Chronobiology International, Early Online: 1–6, (2014) ! Informa Healthcare USA, Inc. ISSN: 0742-0528 print / 1525-6073 online DOI: 10.3109/07420528.2014.914035
ORIGINAL ARTICLE
Validity of the Japanese version of the Munich ChronoType Questionnaire Shingo Kitamura1, Akiko Hida1, Sayaka Aritake1, Shigekazu Higuchi2, Minori Enomoto1, Mie Kato1, Ce´line Vetter3*, Till Roenneberg3, and Kazuo Mishima1
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
1
Department of Psychophysiology, National Institute of Mental Health, National Center of Neurology and Psychiatry, Kodaira, Tokyo, Japan, 2Department of Human Science, Faculty of Design, Kyushu University, Shiobaru, Minami-ku, Fukuoka, Japan, and 3Institute of Medical Psychology, Ludwig-Maximilian-University, Munich, Germany
To assess circadian preference with a score, the Morningness-Eveningness Questionnaire (MEQ) has been used for more than 3 decades now. More recently, the Munich ChronoType Questionnaire (MCTQ) was developed: it asks for sleep-wake behavior on work and free days and uses the midpoint of sleep on free days (MSF), corrected for sleep debt accumulated during the work week as an indicator of chronotype (MSFsc). In this study, we developed a Japanese version of the MCTQ by using a translation/back-translation approach including an examination of its semantic validity. In a subsequent questionnaire survey, 450 adult men and women completed the Japanese versions of the MCTQ and MEQ. Results showed that MEQ scores were significantly negatively correlated with mid-sleep parameters assessed by the MCTQ, on both, work and free days, as well as with the chronotype measure MSFsc (r ¼ 0.580 to 0.652, all p50.001). As in the original German version, the strongest correlation was observed between MEQ score and MSF. A physiological validation study using dim light melatonin onset as a circadian phase marker (N ¼ 37) showed a high correlation between chronotype as assessed with the MSFsc (r ¼ 0.542, p50.001), and less so for MEQ score (r ¼ 0.402, p ¼ 0.055). These results demonstrate the validity of the Japanese MCTQ and provide further support of the adequacy of the MCTQ as a chronotype measure. Keywords: Diurnal preference, chronotype, morningness-eveningness, questionnaire, validation, dim light melatonin onset, circadian phase, sleep
INTRODUCTION
activities. The MEQ also asks respondents to describe their most likely response or condition for given situations (e.g. alertness after wake up and appetite). The resulting score allows the categorization into morning, neither and evening types. The Munich ChronoType Questionnaire (MCTQ, Roenneberg et al., 2003) asks respondents to provide information about actual sleep/wake behavior on work days and free days. Chronotyping is based on the assumption that individual phase of entrainment is evident in sleep timing on free days (Roenneberg et al., 2012). Thus, only participants indicating that they wake up without alarm clocks (or other constraints) can be chronotyped. In fact, a survey of approximately 5000 individuals revealed a significant correlation between the MEQ, an indicator of preferential, behavior, and the MCTQ, reflecting actual behavior (Zavada et al., 2005).
Human rest and activity cycles show a high interindividual variability, a phenomenon known as diurnal preference, circadian typology or chronotype. Behavioral differences, as observed in sleep timing, also translate into a differential timing of peak and troughs in accordance with one’s chronotype, as has been shown for the endogenous circadian rhythms of plasma melatonin, cortisol, and body temperature as well as length of the circadian period (Adan et al., 2012; Di Milia et al., 2013). To assess diurnal preference, the most widely used is the Morningness-Eveningness Questionnaire (MEQ) (Horne & Ostberg, 1976). The MEQ assumes situation with various external demands and asks respondents to specify preferred times for going to bed, waking up, going to work, and performing exercise, among other
Submitted February 5, 2014, Returned for revision March 24, 2014, Accepted April 8, 2014
*Now at: Channing Division of Network Medicine, Brigham and Women’s Hospital, and Harvard Medical School, Boston, MA, USA. Correspondence: Kazuo Mishima, Department of Psychophysiology, National Institute of Mental Health, National Center of Neurology and Psychiatry, 4-1-1 Ogawa-Higashi, Kodaira, Tokyo 187-8553, Japan. Tel: +81-42-346-2071. Fax: +81-42-346-2072. E-mail: [email protected]
1
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
2
S. Kitamura et al.
Additionally, the MCTQ also corrects for sleep debt accumulated over the work week that in turn leads; otherwise catch up sleep on free days would lead to an over-estimation of chronotype, especially in late chronotypes (Roenneberg et al., 2007a, 2012). To date, the MCTQ has been used in a variety of studies on diurnal rhythms in brain function (Peres et al., 2011) and cognition (Vetter et al., 2012), to understand the effects of Daylight Savings Time (Kantermann et al., 2007) and aging on the circadian system (Roenneberg et al., 2004), to study the effects of social jetlag on metabolism (Roenneberg et al., 2012), to study the impact of urbanization and geographical location (Roenneberg et al., 2007b), and to better grasp the influence of light at the work place (Vetter et al., 2011). The validity of MCTQ has been demonstrated for the English version (Zavada et al., 2005). However, because chronotype is dependent upon the local by light/dark cycle (Roenneberg et al., 2012) as well as genetics and age, we therefore aimed at (1) creating a Japanese version of the questionnaire and (2) validate it amongst a Japanese sample.
METHODS Translation of the MCTQ Two of the authors (KM, AH) translated the original English version of MCTQ into Japanese with the permission of the original questionnaire’s first author (TR). Subsequently, the first Japanese version was professionally back-translated into English. Two of the authors (CV, TR) reviewed the back-translated English version to check for differences in the semantic content from the original English version, and if there was any difference, the wording was corrected. To verify its semantic validity, a pilot study was done using this version of the Japanese MCTQ (N ¼ 33, all male, mean age: 23.3 ± 5.0 years; range: 19–36 years). Participants were the people who visited the National Center of Neurology and Psychiatry to participate research and subsequently volunteered to answer the second version of the Japanese version of the MCTQ. We reviewed whether their responses were in line with the intent of the questions.
participants provided informed written consent. The study was approved by the Institutional Ethics Committee of the National Center of Neurology and Psychiatry; the protocol and all the procedures were in agreement with the Declaration of Helsinki and met the ethical standards of Chronobiology International (Portaluppi et al., 2010).
Questionnaires Both, the Japanese version of the MEQ (Ishihara et al., 1984) and of the MCTQ, were implemented online. The 19-item MEQ is the standard questionnaire for assessing circadian/diurnal preference with scores ranging from 16–86, with higher (lower) scores indicating a morning (evening) chronotype. With regard to the MCTQ, we extracted three relevant phase markers (Roenneberg et al., 2012), namely the mid-point of sleep on work (MSW, computed as sleep onset on work days +0.5 sleep duration on work days) and on free days (MSF, computed as sleep onset +0.5 sleep duration on free days). If participants slept less on work days than on free days, we computed mid-sleep on free days, corrected for catch up sleep (MSFsc) as a marker for chronotype marker (MSFsc ¼ MSF (Sleep duration on free days average weekly sleep duration)/ 2: see also Roenneberg et al. (2012). Statistical analysis Normality of the MEQ score and MCTQ parameters was analyzed using the Kolmogorov–Smirnov test. One-way analysis of variance (ANOVA) was used for comparisons of MEQ and MCTQ parameters between sexes. Pearson’s product-moment correlation coefficient (r) was used as an indicator of the correlation (i.e. validity) between individual MCTQ parameters (MSW, MSF, and MSFsc) and MEQ score, as well as age with bonferroni correction. The significance probabilities were determined with appropriately adjusted p values (Wright, 1992). Results are shown as mean ± standard error of mean (SEM), with significance set at 0.05. Statistical analysis was performed using SPSS 11.5 J for Windows (Chicago, IL).
