ADVERTISEMENT

Issue Navigator

Volume 10 No. 02
Earn CME
Accepted Papers
Classifieds







Scientific Investigations

Self-evaluated and Close Relative-Evaluated Epworth Sleepiness Scale vs. Multiple Sleep Latency Test in Patients with Obstructive Sleep Apnea

http://dx.doi.org/10.5664/jcsm.3446

Yun Li, M.D.1; Jihui Zhang, M.D., Ph.D.2; Fei Lei, M.D.1; Hong Liu, M.D.1,3; Zhe Li, M.D.1; Xiangdong Tang, M.D., Ph.D.1
1Sleep Medicine Center, West China Hospital, Sichuan University, Chengdu, Sichuan, China; 2Department of Psychiatry, The Chinese University of Hong Kong, Shatin, Hong Kong SAR, China; 3Department of Internal Medicine, First People's Hospital of Yibin, Yibin, Sichuan, China

ABSTRACT

Objectives:

The aims of this study were to determine (1) the agreement in Epworth Sleepiness Scale (ESS) evaluated by patients and their close relatives (CRs), and (2) the correlation of objective sleepiness as measured by multiple sleep latency test (MSLT) with self-evaluated and close relative-evaluated ESS.

Methods:

A total of 85 consecutive patients with obstructive sleep apnea (OSA) (70 males, age 46.7 ± 12.9 years old) with an apnea-hypopnea index (AHI) > 5 events per hour (mean 38.9 ± 26.8/h) were recruited into this study. All participants underwent an overnight polysomnographic assessment (PSG), MSLT, and ESS rated by both patients and their CRs. Mean sleep latency < 8 min on MSLT was considered objective daytime sleepiness.

Results:

Self-evaluated global ESS score (ESSG) was closely correlated with evaluation by CRs (r = 0.79, p < 0.001); the mean ESSG score evaluated by patients did not significantly differ from that evaluated by CRs (p > 0.05). However, Bland-

Altman plot showed individual differences between self-evaluated and CR-evaluated ESS scores, with a 95%CI of -9.3 to 7.0. The mean sleep latency on MSLT was significantly associated with CR-evaluated ESSG (r = -0.23, p < 0.05); significance of association with self-evaluated ESSG was marginal (r = -0.21, p = 0.05).

Conclusions:

CR-evaluated ESS has a good correlation but also significant individual disagreement with self-evaluated ESS in Chinese patients with OSA. CR-evaluated ESS performs as well as, if not better than, self-evaluated ESS in this population when referring to MSLT.

Citation:

Li Y; Zhang J; Lei F; Liu H; Li Z; Tang X. Self-evaluated and close relative-evaluated Epworth Sleepiness Scale vs. multiple sleep latency test in patients with obstructive sleep apnea. J Clin Sleep Med 2014;10(2):171-176.


Excessive daytime sleepiness (EDS) is one of the most common symptoms in patients suffering from obstructive sleep apnea (OSA). Four percent of men and 2% of women have OSA with symptoms of EDS.1 EDS is known to be associated with reduced quality of life,25 impaired cognitive function,6 and increased risk of various accidents,7 as well as cardiovascular and cerebrovascular morbidities8 in patients with OSA.

As the degree of EDS is correlated with the severity of OSA, the determination of the severity of EDS in patients with OSA is important in the evaluation of disease progress and severity,9 prediction of outcomes,1014 and indication of treatment response. Among the various methods to define the degree of EDS, the Epworth Sleepiness Scale (ESS) and the multiple sleep latency test (MSLT) are the most commonly used tools for evaluating subjective and objective daytime sleepiness in patients with OSA.15,16 A number of clinical studies have found that ESS score is closely associated with MSLT in individuals with OSA1720; however, some studies report no significant correlation between ESS and MSLT scores in subjects with OSA.18,2022 This inconsistency might be due to the severity of OSA,21 age of participants,23 and/or the sample size of the study.17,21 These findings raise the concerns about the disagreement between objective and subjective sleepiness in patients with OSA.

BRIEF SUMMARY

Current Knowledge/Study Rationale: To determine the agreements in Epworth Sleepiness Scale (ESS) evaluated by patients and their close relatives (CR) and the correlations of objective sleepiness as measured by multiple sleep latency test (MSLT) with self-evaluated and CR-evaluated ESS.

