The Effects of Irregular Bedtime on Sleep Quality, Daytime Sleepiness, and Fatigue in University Students in Taiwan

Background: Having an irregular bedtime (IB) is a prevalent problem in young adults and may be a potential detrimental factor affecting sleep quality. The goal of this study was to explore the effect of an IB on sleep quality, daytime sleepiness, and fatigue of undergraduate students in Taiwan. Methods: In total, 160 students completed the survey which included four parts: the Pittsburg Sleep Quality Index (PSQI), Epworth Sleepiness Scale (ESS), Fatigue Severity Scale (FSS), and rating of the frequency of an IB (FIB) via a semi-structural interview. An additional CHQ-12 (Chinese health questionnaire-12) was used to prevent the interference of recent possible psychological events in sleep assessment. Results: We found that high FIB was associated with a significant decrease in the average daily sleep time (ADST). The worst sleep quality with highest PSQI was noted in the group who had high FIB with ADST less than 7 hours. Differences in the ESS and FSS in among the FIB groups were insignificant. Conclusions: These results suggest that the students who have high FIB is associated with poor sleep quality especially in short sleepers. Background Some behaviors or activities such as irregular sleep schedule, frequent or prolonged daytime naps, excessive alcohol consumption near bedtime, and staying on the bed for non-sleep-related activities are known as “inadequate sleep hygiene” behaviors and are considered to be factors harmful to normal sleep [1-3]. Maintaining adequate sleep hygiene is considered to be an important adjuvant therapy in treating patients with insomnia or other sleep disturbances [1, 3, 4]. However, in normal subjects, evidence of these concepts is surprising inconsistent [5-8]. Human sleep is controlled by two main processes: the sleep homeostatic drive, which is determined by the length of sleep and wake time and the circadian system, which is an intrinsic pacemaker, and involves the pathway from the suprachiasmatic nucleus to the hypothalamus [9]. It is believed that regularizing the bedtime can strengthen the circadian rhythm and is beneficial to sleep quality [10]. It should be considered that the subjects who have high frequency of irregular bedtime (FIB) may also be prone to having a problem of chronic sleep insufficiency, and many studies have proven to impairments of cognition, vigilance, memory, and disturbing mood due to the accumulation of sleep deprivation [11-14]. Even a single shifted sleep schedule may cause a person to experience difficulties with sleep initiation and maintenance. A series of early studies conducted by Taub et al. showed that even young adults with the same amount of sleep but an acute shift of 2 hours schedule were found to have decreased cognitive and psychological functioning in the laboratory setting [15-18]. However whether this conclusion can be applied to real life over a long duration where the frequencies of shifted nights are often irregular is still unclear. Another condition related to sleep shifting in a period is work shifting. Studies have shown that shifting the sleep-awake schedule of shift workers also has adverse effects on sleep and general health such as a decrease in sleep quality, altered sympathetic activity, an increased cardiovascular event risk, and reduced cognitive performance [19-21]. However, the shifting sleep/awake pattern of shift workers greatly differs from the observed pattern of an irregular sleep schedule in young adults whose phase shifts are shorter but the pattern is more irregular. The suitability of applying physiological changes in shift workers to the population with an irregular sleep schedule is also doubtful. Although the exact prevalence is variable, it has been observed that many adolescents and young adults have irregular sleep schedules and a tendency to sleep late (phase delayed) [22-26]. A remarkable ratio of sleep problems and poor sleep quality in university students of many Western societies has been found by many studies. Ethnicity, social factors, and cultural effects can affect one’s sleep habits [24, 26]. To our knowledge, there has been no study exploring the epidemiological data on the sleep habits and sleep quality of university students in Taiwan, a modern Oriental society. The purpose of this study was to explore epidemiological data about the FIB and sleep quality in