Study 1: Comparison with MEQ Japanese adults in the general population were enrolled for the evaluation of convergent validity of the newly developed Japanese version of the MCTQ, using the MEQ as the standard.
Study 2: Comparison with physiological circadian phase The convergent validity of the new Japanese MCTQ was assessed using dim light melatonin onset (DLMO) time. In parallel, MCTQ results are compared with those obtained for the MEQ.
Participants Japanese adult men and women aged 20 years were recruited by means of a leaflets, as well as online and print advertisement. Out of 504 respondents, 450 provided answers to the two entire questionnaires (203 men, age 36.4 ± 12.3 years, range 20–77 years; 247 women, age 35.1 ± 11.6 years, range 20–81 years). All
Participants Forty-nine participants were recruited via a leaflet, advertisement, mail, or website aside from subjects of STUDY 1. After the exclusion of 12 participants (5 because of incomplete survey data, 7 participants with undetermined DLMO during the study period because not reaching the threshold), data was analyzed Chronobiology International
Validity of Japanese version of MCTQ for 37 participants (14 men, 30 women; age 53.0 ± 9.6 (34–68) years). The protocol was approved by the Institutional Ethics Committee of the National Center of Neurology and Psychiatry, and written informed consent was obtained from all subjects. The protocol and all the procedures were in agreement with the Declaration of Helsinki and met the ethical standards of Chronobiology International (Portaluppi et al., 2010).
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
Questionnaire As in Study 1, the Japanese versions of the MCTQ and MEQ were used and MSW, MSF, and MSFsc were calculated as MCTQ parameters. Assessment of sleep parameters Participants wore an actigraph accelerometer (Actiwatch-L, Mini-Mitter, Bend, OR) to record their sleep schedule for 4 consecutive work days and supplemented the data with a self-completed sleep log. Mean bedtime was calculated for individual participants from their sleep-wake data. Estimation of physiological circadian phase Participants performed in-home saliva sampling over the weekend after the 4-day sleep parameter study, collecting saliva using a Salivette tube (Sarstedt, Newton, NC) once every hour starting 6 h before their mean bedtime determined from their sleep log for 4 days preceding sampling, for a total of 7 saliva samples. During the period starting 15 min before saliva sampling, participants were prohibited from eating and drinking anything other than water, and throughout the sampling were prohibited from brushing their teeth, exercising, and smoking under dim light conditions or with goggles that completely covered their field of view. Participants were also prohibited from consuming alcohol and caffeinated drinks (coffee, tea, and green tea) from the morning to the completion of saliva sampling. Saliva samples were stored frozen at 10 C and transported frozen to the National Center of Neurology and Psychiatry the following day. After thawing, samples were separated by centrifugation and quantitated by radioimmunoassay (Buhlmann Laboratories, Scho¨nenbuch, Switzerland). The threshold for melatonin onset was set at 3 pg/ml (Benloucif et al., 2008), and actual onset time for each participant was calculated by linear interpolation between melatonin levels before and after exceeding the threshold value. Dim-light melatonin onset was used as the marker of circadian phase in each participant (Arendt, 2005). Statistical analysis Normality of the DLMO was analyzed using the Kolmogorov–Smirnov test. Pearson’s product-moment correlation coefficient (r) was used to analyze the correlation between DLMO and MSW, MSF, MSFsc, or !
Informa Healthcare USA, Inc.
3
MEQ score with Bonferroni correction. The significance probabilities were determined with appropriately adjusted p values (Wright, 1992). Results are expressed as mean ± SEM, with significance set at 0.05. Statistical analysis was performed using SPSS 11.5 J for Windows.
RESULTS Semantic validation of the Japanese version of the MCTQ: pilot study Analysis of the survey answers from the 33 participants showed the majority of responses were acceptable, but some responses did not reflect the intent of the questions. For example, responses for bedtime and time to sleep were reversed in 10 subjects (30.3%) presumably because these subjects misunderstood ‘‘time to get ready to fall asleep’’ (i.e. ‘‘time for lights out’’) as ‘‘time to prepare the bed for sleeping’’, even though these questions were worded specifically to avoid incorrect estimation of the interval between bedtime and sleep onset time among subjects who engage in activities unrelated to sleeping such as reading in bed. Because such misunderstanding is seldom seen in the English version (Roenneberg, personal communication, 2011), this problem was likely associated with the traditional Japanese system of laying out a futon to sleep on, which is not a fixed piece of furniture like a Western-style bed. After careful deliberation and with the permission of the original first author, ‘‘lights out’’ was included in these problematic questions to clarify the meaning. In addition, only 24 subjects (72.7%) responded that they ‘‘work regularly’’, even though the survey question was accompanied by a note saying that ‘‘this includes being, for example, a student, a housewife, or househusband’’. This indicates that the note was unclear to most of the subjects. Therefore, to achieve a near maximum effect with a minimum revision by addressing the problem among students, we obtained permission from the original first author to change the wording to ‘‘work (go to school) regularly’’ with no modification to the note. Furthermore, some participants responded affirmatively to a question asking if they were ‘‘a shift worker’’ (i.e. a worker with rotating shifts) even though they were part-time workers with a fixed shift schedule. To solve this problem, we modified the phrase to ‘‘a worker with rotating shifts’’ because night shift workers were rarely seen in Japan. With these modifications, the items of the Japanese version of the MCTQ were finalized. Comparison with MEQ: Study 1 Mean MEQ score for the 450 subjects was 52.03 ± 0.42 (range, 29–77) with a normal distribution (D(450) ¼ 0.038, p ¼ 0.153) The mean values for the MCTQ parameters MSW, MSF, and MSFsc were 3.64 ± 0.06 h (0.21 h–9.25 h), 4.66 ± .07 h (0.25 h–11.50 h), and 4.31 ± 0.07 h (0.25 h–10.07 h), respectively. Distribution was normal for MSF (D(450) ¼ 0.035,
4
S. Kitamura et al.
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
FIGURE 1. Distributions of (A) MEQ score, (B) MSW, (C) MSF, and (D) midpoint of sleep on free days corrected for sleep debt accumulated through workdays (MSFsc). Distribution were normal for MEQ (D(450) ¼ 0.038, p ¼ 0.153), MSF (D(450) ¼ 0.035, p ¼ 0.200) and MSFsc (D(450) ¼ 0.034, p ¼ 0.200), but not for MSW (D(450) ¼ 0.058, p ¼ 0.001).