Study Impact: CR-evaluated ESS performs as well as self-evaluated ESS in patients with obstructive sleep apnea (OSA) when referring to MSLT. CR-evaluated ESS scores maybe a supplement for assessing daytime sleepiness in patients with OSA.

The ESS is a tool measuring how likely a patient estimates he/she is to doze off or fall asleep in eight different daily situations.16 However, if the patient has little awareness of his/her own daytime sleepiness or believes that his/her sleep patterns are normal, the ESS may give unreliable results.23 Since a close relative (CR) of a patient can easily observe the patient's episodes of dozing off or falling asleep in these daily situations, CR might be a supplementary source in assessing patients' daytime sleepiness. However, previous studies have shown contradictory results in agreement between self-evaluated and CR-evaluated ESS scores. Two studies found consistency between self-evaluated and CR-evaluated scores,24,25 while six others19,23,2629 found some degree of disagreement despite moderate to high correlation. For example, CRs of patients with OSA tends to overestimate the severity of the patient's sleepiness, with a mean of difference in ESS ranging from 1.2 to 2.6 (p < 0.05).19,23,2629 Several factors including age, relationship between patient and CR, and education level of patient and CR, might account for the inconsistent results. To our knowledge, no previous study has tested the correlations of self-evaluated ESSG scores and CR-evaluated ESSG with MSLT scores. We thus conducted this study to determine the agreement of ESS score evaluated by patients and their CRs. In particular, the present study was to determine whether witnessed sleepiness based on ESSG evaluated by CR is closer than self-evaluated ESSG score to the mean sleep latency in patients with OSA.

METHODS

Setting

This was a cross-sectional observational study and was conducted in Sleep Medicine Center, West China Hospital of Sichuan University, China. Informed consent was obtained from all subjects and CRs. This study was approved by the Research Ethics Board of the West China Hospital of Sichuan University.

Subjects

Eighty-five (15 female and 70 male, mean age ± SD 46.60 ± 12.85 years) consecutive Chinese patients admitted form July 2012 to September 2012 were recruited into current study. Inclusion criteria were (1) newly diagnosed OSA, defined by apnea-hypopnea index (AHI) > 5 according to overnight polysomnography (PSG); (2) patient and CR between 18 to 65 years of age; (3) living at home with a CR (spouse, partner, or a family member), with CR spending > 10 h/day with the patient during the previous 3 months; (4) patient and CR fluent Mandarin speakers; (5) patient and CR free of acute medical problems and mental diseases as determined by a physician. Exclusion criteria for both patients and CRs were (1) the presence of a major psychiatric disorder based on physician's diagnosis; (2) hearing loss; (3) visual loss; (4) acute illness; (5) inability to speak or understand Mandarin fluently. The patients who met MSLT diagnostic criteria of narcolepsy (MSL ≤ 8 min, ≥ 2 SOREMPs) based on ICSD-2 were also excluded from this study.

After undergoing an overnight PSG recording, MSLT and ESS data were collected. All patients and CRs were asked for their highest educational level.

Overnight Polysomnography (PSG)

The diagnosis of OSA (AHI > 5) was established by a standard overnight PSG. Overnight PSG recording techniques and standard parameters were performed according to the American Academy of Sleep Medicine (AASM) Manual for the Scoring of Sleep and Associated Events.30 Sleep data were collected and scored via the Alice 5 Diagnostic Sleep System (Philips Respironics, Bend, OR, USA). The measures consisted of continuous recordings from 4 electroencephalographic (EEG) leads (F4-M1, C4-M1, O2-M1, F3-M2, C3-M2, O1-M2), bilateral electro-oculography (EOG), electrocardiography (ECG), electromyography (EMG) (submental, anterior tibialis), oral airflow, rib cage and abdominal movements, and measures obtained from an arterial oxygen saturation sensor on the left index finger. The hours during which the subjects slept in the sleep medicine center were based on their habitual sleep schedule at home. All sleep parameters recorded by PSG were analyzed by a senior technician according to the criteria of the AASM.30

Multiple Sleep Latency Test (MSLT)