Chinese undergraduate students in the Eastern society, and also to analyze whether an IB can directly affect one’s daytime functioning and sleep quality. Methods Sampling and study population All participants were recruited from a medical university in Taipei, the capital of Taiwan. A semi-structured survey was given to all first-year medical students. In total, 197 students were interviewed, and 160 (81.2 %) students completed the survey. There were 81 males and 79 females. Their mean age was 20.3±1.9 years. The Taipei Medical University Hospital Review Board approved the study. Measures and data management The survey included three questionnaires which evaluated the subjects’ sleep quality, daytime sleepiness, fatigue, and the FIB. The Pittsburg Sleep Quality Index (PSQI) which has been accepted as being a valid and reliable survey was applied to assess sleep quality [27, 28]. The PSQI includes 19 items, and the score ranges from 0 to 21 (good quality to poor quality). Sleep onset latency (SOL) and sleep efficiency (SE), defined as actual sleep time divided by the time in bed can also be obtained with the PSQI. To test for daytime sleepiness, we used the Epworth Sleepiness Scale (ESS), a widely used and reliable predictor of daytime sleepiness [29]. It is based on a four-point scale to rank a subject’s chances of falling asleep in different scenarios. The other self-reporting scale used to evaluate fatigue was the Fatigue Severity Scale (FSS), which includes nine questions with a total score of from 9 to 63 (no fatigue to severe fatigue) and which has successfully been applied to the clinical evaluation of fatigue [30, 31]. Frequency of variable bedtime (FIB) which was defined as the frequency of shifting their bedtime more than 1 hour from their usual bedtime in the past 2 weeks into three categories (‘low 1 night per week’, ‘intermediate 1 to 3 nights per week’; ‘high 3 nights per week or a regular bedtime cannot defined by subjects’). The ADST in the most recent 2 weeks was estimated by subjects. An additional CHQ-12 (Chinese health questionnaire-12) was used to prevent the interference of recent possible psychological events in sleep assessment. CHQ-12 included 12 items which is developed to screen the possibility of mild psychological disorders or distress and has been shown as a validate tool in Chinese population [32]. Eleven students were excluded with the CHQ-12 was more than 2 points. Statistical analysis After data collection, five main outcome scores served as the dependent variables: SOL, SE, PSQI, ESS, and FSS. Student’s t-test was applied to compare these variables between the two genders. In order to exclude the potential effects of sleep insufficiency, we grouped the participants into three subgroups: an ADST less than 7 hours, between 7 and 8 hours, and more than 8 hours. Chi-square test and Spearman’s correlation were applied to analyze the distribution between the two variables: FIB and ADST. Two-way ANOVA was applied to compare the differences in SOL, SE, FSS, ESS, and PSQI in each FIB and ADST group. LSD analysis for the post hoc analysis was applied. The significance level was set as p < 0.05. Results There were 81 (51%) females and 79 males (49%) in the sample. There was no statistical difference in any variable between the two genders, so the data for males and females were combined for all subsequent analysis. The SOL of all subjects was 14.2±10.6 mins, and the ADST was 6.7±1.3 hours. The FSS, ESS, and PSQI were 38.2±8.9, 6.3±3.3, and 4.9±2.4 respectively. Data on variables for all subjects are summarized in Table 1. Regarding the FIB, 26.9% of students had a low FIB (< 1 night per week), 38.8% were in the intermediated group (1 ~ 3 nights per week), and 34.4% were in the high-FIB group (> 3 nights per week). Nearly half (46.9%) of the students had an ADST less than 7 hours in the survey. It was worth noting that the ADST was statistically negatively correlated with the degree of FIB (r = -0.22, p < 0.05). The distribution of ADST and FIB values of all subjects are listed in Table 2. Results of the two-way ANOVA analysis of differences in the five dependent variables in two variables (FIB and ADST) are showed in Table 3. There was statistical significance noted in the difference in PSQI (r 2 = 0.193, adjusted r 2 = 0.150, p < 0.01), and SE (r 2 = 0.329, adjusted r 2 = 0.294, p < 0.01) in the overall effect. The interactive effect of FIB and ADST on the five variables did not