FIGURE 2. Correlation between DLMO time (43.0 pg/ml) and (A) MEQ score, (B) MSW, (C) MSF, and (D) midpoint of sleep on free days corrected for sleep debt accumulated through workdays (MSFsc). Solid and dashed lines in panels A–D represent linear regressions, while the dotted diagonals in panels B–D indicate the 1:1 relationships. MCTQ parameters were significantly correlated with DLMO time but MEQ score showed non-significant trend for correlation with DLMO time (Table 1).
p ¼ 0.200) and MSFsc (D(450) ¼ 0.034, p ¼ 0.200), but not for MSW (D(450) ¼ 0.058, p ¼ 0.001) (Figure 1). MEQ score was significantly negatively correlated with each of MSW, MSF, and MSFsc, and regardless of work days or free days, the midpoint of sleep was early for morning chronotypes and late for evening chronotypes (all p50.001). The correlation became stronger in the ascending order of MSW (r ¼ 0.580), MSFsc (r ¼ 0.612), and MSF (r ¼ 0.652). No difference by sex was observed for MEQ, MSW, MSF, or MSFsc (MEQ, F(1,440) ¼ 0.865, p ¼ 0.353; MSW, F(1,440) ¼ 0.139, p ¼ 0.709; MSF, F(1,440) ¼ 0.004, p ¼ 0.952; MSFsc, F(1,440) ¼ 0.146, p ¼ 0.703). However, age was significantly correlated with MEQ (r ¼ 0.228), MSW (r ¼ 0.202), MSF (r ¼ 0.338), and MSFsc (r ¼ 0.317), with advances in the morning chronotype and the midpoint of sleep seen with increasing age (all p50.001).
Comparison with physiological circadian phase: Study 2 Data from 37 subjects was analyzed because salivary melatonin levels did not exceed the threshold
(3.0 pg/ml) in 7 of the 44 participants who had collected saliva samples. Mean DLMO for these 37 subjects was 20.80 ± 0.24 h, and a normal distribution was shown in the range of 17.19 h–23.88 h (D(37) ¼ 0.133, p ¼ 0.096). DLMO showed non-significant trend for negative correlation with MEQ score (r ¼ 0.402, p ¼ 0.055), with lower MEQ being associated with later DLMO. On the other hand, we observed a significant positive correlation between DLMO and the individual MCTQ parameters, with the midpoint of sleep becoming delayed as DLMO was delayed regardless of work days or free days. The correlation coefficient became higher in the ascending order of MSW (r ¼ 0.462, p ¼ 0.012), MSF (r ¼ 0.472, p ¼ 0.010), and MSFsc (r ¼ 0.542, p ¼ 0.002), all of which were higher than that of MEQ score (Figure 2, Table 1).
DISCUSSION In this study, we developed the Japanese version of the MCTQ for assessing chronotype and analyzed the convergent validity between the Japanese versions of the MCTQ and the MEQ, the standard questionnaire for chronotype, as well as DLMO, one of the most Chronobiology International
Validity of Japanese version of MCTQ TABLE 1. Relationship of MEQ score and MCTQ parameters with DLMO time.
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
MEQ MSW MSF MSFsc
r
p
Slope
0.402 0.462 0.472 0.542
0.055 0.012 0.010 0.002
n/a 0.265 0.387 0.436
prominent physiological circadian phase markers. Although our pilot study revealed that questions on bedtime and time to sleep (lights-out), regular work days, and involvement in shiftwork were unclear in the Japanese version of the MCTQ, with the permission of the original first author, these questions were modified to make the content clear. In Study 1, the MCTQ parameters MSW, MSF, and MSFsc were significantly correlated with MEQ score (approximately jrj ¼ 0.6). These findings are consistent with those of a previous study involving around 5000 students (Zavada et al., 2005), demonstrating that the validity of the Japanese version of the MCTQ developed in this study is compatible with the validity of the original version. The MCTQ parameter that showed the strongest correlation with MEQ score was MSF (r ¼ 0.652), not MSFsc (r ¼ 0.612). A similar correlation was also seen with the original version because, according to Professor Roenneberg, the MEQ does not take into account sleep debt that accumulates over work days (Roenneberg et al., 2007a). Because MEQ items are concerned with preferred sleep-wake patterns and preferred mental and physical conditions upon waking under given circumstances with no social obligations attached, responses to the MEQ’s questions are potentially affected not only by individual chronotype, but also by factors such as poor wakefulness and sleep extension to compensate for accumulated sleep debt due to shortened sleep time. In particular, evening chronotypes frequently accumulate sleep debt on work days due to sleep interruption caused by social restrictions such as work, school, and childrearing. As a compensatory mechanism, evening chronotypes tends to extend their sleep time on free days (Ishihara et al., 1987; Park et al., 1997; Taillard et al., 1999). When sleep is extended in this way (i.e. to catch up on sleep), waking time is delayed much more than bedtime is. This extended sleep results in shorter periods of light exposure in the morning and a delay in sleep timing the following night due to reduced sleep pressure. This disturbance in sleep pattern over a 2-day weekend causes a phase delay of up to 30–45 min (Crowley & Carskadon, 2010; Taylor et al., 2008; Yang et al., 2001), and this delay occurs more frequently in evening chronotypes than in morning chronotypes (Roepke & Duffy, 2010). Because it is necessary to get up early the following weekday morning, social jetlag develops because of the desynchronization that occurs with a phase delay in biological rhythms (Roenneberg !
Informa Healthcare USA, Inc.
5
et al., 2003; Wittmann et al., 2006). Consequently, as with the MEQ, MSF (not corrected for sleep debt) in individuals with elevated sleep debt reflects the evening chronotype more strongly. The results of the present study therefore corroborate the claim proposed by Professor Roenneberg. In Study 2, MEQ score was significantly correlated with DLMO time (jrj ¼ 0.4), a representative indicator of circadian rhythm in individuals. This is consistent with the findings of previous studies showing that endogenous circadian phases vary by chronotype (Liu et al., 2000; Martin & Eastman, 2002), and the degrees of correlation were also compatible. However, compared with the MEQ, individual MCTQ parameters, especially MSFsc, showed a much stronger correlation with DLMO (r ¼ 0.462–0.542). Although the correlation between the phase of melatonin secretion and the midpoint of sleep has been reported by a number of studies (Burgess et al., 2003; Martin & Eastman, 2002; Terman et al., 2001), the correlation was found to be stronger when using sleep logs recorded only on weekdays than when using sleep logs recorded on weekdays and weekends (Martin & Eastman, 2002). A possible explanation for the discrepancy between that study and the present study which showed this correlation was stronger with MSFsc than MSW might be that the sleep-wake schedule on work days among the participants in Study 2 was not entrained to the phase of their circadian rhythm (social jetlag). In fact, social jetlag (MSFMSW, the indicator of social jetlag used in the MCTQ) in 38.6% of subjects was 0 (no delay in sleep timing on free days as oppose to work days), indicating that in more than half the participants sleep was socially limited during work days. This means that the MCTQ parameters can provide useful information about the characteristics of an individual’s circadian rhythms even in a group of subjects including those engaged in a work day-specific sleep–wake cycle different from their endogenous circadian phase and thus with an altered internal phase relationship. This study has several limitations. First, the sample sizes were relatively small, especially in Study 2, and were not representative of the population due to the lack of systematic selection, even though the effect sizes (equal to r values) showed sufficiently high values. Second, this study adopted a cross-sectional design, not a longitudinal design. Although the chronotype could have changed with age, predictability could have been determined by carefully considering the effects of aging. As Di Milia et al. (2013) suggested, further research with larger representative populations and/or longitudinal studies are needed to confirm our results.