According to the AASM Practice Parameters for Clinical Use of the MSLT,31 the MSLT was performed during daytime on the day following the overnight PSG recording and comprised four 20-min nap trials at intervals of 2 h. The first 30-s epoch of stage 1 or one epoch of any other sleep stage was required to establish sleep onset. When a subject did not fall asleep during the 20 min allowed, a score of 20 min was assigned to that nap. Mean sleep latency was calculated using MSLT data from all 4 tests. We used a score ≤ 8 in MSLT as the “gold standard” for the diagnosis of EDS.31

Epworth Sleepiness Scale (ESS)

Subjective daytime sleepiness was measured with the Chinese version of ESS.32 This is an 8-item self-administered questionnaire asking the subject how likely he/she is to doze off or fall asleep in different situations of daily life. This questionnaire was filled out by both patients and CRs between nap times in the morning during the MSLT. The patients and their CRs were asked to complete the questionnaire separately in order to prevent discussion between patients and CRs. Both subjects and CRs were allowed to interact with the technicians to clarify the doubts about the instructions and items on the ESS. The difference between self-reported and CR-evaluated ESSG scores was calculated as the pair of ESSG scores in self-evaluated minuend that in CR-evaluated.

Statistical Analysis

Data were analyzed using SPSS 17.0 software. As MSLT and ESS scores cannot be assumed to follow normal distributions, we performed nonparametric analyses using the Spearman rank correlation, χ2 test, or Fisher exact test, wherever appropriate. Agreement between patient and CR ESSG score was assessed graphically according to the Bland and Altman method. The Y-axis indicates the difference between self-evaluated and CR-evaluated ESSG, which is plotted against the mean of scores of each pair.33 Using scores < 8 on MSLT as the gold standard for diagnosis of EDS, ESSG > 10 as the test variable for assessing sensitivity, specificity, positive predictive value (PV+), and negative predictive value (PV-) in self-evaluated, CR-evaluated, both self- and CR-evaluated, and self- or CR-evaluated ESSG > 10, respectively. Data were considered significant when p-values were less than 0.05.

RESULTS

Eight-five patients' data were obtained (Table 1). The mean ± SD of AHI in all patients was 38.9 ± 26.8 events/hour. Among them, 21.2% had mild OSA, 29.4% had moderate OSA, and 49.4% had severe OSA. In this study, the educational level of most patients (81.2%) and CRs (92.1%) were above junior high school degree, so we believed that they could fully understand each item of the ESS.

General information and PSG parameters in patients.

jcsm.10.2.171.t01.jpg

table icon
Table 1

General information and PSG parameters in patients.

(more ...)

There was no difference between patient-evaluated ESSG and CR-evaluated ESSG (9.81 ± 6.12 vs. 10.99 ± 6.62, p > 0.05). As shown in Figure 1, the correlation between ESSG scores evaluated by patient and CR was close, with a Spearman coefficient of 0.79 (p < 0.001). Agreement between subjects' self-evaluated and CR-evaluated ESSG scores is shown in Figure 2 (Bland-Altman plot). As shown in Figure 2, CR-evaluated ESSG scores were larger than self- evaluated scores in 54.1% (n = 46), remained the same in 14.1% (n = 12), and were lower in 31.8% (n = 27) of cases. The mean of the difference between the ESSG scores evaluated by patients and CR was -1.18 ± 4.15 (p > 0.05), and the 95% upper and lower limits of the corresponding interval of variation were 7.0 and -9.3, as shown on Figure 2. This implies that on average, CR evaluation of ESSG was about 1 point higher than patient self-evaluation. Moreover, most of the differences were contained between 7.0 and -9.3. This demonstrates a large variability in the ESSG scores evaluated by patients and CRs. Visual inspection of the graph suggested no systematic variation in the disagreement as the mean values increased, and this trend was also confirmed by Spearman correlation analysis (r = -0.04, p = 0.75).

Correlation between patent's self-evaluated ESSG and CR-evaluated ESSG scores

jcsm.10.2.171a.jpg

jcsm.10.2.171a.jpg
Figure 1

Correlation between patent's self-evaluated ESSG and CR-evaluated ESSG scores

(more ...)

Bland-Altman plot: ESS scores of patients and their CRs estimating the patients' sleepiness.

