https://orcid.org/0000-0003-3610-9940
ABSTRACT
We hypothesized that social jetlag would be associated with prolonged sleep duration on weekends and irregularities in wake-up time and/or bedtime on weekdays. In total, 1,200 Japanese male college students were included in this study. Participants completed an eight-day sleep diary in which they recorded their wake-up time, bedtime, and sleep duration every day for a week. Mean wake-up time, bedtime, and sleep duration, standard deviations in wake-up time and bedtime, the coefficient of variation in sleep duration, social jetlag, and chronotype were calculated over seven nights. Multiple regression was used to explore the factors influencing social jetlag. Stepwise selection analysis was performed to analyze the parameters identified on multiple regression analysis. The mean chronotype of the included participants was 5.3 ± 1.5; their mean social jetlag value was 1.1 ± 1.0. The mean wake-up time, bedtime, and sleep duration values were 8.5 ± 1.1, 25.0 ± 1.1, and 7.5 ± 1.1 h, respectively. Multiple regression analysis showed five indicators exerting a statistically significant influence on social jetlag. The standard deviation in wake-up time was adopted as the first factor in the stepwise selection analysis. These results show that social jetlag is associated with not only prolonged sleep duration on weekends but also irregularities in wake-up time during the week.
Introduction
Lack of sleep is a global public health concern. Insufficient sleep duration is associated with the onset of lifestyle-related diseases, as this adversely affects hormone secretion and autonomic nervous system functions in the body. Moreover, insufficient sleep leads to deterioration of mental functions, which manifests as drowsiness, lack of motivation, and memory loss during the day (Spiegel et al. Citation1999; Suka et al. Citation2003). Many people try to compensate for insufficient sleep on workdays by prolonging their sleep duration on weekends. However, prolonging sleep duration on weekends does not improve the decline in metabolic function caused by lack of sleep (Depner et al. Citation2019). In addition, previous reports have indicated that prolonging sleep on weekends decreases insulin sensitivity, leads to increased food intake at dinner (Depner et al. Citation2019), and reduces the Sleep Regularity Index (Phillips et al. Citation2017). Moreover, irregularities in sleep habits are observed with social jetlag and late chronotypes (Yamanaka Citation2016).
Social jetlag indicates a discrepancy between social time and an individual’s biological clock and is associated with mental and physical disorders (Wittmann et al. Citation2006). Social jetlag is assessed by calculating the difference between the midpoints of sleep duration on workdays and weekends (Wittmann et al. Citation2006). Social jetlag has been linked to unhealthy behaviors, such as smoking (Wittmann et al. Citation2006) or physical inactivity (Rutters et al. Citation2014) and has also been suggested to be a risk factor for obesity (Roenneberg et al. Citation2012), metabolic dysfunction that may lead to a predisposition to diabetes, atherosclerotic cardiovascular disease (Islam et al. Citation2018; Parsons et al. Citation2015; Wong et al. Citation2015), and depression (Levandovski et al. Citation2011). In Japan, the average absolute social jetlag value has been reported as 0.91 ± 0.89 h (Komada et al. Citation2019). In addition, the absolute social jetlag experienced by Japanese individuals in their 20s is higher than that experienced by other age groups (Komada et al. Citation2019). Young people often have a late chronotype and significant social jetlag because they tend to prolong sleep duration on weekends to compensate for the sleep debt accumulated on workdays.
It is presumed that college students have different wake-up times and bedtimes between school days due to differences in the start time of each school day. These differences in wake-up times and bedtimes on school days may be related to disturbances in the circadian rhythm as well as social jetlag. Therefore, it is presumed that the social jetlag experienced by college students is related not only to the prolongation of sleep time on non-school days but also to irregularities in wake-up times and bedtimes on school days. However, previous indicators for social jetlag does not reflect irregularities in wake-up time and bedtime on school days (Fischer et al. Citation2021).