CONCLUSION In this study, we developed the Japanese version of the MCTQ to evaluate chronotype. The Japanese version of
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
6
S. Kitamura et al.
the MCTQ was significantly correlated with the phase of melatonin secretion and with MEQ score, with a stronger correlation with melatonin. The results of this study indicate that the approach of the original MCTQ of using the midpoint of sleep as an indicator of chronotype is a legitimate way to capture the characteristics of an individual’s circadian rhythms, and that this useful attribute is well preserved in the newly developed Japanese version. Information generated by accurate assessment of individual chronotype is extremely important for the basic study of biological clock mechanisms and the clinical diagnosis of sleep and circadian rhythm disorders. Chronotype is also an attribute that needs to be carefully addressed when establishing an individual’s optimal sleep–wake cycle. A questionnaire survey can be an important evaluation tool in field studies and clinical and epidemiological studies demanding easy, largescale data collection and rapid feedback. Now the validity of the Japanese version of the MCTQ has been established in this study, we anticipate there will be further advances in the study of chronotype, including cross-cultural study, in the future.
DECLARATION OF INTEREST The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. A part of this study is the result of ‘‘Understanding of Molecular and Environmental Bases for Brain Health’’ carried out under the Strategic Research Program for Brain Sciences by the Ministry of Education, Culture, Sports, Science and Technology of Japan. This study was supported in part by Grants-in-Aid for Scientific Research (#22791161 and #23770289) from Japan Society for the Promotion of Science and an Intramural Research Grant (No. 23-3) for Neurological and Psychiatric Disorders from the National Center of Neurology and Psychiatry.
REFERENCES Adan A, Archer SN, Hidalgo MP, et al. (2012). Circadian typology: A comprehensive review. Chronobiol Int. 29:1153–75. Arendt J. (2005). Melatonin: Characteristics, concerns, and prospects. J Biol Rhythms. 20:291–303. Benloucif S, Burgess HJ, Klerman EB, et al. (2008). Measuring melatonin in humans. J Clin Sleep Med. 4:66–9. Burgess HJ, Savic N, Sletten T, et al. (2003). The relationship between the dim light melatonin onset and sleep on a regular schedule in young healthy adults. Behav Sleep Med. 1: 102–14. Crowley SJ, Carskadon MA. (2010). Modifications to weekend recovery sleep delay circadian phase in older adolescents. Chronobiol Int. 27:1469–92. Di Milia L, Adan A, Natale V, Randler C. (2013). Reviewing the psychometric properties of contemporary circadian typology measures. Chronobiol Int. 30:1261–71.
Horne JA, Ostberg O. (1976). A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms. Int J Chronobiol. 4:97–110. Ishihara K, Miyasita A, Inugami M, et al. (1987). Differences in sleep-wake habits and EEG sleep variables between active morning and evening subjects. Sleep. 10:330–42. Ishihara K, Saitoh T, Inoue Y, Miyata Y. (1984). Validity of the Japanese version of the Morningness-Eveningness Questionnaire. Percept Mot Skills. 59:863–6. Kantermann T, Juda M, Merrow M, Roenneberg T. (2007). The human circadian clock’s seasonal adjustment is disrupted by daylight saving time. Curr Biol. 17:1996–2000. Liu X, Uchiyama M, Shibui K, et al. (2000). Diurnal preference, sleep habits, circadian sleep propensity and melatonin rhythm in healthy human subjects. Neurosci Lett. 280:199–202. Martin SK, Eastman CI. (2002). Sleep logs of young adults with selfselected sleep times predict the dim light melatonin onset. Chronobiol Int. 19:695–707. Park YM, Matsumoto K, Seo YJ, et al. (1997). Scores on morningness-eveningness and sleep habits of Korean students, Japanese students, and Japanese workers. Percept Mot Skills. 85:143–54. Peres I, Vetter C, Blautzik J, et al. (2011). Chronotype predicts activity patterns in the neural underpinnings of the motor system during the day. Chronobiol Int. 28:883–9. Portaluppi F, Smolensky MH, Touitou Y. (2010). Ethics and methods for biological rhythm research on animals and human beings. Chronobiol Int. 27:1911–29. Roenneberg T, Allebrandt KV, Merrow M, Vetter C. (2012). Social jetlag and obesity. Curr Biol. 22:939–43. Roenneberg T, Kuehnle T, Juda M, et al. (2007a). Epidemiology of the human circadian clock. Sleep Med Rev. 11:429–38. Roenneberg T, Kuehnle T, Pramstaller PP, et al. (2004). A marker for the end of adolescence. Curr Biol. 14:R1038–9. Roenneberg T, Kumar CJ, Merrow M. (2007b). The human circadian clock entrains to sun time. Curr Biol. 17:R44–5. Roenneberg T, Wirz-Justice A, Merrow M. (2003). Life between clocks: Daily temporal patterns of human chronotypes. J Biol Rhythms. 18:80–90. Roepke SE, Duffy JF. (2010). Differential impact of chronotype on weekday and weekend sleep timing and duration. Nat Sci Sleep. 2010:213–20. Taillard J, Philip P, Bioulac B. (1999). Morningness/eveningness and the need for sleep. J Sleep Res. 8:291–5. Taylor A, Wright HR, Lack LC. (2008). Sleeping-in on the weekend delays circadian phase and increases sleepiness the following week. Sleep Biol Rhythms. 6:172–9. Terman JS, Terman M, Lo ES, Cooper TB. (2001). Circadian time of morning light administration and therapeutic response in winter depression. Arch Gen Psychiatry. 58:69–75. Vetter C, Juda M, Lang D, et al. (2011). Blue-enriched office light competes with natural light as a zeitgeber. Scand J Work Environ Health. 37:437–45. Vetter C, Juda M, Roenneberg T. (2012). The influence of internal time, time awake, and sleep duration on cognitive performance in shiftworkers. Chronobiol Int. 29:1127–38. Wittmann M, Dinich J, Merrow M, Roenneberg T. (2006). Social jetlag: Misalignment of biological and social time. Chronobiol Int. 23:497–509. Wright SP. (1992). Adjusted p values for simultaneous inference. Biometrics. 48:1005–13. Yang CM, Spielman AJ, D’Ambrosio P, et al. (2001). A single dose of melatonin prevents the phase delay associated with a delayed weekend sleep pattern. Sleep. 24:272–81. Zavada A, Gordijn MC, Beersma DG, et al. (2005). Comparison of the Munich Chronotype Questionnaire with the Horne– Ostberg’s Morningness-Eveningness Score. Chronobiol Int. 22: 267–78. Chronobiology International
ORIGINAL ARTICLE
Validity of the Japanese version of the Munich ChronoType Questionnaire Shingo Kitamura1, Akiko Hida1, Sayaka Aritake1, Shigekazu Higuchi2, Minori Enomoto1, Mie Kato1, Ce´line Vetter3*, Till Roenneberg3, and Kazuo Mishima1
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
1
Department of Psychophysiology, National Institute of Mental Health, National Center of Neurology and Psychiatry, Kodaira, Tokyo, Japan, 2Department of Human Science, Faculty of Design, Kyushu University, Shiobaru, Minami-ku, Fukuoka, Japan, and 3Institute of Medical Psychology, Ludwig-Maximilian-University, Munich, Germany
