Cochran-Armitage test, respectively. We estimated the sensitivity and specificity for 3% ODI 15 in relation to sleep condition and BMI. Next, to evaluate the effects of various factors on 3% ODI 15, a multiple logistic regression analysis was performed by including 3% ODI 15 as the dependent variable and the above-mentioned factors as independent variables. Furthermore, workers were divided into two groups with respect to BMI (BMI 25 kg/m 2 and BMI <25 kg/m 2 ), and multiple logistic regression analyses were performed. A p value <0.05 denoted the presence of a significant difference. Statistical analyses were performed using JMP 7 software (SAS Institute Inc., Cary, NC). The protocol of this study was explained in detail to the subjects, and written informed consent was obtained from all subjects. This study was carried out in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Review Board of the Japan Society for Occupational Health. Results Valid questionnaires were obtained from 3,761 workers (response rate: 96.0%) in the target population. The responders comprised 728 train drivers, 611 rail conductors, 1,197 bus drivers, and 1,225 mechanics. Their mean age ± SD was 39.9 ± 8.2. Based on the analysis of the response to the questionnaires, 969 workers were selected for pulse oximetry (695 workers with severe snoring 4 days/wk, 441 workers with sleep apnea 2 days/mo, and 155 workers with ESS scores 11). Only 803 workers consented to pulse oximetry, and the oximetry data of 88 workers were excluded because no valid data were collected due to a short recording time, improper use of the oximeter or cancellation on the day of the test. Thus, data of pulse oximetry from 715 workers (110 train drivers, 83 rail conductors, 410 bus drivers, and 112 mechanics) were available for analysis in the present study. Table 2 shows the results of 3% ODI according to age, BMI, work condition, smoking status, sleep time, ESS score, and report of severe snoring and sleep apnea. The prevalence of 3% ODI 15 increased significantly with BMI. In addition, the prevalence was significantly higher in subjects who reported severe snoring ( 4 days/wk) and sleep apnea ( 2 days/mo). The result of 3% ODI did not correlate with age, work condition, smoking status, sleep time, or ESS score. Table 3 shows the sensitivity and the specificity for 3% ODI 15 in relation to sleep condition and BMI. The sensitivities of BMI 25, ESS score 11, report of severe snoring ( 4 days/wk) and sleep apnea ( 2 days/mo) were 74.1%, 16.7%, 93.4%, and 75.3%, respectively. The specificities of these factors were 55.7%, 82.4%, 21.8%, and 53.3%, respectively. The sensitivity of ESS score 11 was extremely low. The combination of the above items except ESS score 11 increased the specificities but decreased the sensitivities. For BMI 25 and/or reported severe snoring ( 4 days/wk), the sensitivity improved to 99.0%. When at least one of the above three items was considered, the sensitivity was still 99.0%, and the specificity diminished to 8.5%. Multiple logistic regression analysis was used to assess the relationships between 3% ODI 15 and the above factors (Table 4). A significant relationship was seen between 3% ODI 15 and BMI 25 (odds ratio [OR]: 3.69, 95% confidence interval [CI]: ); reported severe snoring ( 4 days/wk, OR: 2.95, 95% CI: ); and reported sleep apnea ( 2 days/mo, OR: 2.79, 95% CI: ). A significant relationship was also seen with age 50 yr (OR: 1.74, 95% CI: ). A 3% ODI 15 did not relate to work condition, smoking status, sleep time, or ESS score. Workers were further divided into two groups based on BMI (BMI 25 and BMI <25 kg/m 2 ), and multiple logistic regression analysis for 3% ODI 15 was conducted (Table 5). For the BMI 25 group, 3% ODI 15 correlated significantly with reported sleep apnea ( 2 days/mo, OR: 2.53, 95% CI: ). For the BMI <25 group, 3% ODI 15 correlated significantly with reported severe snoring ( 4 days/wk, OR: 9.12, 95% CI: ) and sleep apnea ( 2 days/mo, OR: 8.81, 95% CI: ). When compared to the BMI 25 group, the OR was higher for both factors. For the BMI <25 group, the OR for current smoking was significantly