The mean of the patients' and CRs' ESSG scores are plotted against the difference between the patients' and the CRs' ESSG scores. The lines are at the mean difference and 2 SD above and below the mean.

jcsm.10.2.171b.jpg

jcsm.10.2.171b.jpg
Figure 2

Bland-Altman plot: ESS scores of patients and their CRs estimating the patients' sleepiness.The mean of the patients' and CRs' ESSG scores are plotted against the difference between the patients' and the CRs' ESSG scores. The lines are at the mean difference and 2 SD above and below the mean.

(more ...)

Table 2 shows correlation of mean sleep latency with self-evaluated and CR-evaluated ESSG scores. Among the 85 patients and CRs in this study, higher scores on ESSG were associated with shorter mean sleep latency. We found there was a significant but weak correlation between CR-evaluated ESSG score and mean sleep latency of MSLT (r = -0.23, p < 0.05); significance of this association between self- evaluated ESSG score and mean sleep latency was marginal (r = -0.21, p = 0.05). The mean sleep latency of MSLT was found significantly correlated with item 1 (sitting and reading, r = -0.23, p < 0.05) and item 4 (as a passenger in a car for an hour without a break, r = -0.26, p < 0.05) in self-evaluated ESSG, and only item 3 (sitting, inactive in a public place, r = -0.27, p < 0.05) in CRs-evaluated ESSG. The results suggested that most of the items of ESS evaluated by patient and CR had weak correlation with mean sleep latency.

Correlations between mean sleep latency of MSLT, self-evaluated and CR-evaluated ESS score, and individual ESS sub items.

jcsm.10.2.171.t02.jpg

table icon
Table 2

Correlations between mean sleep latency of MSLT, self-evaluated and CR-evaluated ESS score, and individual ESS sub items.

(more ...)

Thirty-eight self-evaluated, 43 CR-evaluated, 33 both self- and CR- evaluated, and 43 self- or CR-evaluated ESSG scores were 11 (the lower limit of the abnormal range) or higher.16,25 Self- and CR-evaluated ESSG were both ≥ 11 and had a significant association with MSLT < 8 (p < 0.05); none of the others had a significant association (p > 0.05). The sensitivity and specificity of self-evaluated, CR-evaluated, both self-and CR-evaluated, and self- or CR-evaluated ESSG scores > 10 in detecting MSLT < 8 were low. We found that the specificity in self-evaluated ESSG (0.64) was higher than CR-evaluated ESSG (0.55), which suggested self-evaluated ESSG > 10 was better than that of CR-evaluated ESSG > 10 in preventing missed detection of mean sleep latency < 8. The optimal ESSG for predicting mean sleep latency < 8 was in both self- and CR-evaluated ESSG > 10 (sensitivity 0.49, specificity 0.71). All of the positive and negative predictive values were low (Table 3).

ESSG > 10 for “abnormal” ESS scores and resulting sensitivity, specificity, p value, positive predictive value (PV+), and negative predictive value (PV-) for mean sleep latency < 8 min.

jcsm.10.2.171.t03.jpg

table icon
Table 3

ESSG > 10 for “abnormal” ESS scores and resulting sensitivity, specificity, p value, positive predictive value (PV+), and negative predictive value (PV-) for mean sleep latency 10 for “abnormal” ESS scores and resulting sensitivity, specificity, p value, positive predictive value (PV+), and negative predictive value (PV-) for mean sleep latency

(more ...)

Table 4 shows the correlation of sleep parameters with ESSG scores evaluated by patients and CRs, and the mean sleep latency on MSLT. Self-evaluated ESSG score, CR-evaluated ESSG score, and mean sleep latency were found significantly correlated with total sleep time, the time of wake after sleep onset, sleep efficiency, and arousal index. Self-evaluated or CR-evaluated score was significantly correlated with AHI, longest apnea duration, oxygen desaturation index (ODI), and lowest SpO2; it was not significantly correlated with mean sleep latency. These findings suggested that nocturnal hypoxia had greater effect on subjective daytime sleepiness than objective daytime sleepiness. Moreover, only the CR-evaluated ESSG score was significantly correlated with REM latency and percentage of stage N3.

Spearman correlation coefficient (r) among MSL, ESSG evaluated by patients, and CRs, and PSG parameters measured.

jcsm.10.2.171.t04.jpg

table icon
Table 4

Spearman correlation coefficient (r) among MSL, ESSG evaluated by patients, and CRs, and PSG parameters measured.

(more ...)