To assess circadian preference with a score, the Morningness-Eveningness Questionnaire (MEQ) has been used for more than 3 decades now. More recently, the Munich ChronoType Questionnaire (MCTQ) was developed: it asks for sleep-wake behavior on work and free days and uses the midpoint of sleep on free days (MSF), corrected for sleep debt accumulated during the work week as an indicator of chronotype (MSFsc). In this study, we developed a Japanese version of the MCTQ by using a translation/back-translation approach including an examination of its semantic validity. In a subsequent questionnaire survey, 450 adult men and women completed the Japanese versions of the MCTQ and MEQ. Results showed that MEQ scores were significantly negatively correlated with mid-sleep parameters assessed by the MCTQ, on both, work and free days, as well as with the chronotype measure MSFsc (r ¼ 0.580 to 0.652, all p50.001). As in the original German version, the strongest correlation was observed between MEQ score and MSF. A physiological validation study using dim light melatonin onset as a circadian phase marker (N ¼ 37) showed a high correlation between chronotype as assessed with the MSFsc (r ¼ 0.542, p50.001), and less so for MEQ score (r ¼ 0.402, p ¼ 0.055). These results demonstrate the validity of the Japanese MCTQ and provide further support of the adequacy of the MCTQ as a chronotype measure. Keywords: Diurnal preference, chronotype, morningness-eveningness, questionnaire, validation, dim light melatonin onset, circadian phase, sleep
INTRODUCTION
activities. The MEQ also asks respondents to describe their most likely response or condition for given situations (e.g. alertness after wake up and appetite). The resulting score allows the categorization into morning, neither and evening types. The Munich ChronoType Questionnaire (MCTQ, Roenneberg et al., 2003) asks respondents to provide information about actual sleep/wake behavior on work days and free days. Chronotyping is based on the assumption that individual phase of entrainment is evident in sleep timing on free days (Roenneberg et al., 2012). Thus, only participants indicating that they wake up without alarm clocks (or other constraints) can be chronotyped. In fact, a survey of approximately 5000 individuals revealed a significant correlation between the MEQ, an indicator of preferential, behavior, and the MCTQ, reflecting actual behavior (Zavada et al., 2005).
Human rest and activity cycles show a high interindividual variability, a phenomenon known as diurnal preference, circadian typology or chronotype. Behavioral differences, as observed in sleep timing, also translate into a differential timing of peak and troughs in accordance with one’s chronotype, as has been shown for the endogenous circadian rhythms of plasma melatonin, cortisol, and body temperature as well as length of the circadian period (Adan et al., 2012; Di Milia et al., 2013). To assess diurnal preference, the most widely used is the Morningness-Eveningness Questionnaire (MEQ) (Horne & Ostberg, 1976). The MEQ assumes situation with various external demands and asks respondents to specify preferred times for going to bed, waking up, going to work, and performing exercise, among other
Submitted February 5, 2014, Returned for revision March 24, 2014, Accepted April 8, 2014
*Now at: Channing Division of Network Medicine, Brigham and Women’s Hospital, and Harvard Medical School, Boston, MA, USA. Correspondence: Kazuo Mishima, Department of Psychophysiology, National Institute of Mental Health, National Center of Neurology and Psychiatry, 4-1-1 Ogawa-Higashi, Kodaira, Tokyo 187-8553, Japan. Tel: +81-42-346-2071. Fax: +81-42-346-2072. E-mail: [email protected]
1
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
2
S. Kitamura et al.
Additionally, the MCTQ also corrects for sleep debt accumulated over the work week that in turn leads; otherwise catch up sleep on free days would lead to an over-estimation of chronotype, especially in late chronotypes (Roenneberg et al., 2007a, 2012). To date, the MCTQ has been used in a variety of studies on diurnal rhythms in brain function (Peres et al., 2011) and cognition (Vetter et al., 2012), to understand the effects of Daylight Savings Time (Kantermann et al., 2007) and aging on the circadian system (Roenneberg et al., 2004), to study the effects of social jetlag on metabolism (Roenneberg et al., 2012), to study the impact of urbanization and geographical location (Roenneberg et al., 2007b), and to better grasp the influence of light at the work place (Vetter et al., 2011). The validity of MCTQ has been demonstrated for the English version (Zavada et al., 2005). However, because chronotype is dependent upon the local by light/dark cycle (Roenneberg et al., 2012) as well as genetics and age, we therefore aimed at (1) creating a Japanese version of the questionnaire and (2) validate it amongst a Japanese sample.
METHODS Translation of the MCTQ Two of the authors (KM, AH) translated the original English version of MCTQ into Japanese with the permission of the original questionnaire’s first author (TR). Subsequently, the first Japanese version was professionally back-translated into English. Two of the authors (CV, TR) reviewed the back-translated English version to check for differences in the semantic content from the original English version, and if there was any difference, the wording was corrected. To verify its semantic validity, a pilot study was done using this version of the Japanese MCTQ (N ¼ 33, all male, mean age: 23.3 ± 5.0 years; range: 19–36 years). Participants were the people who visited the National Center of Neurology and Psychiatry to participate research and subsequently volunteered to answer the second version of the Japanese version of the MCTQ. We reviewed whether their responses were in line with the intent of the questions.
participants provided informed written consent. The study was approved by the Institutional Ethics Committee of the National Center of Neurology and Psychiatry; the protocol and all the procedures were in agreement with the Declaration of Helsinki and met the ethical standards of Chronobiology International (Portaluppi et al., 2010).
Questionnaires Both, the Japanese version of the MEQ (Ishihara et al., 1984) and of the MCTQ, were implemented online. The 19-item MEQ is the standard questionnaire for assessing circadian/diurnal preference with scores ranging from 16–86, with higher (lower) scores indicating a morning (evening) chronotype. With regard to the MCTQ, we extracted three relevant phase markers (Roenneberg et al., 2012), namely the mid-point of sleep on work (MSW, computed as sleep onset on work days +0.5 sleep duration on work days) and on free days (MSF, computed as sleep onset +0.5 sleep duration on free days). If participants slept less on work days than on free days, we computed mid-sleep on free days, corrected for catch up sleep (MSFsc) as a marker for chronotype marker (MSFsc ¼ MSF (Sleep duration on free days average weekly sleep duration)/ 2: see also Roenneberg et al. (2012). Statistical analysis Normality of the MEQ score and MCTQ parameters was analyzed using the Kolmogorov–Smirnov test. One-way analysis of variance (ANOVA) was used for comparisons of MEQ and MCTQ parameters between sexes. Pearson’s product-moment correlation coefficient (r) was used as an indicator of the correlation (i.e. validity) between individual MCTQ parameters (MSW, MSF, and MSFsc) and MEQ score, as well as age with bonferroni correction. The significance probabilities were determined with appropriately adjusted p values (Wright, 1992). Results are shown as mean ± standard error of mean (SEM), with significance set at 0.05. Statistical analysis was performed using SPSS 11.5 J for Windows (Chicago, IL).