In a previous study, we proposed that standard deviations in wake-up time and bedtime and the coefficient of variation in sleep duration are parameters reflecting sleep irregularity, and we reported that social jetlag significantly and positively correlated with these parameters (Nishimura et al. Citation2022). Additionally, social jetlag is presumed to be associated with wake-up time, bedtime, and sleep duration differences between weekdays and weekends. However, the parameters that exert the greatest effects on social jetlag are presently unclear. The aim of this study was to verify our hypothesis that social jetlag is associated with prolonged sleep duration on weekends and irregularities in wake-up times and/or bedtimes using multiple regression analysis.
Materials and methods
Participants
A total of 1,200 Japanese male college students (age range, 19–22 years) were included in this study. The participants completed an eight-day sleep diary, which included records of their wake-up times, bedtimes, and sleep duration (Sasawaki et al. Citation2022; Short et al. Citation2013). The inclusion criteria were as follows: participants who completed the Japanese version of the Munich ChronoType Questionnaire (MCTQ) (Kitamura et al. Citation2014), and participants who provided data spanning five school days and two non-school days in a sleep diary. We excluded participants who provided invalid answers or reported on the use of alarm clocks on non-school days (Kitamura et al. Citation2014). The final sample included 1,092 participants. The timing of classes attended by the participants at the college was 8:50 a.m. (first period), 10:30 a.m. (second period), 1:15 p.m. (third period), 3:10 p.m. (fourth period) and 5:05 p.m. (fifth period). The start or end times of classes, the number of classes, etc., differed depending on the participants.
All procedures were reviewed and approved by the Ethics Committee of the Hiroshima Institute of Technology (approval no. 15–003). The study protocol conformed to the tenets of the Declaration of Helsinki. All participants were informed regarding the benefits and risks of the present study and provided their written informed consent prior to study participation.
Sleep diary
Mean wake-up time, bedtime, and sleep duration, standard deviations in wake-up time and bedtime, and the coefficient of variation in sleep duration (i.e., parameters of sleep variability), as well as social jetlag and chronotype were calculated. This assessment included five school days and two non-school days.
The relative social jetlag was calculated as follows:
relative social jetlag (h) = average midpoints of sleep duration on non-school days − average midpoints of sleep duration on school days (Sasawaki et al. Citation2022).
School days were defined as the nights before a school day (Sunday night to Thursday night), whereas non-school days were defined as the nights before non-school days (Friday and Saturday nights).
Additionally, in this study, the absolute social jet lag was examined. Chronotype, defined as an individual’s preference for sleep at a certain time, was assessed using the average midpoints of sleep duration on non-school days (MSF). If sleep duration was longer on non-school days than on school days, sleep-corrected MSF was used as an indicator of the chronotype (MSFsc) (Roenneberg et al. Citation2019). Furthermore, as an index of daily fluctuations in wake-up and bedtime, the standard deviation of wake-up and bedtime for two consecutive days was calculated.
Statistical analysis
The sleep parameters and chronotypes are expressed as means ± standard deviations. All calculations were performed using the Statistical Package for Social Science for Windows (version 25; IBM, Armonk, NY, USA). In the normality analysis, the normality of the data was confirmed. Pearson product-moment correlation coefficients were used to determine the relationships between the evaluated sleep parameters. Multiple regression analysis (forward selection method) was used to explore the factors influencing social jetlag. First, the Pearson correlation coefficients were calculated for each sleep parameter. If the coefficients of the correlations were ≥0.7, the parameter was removed as an independent variable. Second, the multicollinearity was evaluated by the variance inflation factors (VIF). If the VIF was <5, there was no issue of multicollinearity. Finally, stepwise selection analysis was performed to analyze the parameters influencing social jetlag, considering the parameters identified on multiple regression analysis. The data of standard deviation of wake-up and bedtime by day of the week were not confirmed. Therefore, the data are expressed as median (interquartile range). The Friedman’s test was used to compare the standard deviation of wake-up and bedtime by day of the week. The Bonferroni test was used for post hoc analysis. Statistical significance was set to a two-sided P-value of <.05.