4 102 J Occup Health, Vol. 52, 2010 Table 2. Analysis of factors that influence the results of pulse oximetry Variable 3% ODI p value (N=108) (N=607) N (%) N (%) Age (yr) (9.1) 20 (90.9) (13.3) 202 (86.7) (14.0) 277 (86.0) (21.7) 108 (78.3) Shift duty Yes 35 (15.0) 198 (85.0) No 73 (15.1) 409 (84.9) Smoking Yes 49 (12.9) 330 (87.1) No 59 (17.6) 277 (82.4) BMI (kg/m 2 ) >30 32 (36.4) 56 (63.6) < (18.4) 213 (81.6) <25 28 (7.7) 338 (92.3) Average sleep time (h) 6 79(15.9) 418 (84.1) (13.3) 189 (86.7) ESS score (14.4) 107 (85.6) (15.3) 500 (84.7) Severe snoring 4 days/wk 99 (17.9) 453 (82.1) < days/wk 7 (5.3) 126 (94.7) unknown* 2 (6.7) 28 (93.3) Sleep apnea 2 days/mo 73 (22.1) 258 (77.9) < day/mo 24 (7.5) 295 (92.5) unknown* 11 (16.9) 54 (83.1) : p value by Fisher s exact test; :p value by the Cochran-Armitage test. *: Unknown responses were mainly due to lack of report from the bed partner or family members of the subjects. ODI, oxygen desaturation index; BMI, body mass index; ESS Epworth sleepiness scale. Table 3. Sensitivity and specificity for 3% ODI 15, in relation to reported sleep findings and BMI Sensitivity (%) Specificity (%) BMI 25 kg/m ESS Severe snoring 4 days/wk Sleep apnea 2 days/mo BMI 25 and severe snoring 4 days/wk BMI 25 and sleep apnea 2 days/mo Severe snoring 4 days/wk and sleep apnea 2 days/mo BMI 25, severe snoring 4 days/wk, and sleep apnea 2 days/mo BMI 25 and/or severe snoring 4 days/wk BMI 25 and/or sleep apnea 2 days/mo Severe snoring 4 days/wk and/or sleep apnea 2 days/mo BMI 25 and/or severe snoring 4 days/wk and/or sleep apnea 2 days/mo For abbreviations, see Table 2.

5 Shigemi TANAKA, et al.: Screening Tests for Sleep Apnea Syndrome 103 Table 4. Odds ratio (OR)* and 95% confidence interval (95% CI) of various factors on 3%ODI 15 Variable OR (95% CI) p value Age 50 yr 1.74 ( ) Shift duty 1.05 ( ) Smoking 0.73 ( ) BMI 25 kg/m ( ) <0.001 Sleep time 6 h 1.15 ( ) ESS ( ) Severe snoring 4 days/wk 2.95 ( ) <0.001 Sleep apnea 2 days/mo 2.79 ( ) <0.001 *: Odds ratios adjusted for all variables in the table using the logistic regression model. The analysis was performed using data of 647 subjects who underwent pulse oximetry. The analysis excluded data of 68 subjects due to lack of information on sleep apnea and snoring from the bed partner or family member. For abbreviations, see Table 2. Table 5. Odds ratio (OR)* and 95% confidence interval (95% CI) for various factors on 3%ODI 15 in subjects divided according to BMI Variables BMI 25 kg/m 2 (N=310) BMI <25 kg/m 2 (N=337) OR (95% CI) p value OR (95% CI) p value Age 50 yr 1.48 ( ) ( ) Shift duty 1.05 ( ) ( ) Smoking 0.88 ( ) ( ) Sleep time 6 h 1.09 ( ) ( ) ESS ( ) ( ) Severe snoring 4 days/wk 2.23 ( ) ( ) Sleep apnea 2 days/mo 2.53 ( ) ( ) <0.001 *: Odds ratios adjusted for all variables in the table using the logistic regression model. The analysis excluded data of 68 subjects due to lack of information on sleep apnea and snoring from the bed partner or family member. For abbreviations, see Table 2. small (0.43, 95% CI: ). For both the BMI 25 and BMI <25 groups, the ORs for 50 yr of age were not significant, although the values were larger than 1. Discussion Because the incidence of traffic accidents is high among SAS subjects, it is imperative for transportation companies to assess their employees for SAS and provide appropriate therapy 10, 12, 19). In the present study, we conducted a questionnaire survey and pulse oximetry studies of employees of a Japanese transportation company. The results showed that 3% ODI 15 correlated significantly with reported severe snoring and sleep apnea. The ESS score has been used as an indicator of daytime sleepiness, and is reported to correlate with the hypopnea index and reductions in SaO 2 and to reflect PSG data in SAS subjects 4, 11, 