Study 1: Comparison with MEQ Japanese adults in the general population were enrolled for the evaluation of convergent validity of the newly developed Japanese version of the MCTQ, using the MEQ as the standard.
Study 2: Comparison with physiological circadian phase The convergent validity of the new Japanese MCTQ was assessed using dim light melatonin onset (DLMO) time. In parallel, MCTQ results are compared with those obtained for the MEQ.
Participants Japanese adult men and women aged 20 years were recruited by means of a leaflets, as well as online and print advertisement. Out of 504 respondents, 450 provided answers to the two entire questionnaires (203 men, age 36.4 ± 12.3 years, range 20–77 years; 247 women, age 35.1 ± 11.6 years, range 20–81 years). All
Participants Forty-nine participants were recruited via a leaflet, advertisement, mail, or website aside from subjects of STUDY 1. After the exclusion of 12 participants (5 because of incomplete survey data, 7 participants with undetermined DLMO during the study period because not reaching the threshold), data was analyzed Chronobiology International
Validity of Japanese version of MCTQ for 37 participants (14 men, 30 women; age 53.0 ± 9.6 (34–68) years). The protocol was approved by the Institutional Ethics Committee of the National Center of Neurology and Psychiatry, and written informed consent was obtained from all subjects. The protocol and all the procedures were in agreement with the Declaration of Helsinki and met the ethical standards of Chronobiology International (Portaluppi et al., 2010).
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
Questionnaire As in Study 1, the Japanese versions of the MCTQ and MEQ were used and MSW, MSF, and MSFsc were calculated as MCTQ parameters. Assessment of sleep parameters Participants wore an actigraph accelerometer (Actiwatch-L, Mini-Mitter, Bend, OR) to record their sleep schedule for 4 consecutive work days and supplemented the data with a self-completed sleep log. Mean bedtime was calculated for individual participants from their sleep-wake data. Estimation of physiological circadian phase Participants performed in-home saliva sampling over the weekend after the 4-day sleep parameter study, collecting saliva using a Salivette tube (Sarstedt, Newton, NC) once every hour starting 6 h before their mean bedtime determined from their sleep log for 4 days preceding sampling, for a total of 7 saliva samples. During the period starting 15 min before saliva sampling, participants were prohibited from eating and drinking anything other than water, and throughout the sampling were prohibited from brushing their teeth, exercising, and smoking under dim light conditions or with goggles that completely covered their field of view. Participants were also prohibited from consuming alcohol and caffeinated drinks (coffee, tea, and green tea) from the morning to the completion of saliva sampling. Saliva samples were stored frozen at 10 C and transported frozen to the National Center of Neurology and Psychiatry the following day. After thawing, samples were separated by centrifugation and quantitated by radioimmunoassay (Buhlmann Laboratories, Scho¨nenbuch, Switzerland). The threshold for melatonin onset was set at 3 pg/ml (Benloucif et al., 2008), and actual onset time for each participant was calculated by linear interpolation between melatonin levels before and after exceeding the threshold value. Dim-light melatonin onset was used as the marker of circadian phase in each participant (Arendt, 2005). Statistical analysis Normality of the DLMO was analyzed using the Kolmogorov–Smirnov test. Pearson’s product-moment correlation coefficient (r) was used to analyze the correlation between DLMO and MSW, MSF, MSFsc, or !
Informa Healthcare USA, Inc.
3
MEQ score with Bonferroni correction. The significance probabilities were determined with appropriately adjusted p values (Wright, 1992). Results are expressed as mean ± SEM, with significance set at 0.05. Statistical analysis was performed using SPSS 11.5 J for Windows.
RESULTS Semantic validation of the Japanese version of the MCTQ: pilot study Analysis of the survey answers from the 33 participants showed the majority of responses were acceptable, but some responses did not reflect the intent of the questions. For example, responses for bedtime and time to sleep were reversed in 10 subjects (30.3%) presumably because these subjects misunderstood ‘‘time to get ready to fall asleep’’ (i.e. ‘‘time for lights out’’) as ‘‘time to prepare the bed for sleeping’’, even though these questions were worded specifically to avoid incorrect estimation of the interval between bedtime and sleep onset time among subjects who engage in activities unrelated to sleeping such as reading in bed. Because such misunderstanding is seldom seen in the English version (Roenneberg, personal communication, 2011), this problem was likely associated with the traditional Japanese system of laying out a futon to sleep on, which is not a fixed piece of furniture like a Western-style bed. After careful deliberation and with the permission of the original first author, ‘‘lights out’’ was included in these problematic questions to clarify the meaning. In addition, only 24 subjects (72.7%) responded that they ‘‘work regularly’’, even though the survey question was accompanied by a note saying that ‘‘this includes being, for example, a student, a housewife, or househusband’’. This indicates that the note was unclear to most of the subjects. Therefore, to achieve a near maximum effect with a minimum revision by addressing the problem among students, we obtained permission from the original first author to change the wording to ‘‘work (go to school) regularly’’ with no modification to the note. Furthermore, some participants responded affirmatively to a question asking if they were ‘‘a shift worker’’ (i.e. a worker with rotating shifts) even though they were part-time workers with a fixed shift schedule. To solve this problem, we modified the phrase to ‘‘a worker with rotating shifts’’ because night shift workers were rarely seen in Japan. With these modifications, the items of the Japanese version of the MCTQ were finalized. Comparison with MEQ: Study 1 Mean MEQ score for the 450 subjects was 52.03 ± 0.42 (range, 29–77) with a normal distribution (D(450) ¼ 0.038, p ¼ 0.153) The mean values for the MCTQ parameters MSW, MSF, and MSFsc were 3.64 ± 0.06 h (0.21 h–9.25 h), 4.66 ± .07 h (0.25 h–11.50 h), and 4.31 ± 0.07 h (0.25 h–10.07 h), respectively. Distribution was normal for MSF (D(450) ¼ 0.035,
4
S. Kitamura et al.
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
FIGURE 1. Distributions of (A) MEQ score, (B) MSW, (C) MSF, and (D) midpoint of sleep on free days corrected for sleep debt accumulated through workdays (MSFsc). Distribution were normal for MEQ (D(450) ¼ 0.038, p ¼ 0.153), MSF (D(450) ¼ 0.035, p ¼ 0.200) and MSFsc (D(450) ¼ 0.034, p ¼ 0.200), but not for MSW (D(450) ¼ 0.058, p ¼ 0.001).
FIGURE 2. Correlation between DLMO time (43.0 pg/ml) and (A) MEQ score, (B) MSW, (C) MSF, and (D) midpoint of sleep on free days corrected for sleep debt accumulated through workdays (MSFsc). Solid and dashed lines in panels A–D represent linear regressions, while the dotted diagonals in panels B–D indicate the 1:1 relationships. MCTQ parameters were significantly correlated with DLMO time but MEQ score showed non-significant trend for correlation with DLMO time (Table 1).
p ¼ 0.200) and MSFsc (D(450) ¼ 0.034, p ¼ 0.200), but not for MSW (D(450) ¼ 0.058, p ¼ 0.001) (Figure 1). MEQ score was significantly negatively correlated with each of MSW, MSF, and MSFsc, and regardless of work days or free days, the midpoint of sleep was early for morning chronotypes and late for evening chronotypes (all p50.001). The correlation became stronger in the ascending order of MSW (r ¼ 0.580), MSFsc (r ¼ 0.612), and MSF (r ¼ 0.652). No difference by sex was observed for MEQ, MSW, MSF, or MSFsc (MEQ, F(1,440) ¼ 0.865, p ¼ 0.353; MSW, F(1,440) ¼ 0.139, p ¼ 0.709; MSF, F(1,440) ¼ 0.004, p ¼ 0.952; MSFsc, F(1,440) ¼ 0.146, p ¼ 0.703). However, age was significantly correlated with MEQ (r ¼ 0.228), MSW (r ¼ 0.202), MSF (r ¼ 0.338), and MSFsc (r ¼ 0.317), with advances in the morning chronotype and the midpoint of sleep seen with increasing age (all p50.001).
Comparison with physiological circadian phase: Study 2 Data from 37 subjects was analyzed because salivary melatonin levels did not exceed the threshold
(3.0 pg/ml) in 7 of the 44 participants who had collected saliva samples. Mean DLMO for these 37 subjects was 20.80 ± 0.24 h, and a normal distribution was shown in the range of 17.19 h–23.88 h (D(37) ¼ 0.133, p ¼ 0.096). DLMO showed non-significant trend for negative correlation with MEQ score (r ¼ 0.402, p ¼ 0.055), with lower MEQ being associated with later DLMO. On the other hand, we observed a significant positive correlation between DLMO and the individual MCTQ parameters, with the midpoint of sleep becoming delayed as DLMO was delayed regardless of work days or free days. The correlation coefficient became higher in the ascending order of MSW (r ¼ 0.462, p ¼ 0.012), MSF (r ¼ 0.472, p ¼ 0.010), and MSFsc (r ¼ 0.542, p ¼ 0.002), all of which were higher than that of MEQ score (Figure 2, Table 1).
DISCUSSION In this study, we developed the Japanese version of the MCTQ for assessing chronotype and analyzed the convergent validity between the Japanese versions of the MCTQ and the MEQ, the standard questionnaire for chronotype, as well as DLMO, one of the most Chronobiology International
Validity of Japanese version of MCTQ TABLE 1. Relationship of MEQ score and MCTQ parameters with DLMO time.
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
MEQ MSW MSF MSFsc
r
p
Slope
0.402 0.462 0.472 0.542
0.055 0.012 0.010 0.002
n/a 0.265 0.387 0.436
prominent physiological circadian phase markers. Although our pilot study revealed that questions on bedtime and time to sleep (lights-out), regular work days, and involvement in shiftwork were unclear in the Japanese version of the MCTQ, with the permission of the original first author, these questions were modified to make the content clear. In Study 1, the MCTQ parameters MSW, MSF, and MSFsc were significantly correlated with MEQ score (approximately jrj ¼ 0.6). These findings are consistent with those of a previous study involving around 5000 students (Zavada et al., 2005), demonstrating that the validity of the Japanese version of the MCTQ developed in this study is compatible with the validity of the original version. The MCTQ parameter that showed the strongest correlation with MEQ score was MSF (r ¼ 0.652), not MSFsc (r ¼ 0.612). A similar correlation was also seen with the original version because, according to Professor Roenneberg, the MEQ does not take into account sleep debt that accumulates over work days (Roenneberg et al., 2007a). Because MEQ items are concerned with preferred sleep-wake patterns and preferred mental and physical conditions upon waking under given circumstances with no social obligations attached, responses to the MEQ’s questions are potentially affected not only by individual chronotype, but also by factors such as poor wakefulness and sleep extension to compensate for accumulated sleep debt due to shortened sleep time. In particular, evening chronotypes frequently accumulate sleep debt on work days due to sleep interruption caused by social restrictions such as work, school, and childrearing. As a compensatory mechanism, evening chronotypes tends to extend their sleep time on free days (Ishihara et al., 1987; Park et al., 1997; Taillard et al., 1999). When sleep is extended in this way (i.e. to catch up on sleep), waking time is delayed much more than bedtime is. This extended sleep results in shorter periods of light exposure in the morning and a delay in sleep timing the following night due to reduced sleep pressure. This disturbance in sleep pattern over a 2-day weekend causes a phase delay of up to 30–45 min (Crowley & Carskadon, 2010; Taylor et al., 2008; Yang et al., 2001), and this delay occurs more frequently in evening chronotypes than in morning chronotypes (Roepke & Duffy, 2010). Because it is necessary to get up early the following weekday morning, social jetlag develops because of the desynchronization that occurs with a phase delay in biological rhythms (Roenneberg !
Informa Healthcare USA, Inc.
5
et al., 2003; Wittmann et al., 2006). Consequently, as with the MEQ, MSF (not corrected for sleep debt) in individuals with elevated sleep debt reflects the evening chronotype more strongly. The results of the present study therefore corroborate the claim proposed by Professor Roenneberg. In Study 2, MEQ score was significantly correlated with DLMO time (jrj ¼ 0.4), a representative indicator of circadian rhythm in individuals. This is consistent with the findings of previous studies showing that endogenous circadian phases vary by chronotype (Liu et al., 2000; Martin & Eastman, 2002), and the degrees of correlation were also compatible. However, compared with the MEQ, individual MCTQ parameters, especially MSFsc, showed a much stronger correlation with DLMO (r ¼ 0.462–0.542). Although the correlation between the phase of melatonin secretion and the midpoint of sleep has been reported by a number of studies (Burgess et al., 2003; Martin & Eastman, 2002; Terman et al., 2001), the correlation was found to be stronger when using sleep logs recorded only on weekdays than when using sleep logs recorded on weekdays and weekends (Martin & Eastman, 2002). A possible explanation for the discrepancy between that study and the present study which showed this correlation was stronger with MSFsc than MSW might be that the sleep-wake schedule on work days among the participants in Study 2 was not entrained to the phase of their circadian rhythm (social jetlag). In fact, social jetlag (MSFMSW, the indicator of social jetlag used in the MCTQ) in 38.6% of subjects was 0 (no delay in sleep timing on free days as oppose to work days), indicating that in more than half the participants sleep was socially limited during work days. This means that the MCTQ parameters can provide useful information about the characteristics of an individual’s circadian rhythms even in a group of subjects including those engaged in a work day-specific sleep–wake cycle different from their endogenous circadian phase and thus with an altered internal phase relationship. This study has several limitations. First, the sample sizes were relatively small, especially in Study 2, and were not representative of the population due to the lack of systematic selection, even though the effect sizes (equal to r values) showed sufficiently high values. Second, this study adopted a cross-sectional design, not a longitudinal design. Although the chronotype could have changed with age, predictability could have been determined by carefully considering the effects of aging. As Di Milia et al. (2013) suggested, further research with larger representative populations and/or longitudinal studies are needed to confirm our results.