Results
All study participants were male, and the age range was 19–22 years. shows the sleep parameters and chronotypes of the included participants. The mean MSFsc, a measure of chronotype, was 5.3 (05:18) ± 1.5 h. The difference between the midpoints of sleep duration on school days and on non-school days, a measure of relative social jetlag, was 0.9 ± 1.2 h. Further, the absolute social jetlag value was 1.1 ± 1.0 h. The percentages of participants with absolute social jetlag values of more than 1 h and 2 h were 43.2% and 14.2%, respectively. The mean wake-up time was 8.5 (08:30) ± 1.1 h, the mean bedtime was 25.0 (01:00) ± 1.1 h, and the mean sleep duration was 7.5 ± 1.1 h.
Table 1. Sleep parameters and chronotypes.
shows the Pearson correlation coefficients for sleep parameters. The coefficients of the correlations between chronotype, wake-up time, bedtime, and the difference in wake-up time on school days and non-school days were all ≥0.7. Therefore, chronotype was removed as an independent variable.
Table 2. Pearson correlation coefficients for sleep parameters.
shows the results of the multiple regression analysis. Wake-up time, bedtime, sleep duration, and the difference in wake-up times on school days and non-school days were excluded because of a lack of statistical significance. Since all the VIFs were <3 (range: 1.473–2.324), there was no issue of multicollinearity. Multiple regression analysis demonstrated that the standard deviation in wake-up time, difference in bedtime on school days and non-school days, difference in sleep duration on school days and non-school days, standard deviation in bedtime, and coefficient of variation in sleep duration showed statistically significant effects on social jetlag (F = 677.082, adjusted R-squared = 0.756, P < .001). The standardized partial regression coefficient of the standard deviation in wake-up time was the highest (0.487) of all the evaluated sleep parameters, followed by those for the difference in bedtime on school days and non-school days (0.469), the difference in sleep duration on school days and non-school days (0.265), the standard deviation in bedtime (0.209), and the coefficient of variation in sleep duration (−0.168).
Table 3. Multivariate regression analysis findings.
shows the results of the stepwise selection analysis. The standard deviation in wake-up time was adopted as the first factor in the stepwise selection analysis (F = 1119.006, adjusted R-squared = 0.522, P < .001). Thereafter, the standard deviation in wake-up time and the difference in bedtime on school days and non-school days were adopted in stepwise selection analysis (F = 1230.665, adjusted R-squared = 0.693, P < .001).
Table 4. Results of the stepwise selection analysis.
shows the standard deviation of wake-up and bedtime compared by day of the week. The standard deviation of wake-up time and bedtime were significantly different during the week (P < .05). The standard deviation of wake-up time in Mon-Tue was significantly lower than that of Thu-Fri and Sat-Sun, respectively (P < .05). Further, the standard deviation of wake-up time in Tue-Wed was significantly lower than that of Wed-Thu, Thu-Fri, Fri-Sat, and Sun-Mon (P < .05, respectively). The standard deviation of bedtime in Mon-Tue was significantly lower than that of Thu-Fri (P < .05). Moreover, the standard deviation of bedtime in Tue-Wed was significantly lower than Wed-Thu (P < .05) and the standard deviation of Wed-Thu was significantly higher that of Thu-Fri and Fri-Sat (P < .05, respectively).
Table 5. Standard deviation of wake-up and bedtime compared by day of the week.
Discussion
In this study, we aimed to verify our hypothesis that social jetlag is related not only to prolonged sleep duration on weekends but also to irregularities in wake-up time and bedtime on weekdays. Our results demonstrated that, among all evaluated sleep parameters, the standard deviation in wake-up time showed the highest standardized partial regression coefficient, followed by the difference in bedtime on school days and non-school days and the difference in sleep duration on school days and non-school days. The standard deviation in wake-up time was adopted as the first factor in the stepwise selection analysis. These results support our study hypothesis that social jetlag is associated with prolonged sleep duration on weekends and irregularities in wake-up times and/or bedtimes. The results also clarified that social jetlag is affected more by irregularities in wake-up times than by the prolongation of sleep time on non-school days.