20). However, in the present study, no relationship was noted between 3% ODI 15 and the ESS score. In most previous studies, the relationships between the ESS score and test findings including PSG were evaluated in subjects with clinical symptoms such as daytime sleepiness. However, the present study was conducted during regular check-ups in the workplace, and most of the subjects had severe snoring and sleep apnea as reported by their bed partner. The 3% ODI 15 did not correlate with the ESS score, but with reported severe snoring and sleep apnea. Furthermore, multiple logistic regression analysis showed a significant relationship between severe snoring and sleep apnea, and these factors correlated more closely with 3% ODI 15 than the ESS score. Pulse oximetry is used to monitor instantaneous changes in arterial oxygen saturation during sleep, and is used as a screening tool for SAS/SDB 13 15). In this study, a small and lightweight watch-type pulse oximeter equipped with sufficient memory was used. Such a device does not affect sleep, and testing is relatively inexpensive.

6 104 J Occup Health, Vol. 52, 2010 Pulse oximetry has been widely used as a screening tool at many workplaces in Japan 12, 14, 21). With a pulse oximeter, a dip is defined as a transient decrease in arterial oxygen saturation of more than 3% followed by correction within two minutes, and the number of dips (3%ODI) per hour is used to assess the severity of SDB. Nakamata et al. 16) simultaneously performed PSG and pulse oximetry and reported that when the cutoff value of 3% ODI is set at 15, the specificity and sensitivity for 20 AHI (indication for CPAP therapy) are 100% and approximately 90%, respectively. One objective of the present study was to minimize the time wasted in the transition from screening to diagnosis and treatments for SAS, and accordingly 3% ODI 15 was defined as positive for the diagnosis and treatment 15). In a large-scale epidemiological study published by Young et al. 1), SDB correlated with BMI (obese individuals), age (elderly individuals), and work conditions (shift workers). In the present study, 3% ODI 15 correlated with BMI and age ( 50 yr), in agreement with the results of Young et al. 1) However, our index of SDB did not correlate with work conditions or sleep time. This may be because our subjects did not engage in allnight work. Numerous studies have identified obesity as a risk factor for SAS/SDB 1, 15, 21 23). In the present study, BMI 25 showed a strong relationship with 3% ODI 15 (OR, 3.69). Of the 108 workers with 3% ODI 15, the BMI was less than 25 kg/m 2 in 28 workers. Subjects were divided into two BMI groups (BMI 25 and BMI <25 kg/m 2 ), and multiple logistic regression analysis showed that the ORs for severe snoring and sleep apnea in relation to 3% ODI 15 for the BMI <25 group were markedly greater than those for the BMI 25 group. Therefore, SAS is not limited to obese people alone: even for non-obese people, SAS should be suspected in the presence of reported severe snoring or sleep apnea. Our results also show that, among non-obese workers, current smoking lowered the probability of 3% ODI 15. The reason for this finding is not clear, but smoking status may confound with other factors in relation to ODI. We found no difference in BMI between smokers and nonsmokers. Smoking has been considered to increase the risk of SDB 24). On the other hand, a recent study in Japan reported that the risk of SDB was lower among current smokers than never or former smokers, although the difference was not significant 25). Further studies are needed to analyze the relationship between smoking, respiration during sleep and SAS. In the present study, no relationship was seen between ESS score and 3% ODI 15, irrespective of obesity. The present study had several limitations. First, because we conducted pulse oximetry only for workers selected according to the result of a questionnaire survey, the subjects had at least either