CONCLUSION In this study, we developed the Japanese version of the MCTQ to evaluate chronotype. The Japanese version of
Chronobiol Int Downloaded from informahealthcare.com by Emory University on 06/21/14 For personal use only.
6
S. Kitamura et al.
the MCTQ was significantly correlated with the phase of melatonin secretion and with MEQ score, with a stronger correlation with melatonin. The results of this study indicate that the approach of the original MCTQ of using the midpoint of sleep as an indicator of chronotype is a legitimate way to capture the characteristics of an individual’s circadian rhythms, and that this useful attribute is well preserved in the newly developed Japanese version. Information generated by accurate assessment of individual chronotype is extremely important for the basic study of biological clock mechanisms and the clinical diagnosis of sleep and circadian rhythm disorders. Chronotype is also an attribute that needs to be carefully addressed when establishing an individual’s optimal sleep–wake cycle. A questionnaire survey can be an important evaluation tool in field studies and clinical and epidemiological studies demanding easy, largescale data collection and rapid feedback. Now the validity of the Japanese version of the MCTQ has been established in this study, we anticipate there will be further advances in the study of chronotype, including cross-cultural study, in the future.
DECLARATION OF INTEREST The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. A part of this study is the result of ‘‘Understanding of Molecular and Environmental Bases for Brain Health’’ carried out under the Strategic Research Program for Brain Sciences by the Ministry of Education, Culture, Sports, Science and Technology of Japan. This study was supported in part by Grants-in-Aid for Scientific Research (#22791161 and #23770289) from Japan Society for the Promotion of Science and an Intramural Research Grant (No. 23-3) for Neurological and Psychiatric Disorders from the National Center of Neurology and Psychiatry.
REFERENCES Adan A, Archer SN, Hidalgo MP, et al. (2012). Circadian typology: A comprehensive review. Chronobiol Int. 29:1153–75. Arendt J. (2005). Melatonin: Characteristics, concerns, and prospects. J Biol Rhythms. 20:291–303. Benloucif S, Burgess HJ, Klerman EB, et al. (2008). Measuring melatonin in humans. J Clin Sleep Med. 4:66–9. Burgess HJ, Savic N, Sletten T, et al. (2003). The relationship between the dim light melatonin onset and sleep on a regular schedule in young healthy adults. Behav Sleep Med. 1: 102–14. Crowley SJ, Carskadon MA. (2010). Modifications to weekend recovery sleep delay circadian phase in older adolescents. Chronobiol Int. 27:1469–92. Di Milia L, Adan A, Natale V, Randler C. (2013). Reviewing the psychometric properties of contemporary circadian typology measures. Chronobiol Int. 30:1261–71.
Horne JA, Ostberg O. (1976). A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms. Int J Chronobiol. 4:97–110. Ishihara K, Miyasita A, Inugami M, et al. (1987). Differences in sleep-wake habits and EEG sleep variables between active morning and evening subjects. Sleep. 10:330–42. Ishihara K, Saitoh T, Inoue Y, Miyata Y. (1984). Validity of the Japanese version of the Morningness-Eveningness Questionnaire. Percept Mot Skills. 59:863–6. Kantermann T, Juda M, Merrow M, Roenneberg T. (2007). The human circadian clock’s seasonal adjustment is disrupted by daylight saving time. Curr Biol. 17:1996–2000. Liu X, Uchiyama M, Shibui K, et al. (2000). Diurnal preference, sleep habits, circadian sleep propensity and melatonin rhythm in healthy human subjects. Neurosci Lett. 280:199–202. Martin SK, Eastman CI. (2002). Sleep logs of young adults with selfselected sleep times predict the dim light melatonin onset. Chronobiol Int. 19:695–707. Park YM, Matsumoto K, Seo YJ, et al. (1997). Scores on morningness-eveningness and sleep habits of Korean students, Japanese students, and Japanese workers. Percept Mot Skills. 85:143–54. Peres I, Vetter C, Blautzik J, et al. (2011). Chronotype predicts activity patterns in the neural underpinnings of the motor system during the day. Chronobiol Int. 28:883–9. Portaluppi F, Smolensky MH, Touitou Y. (2010). Ethics and methods for biological rhythm research on animals and human beings. Chronobiol Int. 27:1911–29. Roenneberg T, Allebrandt KV, Merrow M, Vetter C. (2012). Social jetlag and obesity. Curr Biol. 22:939–43. Roenneberg T, Kuehnle T, Juda M, et al. (2007a). Epidemiology of the human circadian clock. Sleep Med Rev. 11:429–38. Roenneberg T, Kuehnle T, Pramstaller PP, et al. (2004). A marker for the end of adolescence. Curr Biol. 14:R1038–9. Roenneberg T, Kumar CJ, Merrow M. (2007b). The human circadian clock entrains to sun time. Curr Biol. 17:R44–5. Roenneberg T, Wirz-Justice A, Merrow M. (2003). Life between clocks: Daily temporal patterns of human chronotypes. J Biol Rhythms. 18:80–90. Roepke SE, Duffy JF. (2010). Differential impact of chronotype on weekday and weekend sleep timing and duration. Nat Sci Sleep. 2010:213–20. Taillard J, Philip P, Bioulac B. (1999). Morningness/eveningness and the need for sleep. J Sleep Res. 8:291–5. Taylor A, Wright HR, Lack LC. (2008). Sleeping-in on the weekend delays circadian phase and increases sleepiness the following week. Sleep Biol Rhythms. 6:172–9. Terman JS, Terman M, Lo ES, Cooper TB. (2001). Circadian time of morning light administration and therapeutic response in winter depression. Arch Gen Psychiatry. 58:69–75. Vetter C, Juda M, Lang D, et al. (2011). Blue-enriched office light competes with natural light as a zeitgeber. Scand J Work Environ Health. 37:437–45. Vetter C, Juda M, Roenneberg T. (2012). The influence of internal time, time awake, and sleep duration on cognitive performance in shiftworkers. Chronobiol Int. 29:1127–38. Wittmann M, Dinich J, Merrow M, Roenneberg T. (2006). Social jetlag: Misalignment of biological and social time. Chronobiol Int. 23:497–509. Wright SP. (1992). Adjusted p values for simultaneous inference. Biometrics. 48:1005–13. Yang CM, Spielman AJ, D’Ambrosio P, et al. (2001). A single dose of melatonin prevents the phase delay associated with a delayed weekend sleep pattern. Sleep. 24:272–81. Zavada A, Gordijn MC, Beersma DG, et al. (2005). Comparison of the Munich Chronotype Questionnaire with the Horne– Ostberg’s Morningness-Eveningness Score. Chronobiol Int. 22: 267–78. Chronobiology International