DISCUSSION

In the current study, we found that (1) there was highly significant correlation but also significant individual disagreement in ESSG score between patient report and CR report; (2) ESSG was negatively correlated with mean sleep latency on MSLT. These results indicated that CR-evaluated ESS is as good as that evaluated by patients themselves. In this regard, CR might be a supplementary source for the assessment of subjective sleepiness in Chinese patients with OSA.

Since ESS relies on subject's retrospective subjective reports, the main disadvantage of ESS is its reliance on the subjective reports rather than objective measurements. Therefore, some factors might influence the ESS scores, such as patients' cognitive level,23 age,23 the time of evaluating ESS,29 and awareness of excessive daytime sleepiness. Due to the psychological factors and the lack of self-awareness of EDS, a self-evaluated ESS score below 10 might not necessarily indicate the absence of sleepiness. Thus, a supplementary informant might provide additional information on patient sleepiness. Previous studies have tried to analyze the difference between ESS evaluated by patients and their CRs; the result of our present study was in line with most previous studies.19,2329 The individual agreement between patients and CRs was not good (Figure 2). The individual disagreement was notable for about 40% of couples with three or more points of difference. Moreover, the CR frequently identified higher sleepiness levels than the patients themselves, with an average difference of 1.2 points on mean ESSG score (Figure 2). This result was in line with findings of previous studies in young and elder patients and their CRs.23,27,28 However, the difference between ESSG scores evaluated by patient and by CR varies significantly, which indicate the agreement in ESSG score is modest between these two informants.

To the best of our knowledge, there has been no previous study focusing on the different accuracy in association of OSA patient's objective EDS (mean sleep latency < 8 min on MSLT) with ESSG score evaluated by patient and CR. In previous studies, the findings in the association between ESSG score and mean MSLT score was contradictory, with a correlation coefficient ranging from -0.51 to -0.15.18,2022 One study has suggested that the correlation between ESSG score and mean sleep latency increased with increasing severity of OSA.21 In our study, regardless of whether the ESS was completed by the patient or CR, the correlation between ESSG score and mean sleep latency of MSLT was weak. The weak correlation between ESSG score and mean sleep latency may be due to the fact that we recruited a whole spectrum of patients with OSA but did not have enough sample size to stratify the sample.

The ESS was often the only tool to determine EDS in both population-based and hospital-based studies in patients with OSA.3437 In population-based studies, EDS determined by ESSG score was independently associated with AHI and snoring.34,35 Hospital-based studies have suggested that hypoxemia, autonomic arousal,36,37 and sleep fragmentation21 might be the main causes of EDS. A recent study among 787 consecutive Chinese patients with OSAS showed that ESS score was significantly associated with AHI, BMI, and oxygen desaturation index.36 In the current study, our results were in line with the previous findings (Table 4).21,3438 Thus, the degree of EDS judged by ESSG scores in patients with OSA might be an effective predictor for the severity level of OSA but not for their objective sleepiness.

Is it adequate to evaluate the ESS by CR? From our results in a sleep laboratory population, the answer is yes. First of all, it does not change the properties of the ESS. Further, the ESS evaluated by CR performed as well as that evaluated by patients themselves when referring to MSLT results. Since the ESS scores can be affected by a variety of factors, such as personality traits, cognitive function, time of day, medication effect, quality and duration of sleep in the preceding night, and the one's insight into the state of pathological sleepiness,16 we speculate the CR-evaluated approach could eliminate the above influencing factors. Therefore, our results suggested ESS scores evaluated by CRs were better—or at least no worse—than subject's self-evaluated ESS score in correlation with objective EDS.

There were several limitations in the present study. First, the relatively small sample in this study limited further stratification of the patients according to the severity of OSA. According to the previous studies,21 patients with more severe OSAS have much more agreement between ESSG scores and the mean sleep latency. Thus, we speculated that different severity level groups might also have different agreement between ESSG evaluated by patient and CR and meant sleep latency on MSLT. Further studies with larger samples size are warranted to address this issue. Second, all the patients in this study were selected from a sleep laboratory in a sleep medicine center; therefore, the results might not be generalized to the persons from community.