The mean MSFsc (a measure of chronotype) was 5.3 (05:18 a.m.), which is 96 min longer than that in a previous study evaluating the average chronotype of MCTQ respondents from Japan (mean age, 45.1 ± 13.4 years) (Komada et al. Citation2019). Moreover, a previous study indicated that the MSFsc progressively shifts to later times throughout puberty and adolescence until it reaches a peak at around the age of 20 years old (Foster and Roenneberg Citation2008). The participants in the present study were young adults with late chronotypes. However, their social jetlag was 1.1 h, which may be only slightly different from that reported in a previous study of somewhat older individuals with a mean age of 45.1 ± 13.4 years (0.9 h) (Komada et al. Citation2019). The percentage of participants with the absolute social jetlag value of more than 1 h was 43.2%, which is 15.9% lower than that reported in a previous study that evaluated absolute social jetlag in respondents from Japan who were 20 years old (Komada et al. Citation2019). Considering that the participants in this study were male college students, it is possible that there were days when the wake-up and bedtime were delayed due to the different start times of classes depending on the day of the week. Alternatively, it might be attributed to some participants who woke up and went to bed early due to club activities on non-school days.
Social jetlag is assessed by calculating the difference between the midpoints of sleep duration on weekdays and weekends (Wittmann et al. Citation2006). It has been reported that the later the individual’s chronotype, the more considerable is the social jetlag (Komada et al. Citation2019; Nishimura et al. Citation2022). Our results showed that, among all evaluated sleep parameters, standard deviation in wake-up time had the highest standardized partial regression coefficient, followed by the difference in bedtime on school days and non-school days and the difference in sleep duration on school days and non-school days. The adjusted R-squared was 0.756. Hence, a highly accurate multiple regression equation was obtained. This indicates that irregularities in wake-up time are related to social jetlag and late chronotype. Moreover, these data suggest that, for the prevention and improvement of social jetlag, the focus should be not only on the modification of prolonged sleep time on non-school days but also on establishing a more consistent wake-up time and bedtime on school days.
As an index of daily fluctuations in wake-up and bedtime, the standard deviation of wake-up and bedtime for two consecutive days was significantly changed during the week. As the school day progressed, the variation in wake-up and bedtime increased. This data suggested that social jetlag may be a factor for irregularities in wake-up and bedtime on Thursdays and Fridays due to accumulated fatigue over the school day.
In the present study, we noted that the standardized partial regression coefficient for the standard deviation in wake-up time was higher than that for the difference in bedtime on school days and non-school days. In addition, the standard deviation in wake-up time was adopted as the first factor in the stepwise selection analysis. These results suggest that social jetlag is affected more by irregularities in wake-up time on weekdays than by the prolongation of sleep duration on non-school days. Therefore, standardizing one’s wake-up time and/or bedtime on school and non-school days may be a more important factor in the prevention and improvement of social jetlag than prolongation of sleep time on non-school days. In the future, interventional studies evaluating standardized wake-up times and/or bedtimes may contribute to health education efforts targeting the prevention and improvement of social jetlag.
This study has some limitations. First, records were logged in a sleep diary for eight days only. Our results may have been affected by the variability inherent to the short duration of the recording period. However, as in a previous study (Sasawaki et al. Citation2022; Short et al. Citation2013), the sleep diary in the present study included five school nights and two non-school nights. Considering the different start times of classes depending on the day of the week, it is possible that there were days when the wake-up and bedtime were delayed. These compositions reflect a typical week for college students. Second, all of the included participants were Japanese male college students. Therefore, our conclusions may not be generalizable to the general Japanese population or to a more international population. Consideration of age and sex may be necessary for the generalization of the findings of the present study. Finally, we did not perform an objective evaluation of sleep. However, it should be noted that sleep diaries are considered highly reliable, as sleep diary data have been proven to correspond to the results of objective polysomnographic measures of sleep (Rogers et al. Citation1993). Additionally, ambient light exposure affects sleep parameters (Korman et al. Citation2022); however, we did not obtain information on the participant’s exposure time to outdoor light, screen time, or seasonal features of light exposure. Also, it is presumed that the lifestyle of college students is affected by various factors, such as body mass index, housing style, method of school attendance, and financial conditions. These factors were not considered in this study. In future studies, these factors should be included because they are needed to conclusively corroborate the findings of this study.
Conclusions
In this study, we examined the sleep factors affecting social jetlag. The multiple regression equation utilized in this study was calculated based on the effects of sleep parameters on social jetlag. The standard deviation in wake-up time was adopted as the first factor in the stepwise selection analysis. These data suggest that social jetlag is related to prolongation of sleep duration on non-school days and to irregularities in wake-up time. Moreover, this study clarified that social jetlag is affected more by irregularities in wake-up time on school days than by prolongation of sleep time on non-school days. Clarifying the impact of these parameters on social jetlag may clarify the lifestyle habits that should be adjusted and/or avoided, and such findings may contribute to the development of new sleep improvement programs.
Author contributions
K. N. designed the research, K.N. wrote the manuscript, S.O. and K.N. analyzed the data, and Y.T. collected the data.
Acknowledgements
We sincerely appreciate the time and investment of all participants of this study. We thank Editage (www.editage.com) for English language editing and publication support.
Disclosure statement
No potential conflict of interest was reported by the authors.
Data availability statement
The data supporting the findings of this study are available from the corresponding author on reasonable request. The data are not publicly available due to privacy and ethical restrictions.
Additional information
Funding
References
- Depner CM, Melanson EL, Eckel RH, Snell-Bergeon JK, Perreault L, Bergman BC, Higgins JA, Guerin MK, Stothard ER, Morton SJ, et al. 2019. Ad libitum weekend recovery sleep fails to prevent metabolic dysregulation during a repeating pattern of insufficient sleep and weekend recovery sleep. Curr Biol. 29:957–967.e4. doi:10.1016/j.cub.2019.01.069. PMID: 30827911.
- Fischer D, Klerman EB, Phillips AJK. 2021. Measuring sleep regularity: Theoretical properties and practical usage of existing metrics. Sleep. 44:zsab103. doi:10.1093/sleep/zsab103. PMID: 33864369.
- Foster RG, Roenneberg T. 2008. Human responses to the geophysical daily, annual and lunar cycles. Curr Biol. 18:R784–R794. doi:10.1016/j.cub.2008.07.003. PMID: 18786384.
- Islam Z, Akter S, Kochi T, Hu H, Eguchi M, Yamaguchi M, Kuwahara K, Kabe I, Mizoue T. 2018. Association of social jetlag with metabolic syndrome among Japanese working population: The Furukawa Nutrition and Health Study. Sleep Med. 51:53–58. doi:10.1016/j.sleep.2018.07.003. PMID: 30099352.
- Kitamura S, Hida A, Aritake S, Higuchi S, Enomoto M, Kato M, Vetter C, Roenneberg T, Mishima K. 2014. Validity of the Japanese version of the Munich Chronotype Questionnaire. Chronobiol Int. 31:845–50. doi:10.3109/07420528.2014.914035. PMID: 24824747.
- Komada Y, Okajima I, Kitamura S, Inoue Y. 2019. A survey on social jetlag in Japan: A nationwide, cross-sectional internet survey. Sleep Biol Rhythms. 17:417–22. doi:10.1007/s41105-019-00229-w.
- Korman M, Tkachev V, Reis C, Komada Y, Kitamura S, Gubin D, Kumar V, Roenneberg T. 2022. Outdoor daylight exposure and longer sleep promote wellbeing under COVID-19 mandated restrictions. J Sleep Res. 31:e13471. doi:10.1111/jsr.13471. PMID: 8646753.