reach statistical significance. However, a decreased in the ADST was associated with a high PSQI (p < 0.01) and a low SE (p < 0.01) but not significantly with ESS or FSS. When considering the main effects of the FIB, there was a statistical difference in PSQI (p = 0.047); particularly when a subject’s ADST less than 7 hours by post hoc analysis (Fig. 1). Differences in the other dependent variables did not reach a significant level. Discussions and conclusions An IB is considered to be one kind of inadequate behavior for maintaining good sleep quality. Our study supports an IB possibly being an independent factor affecting sleep quality. The negative effect of IB in sleep quality is more severe in subjects whose ADST was less than 7 hours. We postulate that the relationship between sleep quality and FIB is due to the irregularity of the participants’ sleep shifts which can cause disturbances in the circadian cycle. Circadian system disturbances, as described by previous studies are characterized by disturbances in sleep architecture, and sleep quality and inability to fall asleep or stay awake. These disturbances are known to further cause fatigue, vigilance problems, and a decrease in productivity and to have negative health side effects [10, 33]. In addition to the circadian disturbance, particularly when a subject tries to get more sleep after shifting the phase later, environmental factors such as light or noise in the morning can affect the overall sleep quality. Interestingly, we found subjects whose ADST is more than 8 hours, the adverse effects of an IB on sleep quality become insignificant. We believe that the effects of an IB may become subtle if a subject can get sufficient sleep. We also found some adverse trends with a high FIB regarding FSS and ESS, although the differences did not reach statistical significance. It is worth noting that students who had a high FIB were significantly associated with a low ADST which puts them at risk for potential sleep insufficiency. This finding can be explained by daytime activities (class or recreation) which restricted their sleep time in the students whose bedtimes were irregular since they were unable to get enough sleep in the morning. According to our data, prevalence of an IB and short ADST are high in undergraduate students in Taiwan. These findings are very similar to previous studies done in other countries and reflect sleep problems in university students may be a universal and prevalent problem in modern society [22, 23, 25]. One study reviewed the published papers regarding university students’ sleep patterns, and found the median sleep time had dropped about one hour in the past 20 years [34]. The lifestyle, social and academic schedules, insufficient sleep education could be the etiology of chronic sleep insufficiency and poor sleep in university students [22, 23, 35]. This study provides evidence that the IB of young adults may have a direct adverse effect on sleep quality. Social cues, subjects’ willingness, activities, and environmental factors all play important roles in setting human circadian rhythms and these factors can entrain the sleep-awake cycle as a way to fine-tune to intrinsic human circadian rhythm [10]. Causes of the development of irregular sleep habits in young adults are still not well-known, and are likely multi-factorial, including biological, behavioral, and social factors. Two different types of people were observed in expression of alertness and sleepiness in the endogenous circadian cycle in the general population: a morning type (M-type) and an evening type (E-type) [36]. One study showed that individual variations in behavior especially in sleep/wake schedules differed between the two types. The E-type tends to vary their bedtime and waking time [37]. Taillard et al. showed that the E-type was associated with an increase of sleep need and irregularity of sleeping habits but not subjective daytime sleepiness [38]. A delayed sleep phase disorder which leads to a sleep phase delay and often-variable bedtimes is another prevalent circadian sleep disorder in adolescents and young adults, and a biological mechanism such as an altered melatonin circadian response was found in some studies [39, 40]. However, no studies have explored the biological mechanism or model of developing a variable sleep schedule. Further studies need to explore the circadian markers and physiological features of subjects who have an IB in