Clinical Implications

In this study, we found ESS scores evaluated by self and CR was significantly correlated. Thus, the ESS assessed by CRs can serve as supplementary information for OSA patient's sleepiness when the patient's information is not accessible (such as due to language, cognitive and somatic problems). Moreover, we found that both self-reported and CR-reported ESSG groups had higher specificity (p < 0.05) in detecting MSL < 8 min when the ESSG was > 10. In this regard, we speculate that the combination of both self-reported and CR-reported ESS might have better consistency with MSLT results.

DISCLOSURE STATEMENT

This was not an industry supported study. This work was supported by the National Natural Science Foundation of China (81170072 and 81328010) and by the Chinese German Joint Center for Sleep Medicine (GZ538). The authors have indicated no financial conflicts of interest.

REFERENCES

1 

Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S, authors. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med. 1993;328:1230–5. [PubMed]

2 

Douglas NJ, Polo O, authors. Pathogenesis of obstructive sleep apnoea/hypopnoea syndrome. Lancet. 1994;344:653–5. [PubMed]

3 

Sassani A, Findley LJ, Kryger M, Goldlust E, George C, Davidson TM, authors. Reducing motor-vehicle collisions, costs, and fatalities by treating obstructive sleep apnea syndrome. Sleep. 2004;27:453–8. [PubMed]

4 

Christou K, Markoulis N, Moulas AN, Pastaka C, Gourgoulianis KI, authors. Reactive oxygen metabolites (ROMs) as an index of oxidative stress in obstructive sleep apnea patients. Sleep Breath. 2003;7:105–10. [PubMed]

5 

Vennelle M, Engleman HM, Douglas NJ, authors. Sleepiness and sleep-related accidents in commercial bus drivers. Sleep Breath. 2010;14:39–42. [PubMed]

6 

Bucks RS, Olaithe M, Eastwood P, authors. Neurocognitive function in obstructive sleep apnoea: A meta-review. Respirology. 2013;18:61–70. [PubMed]

7 

Karimi M, Eder DN, Eskandari D, Zou D, Hedner JA, Grote L, authors. Impaired vigilance and increased accident rate in public transport operators is associated with sleep disorders. Accid Anal Prev. 2013;51:208–14. [PubMed]

8 

Feng J, He QY, Zhang XL, Chen BY, authors. Epworth Sleepiness Scale may be an indicator for blood pressure profile and prevalence of coronary artery disease and cerebrovascular disease in patients with obstructive sleep apnea. Sleep Breath. 2012;16:31–40. [PubMed]

9 

Jacobsen JH, Shi L, Mokhlesi B, authors. Factors associated with excessive daytime sleepiness in patients with severe obstructive sleep apnea. Sleep Breath. 2013;17:629–35. [PubMed]

10 

Kapur VK, Resnick HE, Gottlieb DJ, authors. Sleep disordered breathing and hypertension: does self-reported sleepiness modify the association? Sleep. 2008;31:1127–32. [PubMed]

11 

Feng J, He QY, Zhang XL, Chen BY, authors. Epworth Sleepiness Scale may be an indicator for blood pressure profile and prevalence of coronary artery disease and cerebrovascular disease in patients with obstructive sleep apnea. Sleep Breath. 2012;16:31–40. [PubMed]

12 

Barcelo A, Barbe F, de la Pena M, et al., authors. Insulin resistance and daytime sleepiness in patients with sleep apnoea. Thorax. 2008;63:946–50. [PubMed]

13 

Nena E, Steiropoulos P, Papanas N, et al., authors. Sleepiness as a marker of glucose deregulation in obstructive sleep apnea. Sleep Breath. 2012;16:181–6. [PubMed]

14 

Gooneratne NS, Richards KC, Joffe M, et al., authors. Sleep disordered breathing with excessive daytime sleepiness is a risk factor for mortality in older adults. Sleep. 2011;34:435–42. [PubMed]

15 

Thorpy MJ, author. The clinical use of the Multiple Sleep Latency Test. The Standards of Practice Committee of the American Sleep Disorders Association. Sleep. 1992;15:268–76. [PubMed]

16 

Johns MW, author. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;14:540–5. [PubMed]

17 

Chung KF, author. Use of the Epworth Sleepiness Scale in Chinese patients with obstructive sleep apnea and normal hospital employees. J Psychosom Res. 2000;49:367–72. [PubMed]

18 

Chervin RD, Aldrich MS, Pickett R, Guilleminault C, authors. Comparison of the results of the Epworth Sleepiness Scale and the Multiple Sleep Latency Test. J Psychosom Res. 1997;42:145–55. [PubMed]