- Levandovski R, Dantas G, Fernandes LC, Caumo W, Torres I, Roenneberg T, Hidalgo MP, Allebrandt KV. 2011. Depression scores associate with chronotype and social jetlag in a rural population. Chronobiol Int. 28:771–78. doi:10.3109/07420528.2011.602445. PMID: 21895489.
- Nishimura K, Tamari Y, Yamaguchi H, Onodera S, Nagasaki K. 2022. The relationship between variation in time of waking up and going to bed for a week and social jetlag in young men. Jpn J Biometeorol. 59:15–24. doi:10.11227/seikisho.59.15.
- Parsons MJ, Moffitt TE, Gregory AM, Goldman-Mellor S, Nolan PM, Poulton R, Caspi A. 2015. Social jetlag, obesity and metabolic disorder: Investigation in a cohort study. Int J Obes (Lond). 39:842–48. doi:10.1038/ijo.2014.201. PMID: 25601363.
- Phillips AJK, Clerx WM, O’Brien CS, Sano A, Barger LK, Picard RW, Lockley SW, Klerman EB, Czeisler CA. 2017. Irregular sleep/wake patterns are associated with poorer academic performance and delayed circadian and sleep/wake timing. Sci Rep. 7:3216. doi:10.1038/s41598-017-03171-4. PMID: 28607474.
- Roenneberg T, Allebrandt KV, Merrow M, Vetter C. 2012. Social jetlag and obesity. Curr Biol. 22:939–43. doi:10.1016/j.cub.2012.03.038. PMID: 22578422.
- Roenneberg T, Pilz LK, Zerbini G, Winnebeck EC. 2019. Chronotype and social jetlag: A (self-) critical review. Biology (Basel). 8:54. doi:10.3390/biology8030054. PMID: 31336976.
- Rogers AE, Caruso CC, Aldrich MS. 1993. Reliability of sleep diaries for assessment of sleep/wake patterns. Nurs Res. 42:368–72. doi:10.1097/00006199-199311000-00010. PMID: 8247821.
- Rutters F, Lemmens SG, Adam TC, Bremmer MA, Elders PJ, Nijpels G, Dekker JM. 2014. Is social jetlag associated with an adverse endocrine, behavioral, and cardiovascular risk profile? J Biol Rhythms. 29:377–83. doi:10.1177/0748730414550199. PMID: 25252710.
- Sasawaki Y, Inokawa H, Obata Y, Nagao S, Yagita K. 2022. Association of social jetlag and eating patterns with sleep quality and daytime sleepiness in Japanese high school students. J Sleep Res. doi:10.1111/jsr.13661. PMID: 35672255.
- Short MA, Gradisar M, Lack LC, Wright HR. 2013. The impact of sleep on adolescent depressed mood, alertness and academic performance. J Adolesc. 36:1025–33. doi:10.1016/j.adolescence.2013.08.007. PMID: 24215949.
- Spiegel K, Leproult R, Van Cauter E. 1999. Impact of sleep debt on metabolic and endocrine function. Lancet. 354:1435–39. doi:10.1016/S0140-6736(99)01376-8. PMID: 10543671.
- Suka M, Yoshida K, Sugimori H. 2003. Persistent insomnia is a predictor of hypertension in Japanese male workers. J Occup Health. 45:344–50. doi:10.1539/joh.45.344. PMID: 14676413.
- Wittmann M, Dinich J, Merrow M, Roenneberg T. 2006. Social jetlag: Misalignment of biological and social time. Chronobiol Int. 23:497–509. doi:10.1080/07420520500545979. PMID: 16687322.
- Wong PM, Hasler BP, Kamarck TW, Muldoon MF, Manuck SB. 2015. Social jetlag, chronotype, and cardiometabolic risk. J Clin Endocrinol Metab. 100:4612–20. doi:10.1210/jc.2015-2923. PMID: 26580236.
- Yamanaka Y. 2016. Living environment and circadian clock in humans. Jpn J Biometeorol. 53:69–81. doi:10.11227/seikisho.53.69.