order to clarify the underlying mechanism and model. Our study showed an IB had adverse effects on sleep quality. Although the consequence of an IB may be less than sleep insufficiency, the long-term effect should not be overlooked. There is a need to address the importance of a regular sleep schedule which should be maintained. Students who have an IB should be advised to make their bedtime regular and increase their ADST which might help increase their sleep quality and daytime functioning. A structured approach to enhancing the sleep hygiene and knowledge of university students was suggested but the effects still need to be confirmed [35]. However, one limitation of this study is related to the estimation of the ADST and FIB from the self-reported questionnaires which may have a recall bias. An objective assessment such as actigraphy of these parameters would be required for more-accurate estimations of the ADST and FIB. Moreover, the undergraduate subjects we studied at the medical university might not be a representative sample of the general population of university students. A random population-based sample with a large sample size would be required to confirm the effects of an IB observed in our study. Also, the real causality is still difficult to determine in this cross-sectional study. The possibility that subjects who have poor sleep quality with sleep onset difficulty or sleep fragmentation will develop some inadequate sleep habits such as irregular sleep schedules, etc cannot be ruled out. Longitudinal studies would be required to clarify the direction of causality between the IB and poor sleep quality. Finally, it should be noted that there are many different types of inadequate sleep habits and sleep-related behaviors; only the FIB was taken into account in our study. Other poor sleep habits would likely also have harmful impacts in sleep quality in normal subjects, and these factors should be considered in further studies. Reference 1. Hauri PJ: Advances in the behavioral treatment of insomnia. Sleep MedRev 2003, 7:201-202. 2. Jefferson CD, Drake CL, Scofield HM, Myers E, McClure T, Roehrs T, Roth T: Sleep hygiene practices in a population-based sample of insomniacs. Sleep 2005, 28:611-615. 3. Stepanski EJ, Wyatt JK: Use of sleep hygiene in the treatment of insomnia. Sleep MedRev 2003, 7:215-225. 4. Guedes BM, Soares V, Carlos WJ: [Impact of sleep hygiene on patients with obstructive sleep apnoea syndrome]. RevPortPneumol 2006, 12:147-176. 5. Cheek RE, Shaver JL, Lentz MJ: Lifestyle practices and nocturnal sleep in midlife women with and without insomnia. BiolResNurs 2004, 6:46-58. 6. Cheek RE, Shaver JL, Lentz MJ: Variations in sleep hygiene practices of women with and without insomnia. ResNursHealth 2004, 27:225-236. 7. LeBourgeois MK, Giannotti F, Cortesi F, Wolfson A, Harsh J: Sleep hygiene and sleep quality in Italian and American adolescents. AnnNYAcadSci 2004, 1021:352-354. 8. Manni R, Politini L, Ratti MT, Marchioni E, Sartori I, Galimberti CA, Tartara A: Sleep hygiene in adult epilepsy patients: a questionnaire-based survey. Acta NeurolScand 2000, 101:301-304. 9. Dijk DJ, Lockley SW: Integration of human sleep-wake regulation and circadian rhythmicity. JApplPhysiol 2002, 92:852-862. 10. Borbely AA, Achermann P: Sleep homeostasis and models of sleep regulation. JBiolRhythms 1999, 14:557-568. 11. Carskadon MA: Sleep deprivation: health consequences and societal impact. MedClinNorth Am 2004, 88:767-776. 12. Dement WC: Sleep extension: getting as much extra sleep as possible. ClinSports Med 2005, 24:251-268, viii. 13. Dinges DF, Pack F, Williams K, Gillen KA, Powell JW, Ott GE, Aptowicz C, Pack AI: Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4-5 hours per night. Sleep 1997, 20:267-277. 14. Horne JA: Sleep function, with particular reference to sleep deprivation. AnnClinRes 1985, 17:199-208. 15. Taub JM: Effects of ad lib extended-delayed sleep on sensorimotor performance, memory and sleepiness in the young adult. Physiol Behav 1980, 25:77-87. 16. Taub JM, Berger RJ: Performance and mood following variations in the length and timing of sleep. Psychophysiology 1973, 10:559-570. 17. Taub JM, Berger RJ: Acute shifts in the sleep-wakefulness cycle: effects on performance and mood. PsychosomMed 1974, 36:164-173. 18. Taub JM, Berger RJ: The effects of changing the phase and duration of sleep. JExpPsycholHumPerceptPerform 1976, 2:30-41. 19. Kecklund G, Ekstedt M, Akerstedt T, Dahlgren A, Samuelson B: The effects of double-shifts (15.5 hours) on sleep, fatigue and health. JHumErgol(Tokyo) 2001, 30:53-58. 20. Phillips B, Magan L, Gerhardstein C, Cecil B: Shift work, sleep quality, and worker health: a study of police officers. SouthMedJ 1991, 84:1176-1184, 1196. 21. Shen J, Botly LC, Chung SA, Gibbs AL, Sabanadzovic S, Shapiro CM: Fatigue and shift work. JSleep Res 2006, 15:1-5. 22. Brown FC, Buboltz WC, Jr., Soper B: Relationship of sleep hygiene awareness, sleep hygiene practices, and sleep quality in university students. BehavMed 2002, 28:33-38. 23. Buboltz WC, Jr., Brown F, Soper B: Sleep habits and patterns of college students: a preliminary study. JAmCollHealth 2001, 50:131-135. 24. Hicks RA, Lucero-Gorman K, Bautista J, Hicks GJ: Ethnicity, Sleep Hygiene Knowledge, and Sleep Hygiene Practices. PerceptMotSkills 1999, 88:1095-1096. 25. Lack LC: Delayed sleep and sleep loss in university students. JAmCollHealth 1986, 35:105-110. 26. Manni R, Ratti MT, Marchioni E, Castelnovo G, Murelli R, Sartori I, Galimberti CA, Tartara A: Poor sleep in adolescents: a study of 869 17-year-old Italian secondary school students. JSleep Res 1997, 6:44-49. 27. Buysse DJ, Reynolds CF, III, Monk TH, Berman SR, Kupfer DJ: The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1989, 28:193-213. 28. Smyth C: The Pittsburgh Sleep Quality Index (PSQI). Insight 2000, 25:97-98. 29. Johns MW: A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep 1991, 14:540-545. 30. Laberge L, Gagnon C, Jean S, Mathieu J: Fatigue and daytime sleepiness rating scales in myotonic dystrophy: a study of reliability. JNeurolNeurosurgPsychiatry 2005, 76:1403-1405. 31. Taylor RR, Jason LA, Torres A: Fatigue rating scales: an empirical comparison. PsycholMed 2000, 30:849-856. 32. Chong MY, Wilkinson G: Validation of 30and 12-item versions of the Chinese Health Questionnaire (CHQ) in patients admitted for general health screening. PsycholMed 1989, 19:495-505. 33. Van RO, Mennuni G: [Fatigue and sleep: the point of view of the chronobiologist]. RevMedBrux 2002, 23:A288-A293. 34. Hicks RA, Pellegrini RJ: The changing sleep habits of college students. PerceptMotSkills 1991, 72:1106. 35. Brown FC, Buboltz WC, Jr., Soper B: Development and evaluation of the Sleep Treatment and Education Program for Students (STEPS). JAmCollHealth 2006, 54:231-237. 36. Lavie P, Segal S: Twenty-four-hour structure of sleepiness in morning and evening persons investigated by ultrashort sleep-wake cycle. Sleep 1989, 12:522-528. 37. Carrier J, Monk TH, Buysse DJ, Kupfer DJ: Sleep and morningness-eveningness in the 'middle' years of life (20-59 y). JSleep Res 1997, 6:230-237. 38. Taillard J, Philip P, Bioulac B: Morningness/eveningness and the need for sleep. JSleep Res 1999, 8:291-295. 39. Arendt J, Skene DJ: Melatonin as a chronobiotic. Sleep MedRev 2005, 9:25-39. 40. Wyatt JK: Delayed sleep phase syndrome: pathophysiology and treatment options. Sleep 2004, 27:1195-1203. Tables Table 1 All measured variables in both genders Variable* Female (n = 81) Male (n = 79) Total (n = 160) SOL (mins) 15.2 (11.3) 12.9 (9.8) 14.2 (10.6) SE (%) 89.0 (14.4) 91.1 (10.5) 90.1 (12.6) ADST (hours) 6.8 (1.4) 6.6 (1.1) 6.7 (1.3) ESS 6.7 (3.2) 5.9 (3.4) 6.3 (3.3) FSS 38.0 (8.5) 38.5 (9.4) 38.2 (8.9) PSQI 5.0 (2.4) 4.8 (2.3) 4.9 (2.4) SOL, sleep onset latency; SE, sleep efficiency; ADST, average daily sleep time; ESS, Epworth Sleepiness Scale; FSS, fatigue severity scale; PSQI, Pittsburg Sleep Quality Index. All variables were expressed as mean (standard deviation). * No variables exhibited a statistical difference between the two genders at the level of p<0.05. Table 2 FIB and ADST in all subjects FIB* + ADST* Low intermediate high Total < 7 h/day 14 (8.8) 26 (16.3) 35 (21.9) 75 (46.9) 7 ~ 8 h/day 17 (10.6) 24 (15.0) 10 (6.3) 51 (31.9) 8 h/day 12 (7.5) 12 (7.5) 10 (6.3) 34 (21.3) Total 43 (26.9) 62 (38.8) 55 (34.4) 160 (100) Values are given as the number (%). FIB, frequency of an irregular bedtime; ADST, average daily sleep time. + The low-FIB group was < 1 night/week, while intermediate group was 1 to 3 nights /week, and high group was more than three nights/week. An irregular bedtime was defined as a more than 1-hour variation from the usual bedtime. All variables were expressed as mean (standard deviation). * Chi-Square analysis of the distribution between two variables had statistical significance (χ = 11.68, p = 0.022), and the Spearman correlation was -0.22 (p =