19 

Olson LG, Cole MF, Ambrogetti A, authors. Correlations among Epworth Sleepiness Scale scores, multiple sleep latency tests and psychological symptoms. J Sleep Res. 1998;7:248–53. [PubMed]

20 

Sunwoo BY, Jackson N, Maislin G, Gurubhagavatula I, George CF, Pack AI, authors. Reliability of a single objective measure in assessing sleepiness. Sleep. 2012;35:149–58. [PubMed]

21 

Fong SY, Ho CK, Wing YK, authors. Comparing MSLT and ESS in the measurement of excessive daytime sleepiness in obstructive sleep apnoea syndrome. J Psychosom Res. 2005;58:55–60. [PubMed]

22 

Aurora RN, Caffo B, Crainiceanu C, Punjabi NM, authors. Correlating subjective and objective sleepiness: revisiting the association using survival analysis. Sleep. 2011;34:1707–14. [PubMed]

23 

Onen F, Moreau T, Gooneratne NS, Petit C, Falissard B, Onen SH, authors. Limits of the Epworth Sleepiness Scale in older adults. Sleep Breath. 2013;17:343–50. [PubMed]

24 

Kingshott RN, Sime PJ, Engleman HM, Douglas NJ, authors. Self assessment of daytime sleepiness: patient versus partner. Thorax. 1995;50:994–5. [PubMed]

25 

Johns MW, author. Sleepiness in different situations measured by the Epworth Sleepiness Scale. Sleep. 1994;17:703–10. [PubMed]

26 

Garcia LP, Capote GF, Quintana GM, Fuentes PM, Carmona BC, Sanchez AA, authors. Assessment with the Epworth scale of daytime somnolence in patients with suspected obstructive apnea syndrome during sleep. Differences between patients and their partners. Arch Bronconeumol. 2000;36:608–11. [PubMed]

27 

Walter TJ, Foldvary N, Mascha E, Dinner D, Golish J, authors. Comparison of Epworth Sleepiness Scale scores by patients with obstructive sleep apnea and their bed partners. Sleep Med. 2002;3:29–32. [PubMed]

28 

Kumru H, Santamaria J, Belcher R, authors. Variability in the Epworth sleepiness scale score between the patient and the partner. Sleep Med. 2004;5:369–71. [PubMed]

29 

Martinez D, Breitenbach TC, Lumertz MS, et al., authors. Repeating administration of Epworth Sleepiness Scale is clinically useful. Sleep Breath. 2011;15:763–73. [PubMed]

30 

Iber C, Ancoli-Israel S, Chesson A, Quan SF, authors. 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.

31 

Littner MR, Kushida C, Wise M, et al., authors. Practice parameters for clinical use of the multiple sleep latency test and the maintenance of wakefulness test. Sleep. 2005;28:113–21. [PubMed]

32 

Chen NH, Johns MW, Li HY, et al., authors. Validation of a Chinese version of the Epworth sleepiness scale. Qual Life Res. 2002;11:817–21. [PubMed]

33 

Bland JM, Altman DG, authors. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1:307–10. [PubMed]

34 

Gottlieb DJ, Whitney CW, Bonekat WH, et al., authors. Relation of sleepiness to respiratory disturbance index: the Sleep Heart Health Study. Am J Respir Crit Care Med. 1999;159:502–7. [PubMed]

35 

Gottlieb DJ, Yao Q, Redline S, Ali T, Mahowald MW, authors. Does snoring predict sleepiness independently of apnea and hypopnea frequency? Am J Respir Crit Care Med. 2000;162:1512–7. [PubMed]

36 

Chen R, Xiong KP, Lian YX, et al., authors. Daytime sleepiness and its determining factors in Chinese obstructive sleep apnea patients. Sleep Breath. 2011;15:129–35. [PubMed]

37 

Castiglioni P, Lombardi C, Di Rienzo M, et al., authors. What are the causes of excessive daytime sleepiness in patients with sleep-disordered breathing? Eur Respir J. 2008;32:526–7. [PubMed]

38 

Sun Y, Ning Y, Huang L, et al., authors. Polysomnographic characteristics of daytime sleepiness in obstructive sleep apnea syndrome. Sleep Breath. 2012;16:375–81. [PubMed]