severe snoring 4 days per week, sleep apnea 2 days per month, or a high ESS score. Therefore, we did not obtain information on workers without these signs in the questionnaire. Second, because severe snoring and sleep apnea were evaluated dichotomously for logistic regression analysis, the relationship between 3% ODI 15 and the extent of these symptoms could not be evaluated. Third, because severe snoring and sleep apnea were confirmed according to a report from the bed partner or other family members, they cannot be evaluated objectively. In addition, the same information was not available on subjects who had no bed partner. Lastly, pulse oximetry may inherently underestimate respiratory disturbance in young people and non-obese people 6, 14, 16). Because the ventilatory capacity of young subjects is high, oxygen saturation rapidly recovers with respiration over a short period of time. In the present study, the cutoff value for 3% ODI, as determined by pulse oximetry, was set at 15, but it is necessary to investigate carefully the validity of pulse oximetry taking into account BMI and other factors 15, 16, 21). It is important for large corporations to set measures for the diagnosis and treatment of SAS as part of health management in the workplace, particularly when safe operations are required 12). When the present study was conducted, the types of monitoring devices for SAS/SDB screenings were limited, and we determined that pulse oximetry would be suitable in terms of convenience and cost-effectiveness. In recent years, flow sensors (placed in the nose and mouth to monitor airflow during sleep) that objectively determine apnea and hypopnea during sleep have been developed and are reported to be more feasible for screening of SDB than pulse oximetry 26, 27). The ideal SAS/SDB screening test allows for efficient transition to subsequent comprehensive tests and treatment with as little as possible waiting time. Furthermore, in individuals with positive screening results, it will be necessary to conduct tests for prompt diagnosis and then provide appropriate treatment. Based on the results of the present study, the use of ESS alone as the primary screening test for SAS may result in high false negative rates, suggesting poor suitability as a screening tool. However, reported severe snoring and sleep apnea related significantly to 3% ODI 15, with high sensitivities, and BMI related significantly to the result of pulse oximetry. Similar results were obtained in a previous study conducted in a Japanese workplace, although the criteria for SDB somewhat differed 14). Furthermore, 28 non-obese workers had 3% ODI 15, and the relationship between the results of pulse oximetry of these individuals and reported severe snoring and sleep apnea was more pronounced than in obese workers. When severe snoring and sleep apnea were combined with BMI, the sensitivity for 3% ODI 15 improved markedly, though with a decrease in specificity.

7 Shigemi TANAKA, et al.: Screening Tests for Sleep Apnea Syndrome 105 Of the subjects with 3% ODI 15, only one did not confirm the criteria of severe snoring, sleep apnea, or BMI 25. These results emphasize the usefulness of reported severe snoring and sleep apnea in SAS screening 28). In conclusion, our questionnaire used for SAS screening was useful for assessing SDB in terms of severe snoring and sleep apnea as well as BMI. The combination of these three items markedly increased the sensitivity for SDB. The ESS should be used to monitor symptoms, not as a screening tool for SAS. Acknowledgment: The authors are grateful to all the staff of the transportation company involved in this study. The authors also thank Ms. Yoshiko Yoda for her excellent secretarial assistance. References 1) Young T, Palta M, Dempsy J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993; 328: ) Partinen M, Guilleminault C. Daytime sleepiness and vascular morbidity at seven-year follow-up in obstructive sleep apnea patients. Chest 1990; 97: ) Young T, Peppard P, Palta M. Population-based study of sleep-disordered breathing as a risk factor for hypertension. Arch Intern Med 1997; 157: ) Chin K, Ohi M. Obesity and obstructive sleep apnea syndrome. Intern Med 1999; 38: ) Peppard PE, Yong T, Palta M, Skatrud J. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000; 342: ) Tanigawa T, Tachibana N, Yamagishi K, et al. Relationship between sleep-disordered breathing and blood pressure levels in community-based samples of Japanese men. Hypertens Res 2004; 27: ) Shahar E, Whitney CW, Redline S, et al. Sleepdisordered breathing and cardiovascular disease: Crosssectional results of the Sleep Heart Health Study. Am J Respir Crit Care Med 2001; 163: ) Elmasry A, Janson C, Lindberg E, Gislason T, Tageldin MA, Boman G. The role of habitual snoring and obesity in the development of diabetes: A 10-year follow-up study in a male population. J Intern Med 2000; 248: ) Teran-Santos J, Jimenez-Gomez A, Cordero-Guevara J. The association between sleep apnea and the risk of traffic accidents. Cooperative Group Burgos-Santander. N Engl J Med 1999; 340: ) Philip P. Sleepiness of occupational drivers. Ind Health 2005; 43: ) Johns MW. A new method for measuring daytime sleepiness: The Epworth sleepiness scale. Sleep 1991; 14: ) Tanigawa T, Horie S, Sakurai S, Iso H. Screening for sleep-disordered breathing at workplaces. Ind Health 2005; 43: ) Netzer N, Eliasson AH, Netzer C, Kristo DA. Overnight pulse oximetry for sleep-disordered breathing in adults: A review. Chest 2001; 120: ) Niijima K, Enta K, Hori H, Sashihara S, Mizoue T, Morimoto Y. The usefulness of sleep apnea syndrome screening using a portable pulse oximeter in the workplace. J Occup Health 2007; 49: ) Dingli K, Coleman EL, Vennelle M, et al. Evaluation of a portable device for diagnosing the sleep apnoea/ hypopnoea syndrome. Eur Respir J 2003; 21: ) Nakamata M, Kubota Y, Sakai K, et al. The limitation of screening test for patients with sleep apnea syndrome using pulse oximetry. Nihon Kokyu Kanri Gakkaishi 2003; 12: (in Japanese). 17) Meoli AL, Casey KR, Clark RW, et al. Hypopnea in sleep-disordered breathing in adults. Sleep 2001; 24: ) Fietze I, Dingli K, Diefenbach K, et al. Night-to-night variation of the oxygen desaturation index in sleep apnoea syndrome. Eur Respir J 2004; 24: ) Powell NB, Schechtman KB, Riley WP, Guilleminault C, Chiang P, Weaver EM. Sleepy driver near-misses may predict accident risks. Sleep 2007; 30: ) Sanford SD, Lichstein KL, Durrence HH, Riedel BW, Taylor DJ, Bush AJ. The influence of age, gender, ethnicity, and insomnia on Epworth sleepiness scores: A normative US population. Sleep Med 2006; 7: ) Niijima K, Mizoue T, Kawashima M, Enta K, Sashihara S, Morimoto Y. Predisposing factors of sleepdisordered breathing in Japanese male workers. J Occup Health 2007; 49: ) Nakano H, Ikeda T, Hayashi M, et al. Effect of body mass index on overnight oximetry for the diagnosis of sleep apnea. Respir Med 2004; 98: ) Kim HC, Young T, Matthews CG, Weber SM, Woodward AR, Palta M. Sleep-disordered breathing and neuropsychological deficits. A population-based study. Am J Respir Crit Care Med 1997; 156: ) Wetter DW, Young TB, Bidwell TR, Badr MS, Palta M. Smoking as a risk factor for sleep-disordered breathing. Arch Intern Med 1994; 154: ) Nakayama-Ashida Y, Takegami M, Chin K, et al. Sleepdisordered breathing in the usual lifestyle setting as detected with home monitoring in a population of working men in Japan. Sleep. 2008; 31: ) Nakano H, Tanigawa T, Furukawa T, Nishima S. Automatic detection of sleep-disordered breathing from a single-channel airflow record. Eur Respir J 2007; 29: ) Nakano H, Tanigawa T, Ohnishi Y, et al. Validation of a single-channel airflow monitor for screening of sleepdisordered breathing. Eur Respir J 2008; 32: ) Yue W, Hao W, Liu P, Liu T, Ni M, Guo Q. A casecontrol study on psychological symptoms in sleep apnea-hypopnea syndrome. Can J Psychiatry 2003; 48: