The reduced Morningness–Eveningness Questionnaire: Psychometric properties and related factors in a young Swedish population

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=icbi20

Chronobiology International

The Journal of Biological and Medical Rhythm Research

ISSN: 0742-0528 (Print) 1525-6073 (Online) Journal homepage: https://www.tandfonline.com/loi/icbi20

The reduced Morningness–Eveningness

Questionnaire: Psychometric properties and related factors in a young Swedish population

Katarina Danielsson, Aysegül Sakarya & Markus Jansson-Fröjmark

To cite this article: Katarina Danielsson, Aysegül Sakarya & Markus Jansson-Fröjmark (2019) The reduced Morningness–Eveningness Questionnaire: Psychometric properties and related factors in a young Swedish population, Chronobiology International, 36:4, 530-540, DOI:

10.1080/07420528.2018.1564322

To link to this article: https://doi.org/10.1080/07420528.2018.1564322

© 2019 The Author(s). Published with license by Taylor & Francis Group, LLC.

Published online: 07 Jan 2019.

Submit your article to this journal

Article views: 254

View Crossmark data

The reduced Morningness –Eveningness Questionnaire: Psychometric properties and related factors in a young Swedish population

Katarina Danielsson

, Aysegül Sakarya

, and Markus Jansson-Fröjmark

a

Department of Neuroscience, Psychiatry, Uppsala University, Uppsala, Sweden;

Centre for Psychiatry Research, Department of Clinical Neuroscience, Karolinska Institute, Stockholm, Sweden

ABSTRACT

During puberty, there is a delay in sleep patterns and the circadian clock, and the prevalence of sleep difficulties is high among young adults. Thus, it is important to be able to measure chronotype (CT) in this group, both in the clinic and for broader epidemiological studies, to better understand the sleep difficulties observed. The reduced version of one of the most widely used questionnaire to measure CT, the Horne & Östberg Morningness –Eveningness Questionnaire (MEQ), has been developed and validated in many languages. The aim of the present study was to investigate the reliability and validity of the Swedish reduced MEQ (rMEQ) and to investigate factors correlated with rMEQ scores in a random sample of young Swedish participants. We sent the rMEQ, sleep questions, demographic questions, the Insomnia Severity Index (ISI), the Epworth Sleepiness Scale, and the Hospital Anxiety and Depression Scale (HADS-A and HADS-D) to 1000 randomly selected individuals aged 16 –26 years in Uppsala, Sweden (response rate = 68%).

A parallel analysis of the rMEQ revealed a single factor solution. Inter-item correlations within the questionnaire were between 0.08 and 0.46. One of the rMEQ items displayed weak correla- tions with the other questions. Cronbach ’s alpha of the rMEQ was 0.68. Low rMEQ scores (eveningness) were correlated with more frequent late bedtimes and difficulties getting up in the morning. Lower rMEQ scores were significantly related to male sex, no educational activity or work, nicotine use, younger age, and higher ISI and HADS-D scores. Even though the inter-item correlations of the rMEQ were very low to moderate, the rMEQ had a Cronbach ’s alpha not far off from acceptable levels, and the strong correlations of the rMEQ scores with responses to the sleep questions indicated that the rMEQ had an adequate validity. To conclude, the rMEQ can be used to effectively estimate CT when a short questionnaire is required; however, caution should be taken when interpreting the results considering the reliability of the questionnaire.

ARTICLE HISTORY Received 24 August 2018 Revised 14 December 2018 Accepted 15 December 2018 KEYWORDS

Reduced Morningness – Eveningness Questionnaire;

chronotype; young adults;

adolescents

Introduction

Circadian rhythms are endogenous 24-h oscillations that are primarily regulated by the anterior hypotha- lamic suprachiasmatic nucleus. This regulation is influenced by genetic factors and environmental fac- tors such as light–dark cycles. Circadian rhythms coordinate our sleep–wake preferences, body tem- perature, hormonal secretion, food intake, and circu- lation (Weaver 2002; Young 2018). The preferred time of the day to be active or to sleep defines morning- ness–eveningness, which is also termed chronotype (CT). For example, morning-oriented individuals (morning types/“larks”) go to bed early and reach their peak cognitive and physical performance in the early daytime hours, while evening-oriented indivi- duals go to bed in the early hours of the morning

and function best at later times in the day or the evening. There are also “neither” types who have a balanced circadian rhythm (Randler et al. 2016).

There is a known delay of circadian rhythms during puberty, and eveningness CT is high in adolescence (Borisenkov et al. 2010; Hagenauer et al. 2009).

Eveningness is associated with nicotine, alcohol and drug use, physical inactivity, sleep difficulties, and symptoms of depression and anxiety, whereas morn- ingness is associated with healthy activities and higher conscientiousness (Adan 1994; Urban et al. 2011). CT is also linked to some sleep disorders such as delayed sleep–wake phase disorder (DSWPD) and advanced sleep–wake phase disorder (American Academy of Sleep Medicine 2014). It is of interest to measure CT in adolescents and young adults since sleep distur- bances such as DSWPD and insomnia are common

CONTACT Katarina Danielsson katarina.danielsson@neuro.uu.se Department of Neuroscience, Psychiatry, Uppsala University, Uppsala SE-75185, Sweden

2019, VOL. 36, NO. 4, 530 –540

https://doi.org/10.1080/07420528.2018.1564322

© 2019 The Author(s). Published with license by Taylor & Francis Group, LLC.

This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/),

which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.

in this group. However, many of the CT question- naires are lengthy. To have a short and easy way to measure CT would be beneficial especially for large population studies.

Several questionnaires have been designed to measure CT. The Morningness–Eveningness Questionnaire (MEQ) is one of the most widely used and accepted instruments to measure CT and is reported to be a useful tool for assessing CT in DSWPD (American Academy of Sleep Medicine 2014). The MEQ is considered as the gold stan- dard measure of CT and has been translated into several languages. The Östberg Swedish language MEQ (Ostberg 1973) was used as the basis when the English version was developed by Horne and Östberg in 1976 (Horne and Ostberg 1976).

A number of large-scale studies with healthy sub- jects have reported MEQ to have a good reliability and validity in various countries (Di Milia et al.

2013; Lee et al. 2014; Roveda et al. 2017). The MEQ contains 19 items concerning sleep–wake times, and preferred times for physical and mental activities and subjective alertness, which are scored using a Likert-type response format. It has been suggested that cutoff scores of the MEQ are affected by age, sex, cultural variations, season of birth, and puberty (Adan and Natale 2002; Caci et al. 2009; Kim et al. 2002). A Swedish translation and reliability study of the MEQ by Ståleby and colleagues reported a high reliability of the scale in a healthy adult population (Ståleby et al. 2016).

However, the MEQ has been criticized for its length and that it contains items with varying num- ber of possible choices (Di Milia et al. 2013).

Furthermore, additional concerns about the under- lying factor structure have led to the development of a reduced 5-item version (rMEQ). Using shorter scales can be advantageous for screening larger popu- lations; shorter scales are also much easier for parti- cipants to fill in, which might increase compliance.

Considering that adolescents may be more impatient than older adults, it is even more important to use assessment tools that are short and easy to apply.

The rMEQ was developed by Adan and Almirall (1991) and only includes items 1, 7, 10, 18, and 19 of the original MEQ (Adan and Almirall 1991).

The briefness and psychometric properties of the rMEQ mean that it has been widely used and it has been evaluated in Hungary, France, Spain,

America, Italy, the Kingdom of Saudi Arabia, China, Germany, Poland, India, Iraq, Iran, and Portugal (Adan and Almirall 1991; BaHammam et al. 2011; Biswas et al. 2014; Caci et al. 2009;

Carciofo et al. 2012; Chelminski et al. 2000;

Jankowski 2013; Loureiro and Garcia-Marques 2015; Natale et al. 2006; Rahafar et al. 2015;

Randler 2013; Raoof et al. 2014; Urban et al.

2011). With adequate psychometric properties, the rMEQ has good convergent and construct validity with the MEQ and good test–retest relia- bility (Young 2018). Most of these previous studies found a one-factor structure and similar values for internal consistency. To our knowledge, the rMEQ has not yet been evaluated in Scandinavia.

Furthermore, most previous studies have been conducted in samples where the subjects have had similar daytime commitments, e.g. samples of university students (BaHammam et al. 2011;

Biswas et al. 2014; Caci et al. 2009; Chelminski et al. 2000; Jankowski 2013; Natale et al. 2006;

Rahafar et al. 2015; Raoof et al. 2014), high- school students, or undergraduates (Loureiro and Garcia-Marques 2015). Only a few studies have investigated the rMEQ in groups with larger age spans and different occupations (Adan and Almirall 1991; Carciofo et al. 2012; Randler 2013).

Our primary aim was to test the psychometric properties of the Swedish version of the rMEQ, with a focus on factor structure, reliability and descriptive statistics in both the total sample and across subgroups (age, gender, occupation, tobacco use, alcohol consumption and drug use).

To fairly validate the measurement of CT with the rMEQ, questions about sleep onset and difficulties rising in the morning were used. Additionally, as symptoms of depression, anxiety sleep difficulties and daytime sleepiness often are related to even- ingness CT (Alvaro et al. 2014; Au and Reece 2017;

Taillard et al. 1999; Tutek et al. 2016), we wanted to investigate the relation between these symptoms with rMEQ to further understand its properties.

Materials and methods Participants

Participants were 1000 individuals selected ran-

domly from the Swedish Population Register in

Uppsala, Sweden, aged between 16 and 26 years (mean age = 21.8 years ± 3.1). Between October 2014 and May 2015, participants were sent a questionnaire about sleep and health by post. After two reminders, 671 participants (68%) responded.

Responders received two cinema tickets as com- pensation. An advisory statement concerning human subjects was received from the Regional Ethical Board, Uppsala, Sweden.

Measurement instruments

Participants received a self-reported questionnaire that included demographic questions about age;

sex; occupation; residential status; nicotine, alco- hol, and drug use; and assessment questions regarding sleep patterns.

The rMEQ

Total scores of the 5-item rMEQ range from 4 to 26, whereby a higher score indicates a morningness CT.

The same cutoff scores for determining CT groups were used as in Adan and Almirall (1991) (evening- ness: < 12; neither: 12–17; morning: > 17). We did not translate the rMEQ since the MEQ is based on a Swedish MEQ (Ostberg 1973). Furthermore as one of the authors is from Sweden, we had the possibility to extract the rMEQ items from the Swedish version of the MEQ by Östberg. The questions are displayed in Table 1. The original scoring was used where the first four questions were scored 1–5 and question 5 was scored 0–6.

Sleep variables

The sleep questions used to validate rMEQ were as follows: falling asleep time on most days of the week (scored from 1 to 5, where 1 was before 21:00 p.m. and 5 was later than 02:00 a.m.); the frequency per week of falling asleep at 01:00 a.m.

or later (scored from 0 to 7 days/week); the fre- quency of being late for school or work because of difficulties waking up in the morning (scored from 1 to 5, where 1 was “never late” and 5 was

“late almost every day of the week”); the most common sleep duration (scored from 1 to 5, where 1 was more than 9 h and 5 was less than 6 h); difficulties in falling asleep earlier if needed and in rising at 09:00 a.m. (scored from 1 to 5, where 1 was “not at all difficult” and 5 was

“totally impossible”).

The Insomnia Severity Index (ISI)

The ISI is a 7-item scale that is scored using a 5-point Likert scale and is a standardized measure of sleep disturbance (Bastien et al. 2001). The ISI assesses the following seven domains of insomnia:

severity of initial, middle, and late insomnia; sleep satisfaction; interference of insomnia with daytime functioning; noticeability of sleep problems by others; and distress about sleep difficulties. Total scores range from 0 to 28, whereby higher scores indicate more severe sleep difficulties.

The Epworth Sleepiness Scale (ESS)

The ESS is an 8-item symptom scale that is used to evaluate daytime sleepiness. Each item is scored on a 4-point Likert scale, and the total score ranges from 0 to 24. Higher scores indicate greater day- time sleepiness (Johns 1991).

The Hospital Anxiety and Depression Scale (HADS) The HADS is used to evaluate symptoms of anxi- ety and depression (Zigmond and Snaith 1983).

This scale consists of two subscales (anxiety, HADS-A; and depression, HADS-D), each of which has seven items that are scored on a 4-point Likert scale. Total scores range between 0 and 21, and higher scores indicate more severe symptoms of anxiety or depression.

Table 1. Factor loadings of the rMEQ in an exploratory factor analysis with maximum likelihood extraction and direct oblimin rotation.

Mean (SD)

Factor loadings 1. Considering only your own diurnal

rhythm, at what time would you get up if you were entirely free to plan your day?

2.8 (0.9) 0.57

2. After the first half hour after having woken in the morning, how tired do you feel?

2.1 (0.8) 0.36

3. At what time in the evening do you feel tired and as a result in need of sleep?

3.2 (1.0) 0.50

4. At what time of the day do you think

that you reach your “feeling best” peak? 3.0 (0.8) 0.56 5. One hears about morning and evening

types of people. Which one of these types do you consider yourself to be?

2.2 (1.7) 0.82

SD = standard deviation.

Statistical analysis

Data analyses were performed using SPSS Statistics version 20.0. A parallel analysis was used to assess the factor structure of the rMEQ (O’Connor 2000), and Cronbach’s alpha was used to assess the internal consistency of the rMEQ.

Composite reliability (i.e. the total amount of true score variance in relation to the total scale score variance) was calculated according to guidelines (Raykov 1997). Composite reliability above the 0.70 threshold has been suggested as acceptable (Hair et al. 1998). A one-way analysis of variance was used to identify significant differences in total rMEQ scores between subgroups that were determined according to age (16–19 years old versus 20–26 years old), sex (male versus female), alcohol intake (less than twice a week versus twice or more a week), drugs (tested versus never tested), nicotine (nicotine use versus no nicotine use), and occupational status (educational activity or work versus no educational activity or work).

The number of participants in each subgroup is seen in Table 2. Since age is an interval variable, the correlation between age and rMEQ-score was also analyzed with Pearson correlation coeffi- cient. The correlation between rMEQ-scores and the sleep variables, ESS scores, ISI scores, HADS- A scores, and HADS-D scores were also assessed using Pearson correlation coefficient. The strengths of the correlations were set according to those outlined by Cohen, whereby r = 0.1–0.3 was defined as a weak correlation, r = 0.3–0.5 a moderate correlation, and r > 0.5 a strong cor- relation (Cohen 1988).

Results

Factor structure of the rMEQ

The initial analyses of the factorability of the rMEQ resulted in a Kaiser–Meyer–Olkin measure of sampling adequacy index of 0.74 and a significant Bartlett Test of Sphericity (χ ² = 584.31, p ≤ 0.001), which indicated that the data were suitable for factor analysis. The parallel analysis suggested a one-factor solution. A further exploration of this solution (explaining 45.2% of the variance) revealed four items with low com- munalities (i.e. <0.40; item 1: 0.32, item 2: 0.13, item 3: 0.25, and item 4: 0.31) and one item with adequate communality (item 5: 0.66). The factor loadings of the rMEQ are displayed in Table 1.

Subgroup analyses demonstrated similar factorial findings for the following demographic para- meters: age (teenagers and young adults sepa- rately), sex (women and men separately), and occupational status (persons with no educational activity or work and persons with an educational activity or work separately). These latter analyses are not presented in detail in this paper but can be sent to interested readers.

Internal consistency of the rMEQ

The internal consistency of the rMEQ was α = 0.68.

Cronbach’s alpha, analyzed from the total rMEQ, decreased slightly after removing items 1, 3, and 4 (the value of the alpha ranged between 0.61 and 0.63) and more substantially after removing item 5 (α = 0.55). The inter-item correlations of the rMEQ

Table 2. Descriptive statistics across subgroups.

n (%) rMEQ: total [M (SD)] F

Age (years) 16 –19 186 (27.8) 12.5 (3.5) 9.35**

20 –26 483 (72.2) 13.4 (3.6)

Sex Female 370 (55.4) 13.7 (3.5) 15.83**

Male 299 (44.6) 12.6 (3.6)

Occupational status Educational activity or work 629 (94.4) 13.3 (3.5) 8.98**

No educational activity or work 37 (5.6) 11.5 (4.5)

Alcohol use Less than 2 times per week 589 (87.9) 13.2 (3.5) 1.10

2 times or more per week 81 (12.1) 12.8 (3.8)

Nicotine use No 527 (78.5) 13.5 (3.5) 11.71**

Yes 137 (20.5) 11.9 (3.4)

Drug use No 635 (94.6) 13.2 (3.6) 3.17

Yes 34 (5.4) 12.1 (3.1)

* p < .05. **p < .01.

were between 0.08 and 0.46. The composite reliabil- ity for the rMEQ items was acceptable at 0.71.

Descriptive statistics for the rMEQ in the total sample and in subgroups

In the total sample, the rMEQ scores were nor- mally distributed (kurtosis: 0.03; skewness: −0.21).

The mean rMEQ total score was 13.1 (SD = 3.5) and ranged from 4 to 24. The mean score for each of the five items are displayed in Table 1.

According to the cutoff scores, we identified 118 (32.5%) evening types, 374 (55.7%) neither types, and 79 (11.7%) morning types.

We examined subgroup differences in rMEQ scores, according to age; sex; occupational status; and use of nicotine, alcohol, and drugs (see Table 2). The following participants scored significantly lower on the rMEQ: younger participants (16–19 years old) compared to young adults (BaHammam et al. 2011;

Biswas et al. 2014; Carciofo et al. 2012; Jankowski 2013; Natale et al. 2006; Randler 2013; Raoof et al.

2014), male participants compared to female partici- pants, participants with no educational activity or work compared to those with an educational activity or work, and those who used nicotine compared to those who did not use nicotine. Age had a positive correlation to rMEQ. When age was analyzed as an interval variable, it demonstrated a correlation with rMEQ (r = 0.13, p < 0.01), where lower age was correlated with lower scores on rMEQ.

Validity of the rMEQ

To examine the validity of the rMEQ, scores of the rMEQ and its five items were correlated with sleep onset timing, frequency of falling

asleep at 1:00 a.m. or later, frequency of being late to school/work because of difficulties waking up in the morning, sleep duration, dif- ficulties falling asleep earlier if needed, and difficulties rising at 09:00 a.m. (see Table 3).

A negative correlation was observed between rMEQ score and the sleep variables. Whereas being more evening-oriented was correlated to late sleep onset timing, high frequency of fall- ing asleep at 1:00 a.m. or later, shorter sleep duration, higher frequency of being late to work/school because of difficulties waking up in the morning, difficulties falling asleep earlier if needed, and difficulties rising at 9:00 a.m.

There were similar results for the five rMEQ items when they were tested separately.

Correlations between rMEQ and its items with ISI, ESS, HADS-A, and HADS-D

The rMEQ and its items (except for item 3) were negatively correlated with ISI scores and HADS-D scores. Negative correlations were also demonstrated between two of the rMEQ items (Young 2018;

Randler et al. 2016) and ESS scores and HADS-A scores. The results are displayed in Table 4.

Discussion

The Swedish version of the rMEQ seems to have acceptable psychometric properties regarding validity; however, the internal consistency was only satisfactory and the percentage of explained variance was relatively low. Across the sub- groups male gender, younger age (teenagers), no work or educational activity, and nicotine use were associated with eveningness CT.

Table 3. Correlations between rMEQ scores and questions assessing sleep onset, sleep duration, and rising in the morning.

rMEQ:

total

rMEQ:

item 1

rMEQ:

item 2

rMEQ:

item 3

rMEQ:

item 4

rMEQ:

item 5

Sleep onset timing −0.56** −0.34** −0.15** −0.51** −0.40** −0.46**

Frequency of falling asleep at 1:00 a.m. or later −0.51** −0.34** −0.14** −0.45** −0.36** −0.41**

Sleep duration −0.08* 0.07 −0.17** −0.04 −0.03 −0.08*

Frequency of being late to school/work because of difficulties waking up in the morning

−0.33** −0.21** −0.28** −0.09* −0.18** −0.32**

Difficulties falling asleep earlier if needed −0.36** −0.19** −0.31** −0.16** −0.23** −0.31**

Difficulties rising at 09:00 a.m. −0.50** −0.43** −0.32** −0.28** −0.28** −0.39**

* p < 0.05. **p < 0.01.

Moreover, a correlation was observed between lower rMEQ scores and higher severity of sleep difficulties and depressive symptoms.

Psychometric properties of the rMEQ

In the present study, the reliability and validity of the Swedish version of the rMEQ was evaluated in a random sample of adolescents and young adults.

The rMEQ displayed a one-factor structure and could thus be regarded as unidimensional. This finding was stable even when dividing data into different subgroups depending on sociodemo- graphic data. This is consistent with most research on the rMEQ (Jankowski 2013; Rahafar et al. 2015;

Randler 2013). Our understanding of the useful- ness of the rMEQ could benefit from using con- firmatory factor analysis and item response theory analysis in future research.

The Swedish rMEQ had a satisfactory but not optimal reliability. The Cronbach’s alpha was some- what lower (0.68) than the usually accepted level of 0.70. Other evaluations of the rMEQ have found a Cronbach’s alpha between 0.69 and 0.78 (Caci et al. 2009; Jankowski 2013). In one Hungarian sam- ple, Cronbach’s alpha was even lower at 0.56 (Urban et al. 2011). In that study, the mean age of partici- pants was younger (15.3 years) than in the present study (21.8 years), and the rMEQ had an additional question (morning fatigue/freshness on free days).

Low Cronbach’s alpha values can be caused by a small sample size, a small number of items, the administration of data-collecting tools, or poor inter-relations between items. However, we had an acceptable sample size of 671 participants; thus, the consistency of the questionnaire with only five items might contribute to the low Cronbach’s alpha. In the present study, we do not know how motivated the responders were when they filled out the question- naire and we do not know if the responders filled out the entire questionnaire at once or did a bit every

now and then. We tried to limit the amount of questions but there is always a risk that question- naires appear too lengthy. Furthermore, our sample included a randomly selected group of young adults and adolescents, compared to samples including only college or university students or youngsters in secondary school, which are more often observed in other studies of the rMEQ. This might have affected both the Cronbach´s alpha and the inter-item corre- lations. A weak correlation was observed between items 2 and 3, and item 2 (which measures tiredness when waking up) had a weak correlation with total rMEQ-scores and with each of the items of the rMEQ separately. The finding of a lower inter-item value for item 2 is in accordance with the findings of other studies (Loureiro and Garcia-Marques 2015;

Randler 2013) Morning tiredness is related to CT preference; however, feeling tired in the morning might not only be caused by a late CT, but could also be the result of other health or social problems that disturb sleep. Furthermore, the question might be better adapted to a group with similar morning commitments since the question never addresses at what time the respondent actually wakes up in the morning.

In the present study, the distribution of the total rMEQ scores in the sample were neither > evening- ness > morningness CT. This is in accordance with earlier studies conducted in healthy young adults and adolescents (BaHammam et al. 2011;

Chelminski et al. 2000; Natale et al. 2006).

However, some other studies found an order of neither > morningness > eveningness CT (Loureiro and Garcia-Marques 2015; Raoof et al. 2014; Urban et al. 2011). These three studies were conducted in Hungary, the Middle East, and Portugal. It is known that latitude can influence CT distribution, whereby the CT of individuals living in northern Europe tends to be more eveningness-oriented compared to those living in Central Europe (Borisenkov et al.

2010). Furthermore, our results can be explained by

Table 4. Correlations between rMEQ scores and sleep difficulties, daytime sleepiness, anxiety, and depressive symptoms.

rMEQ: total rMEQ: item 1 rMEQ: item 2 rMEQ: item 3 rMEQ: item 4 rMEQ: item 5

ISI −0.30** −0.18** −0.41** −0.07 −0.20** −0.21**

ESS −0.06 −0.05 −0.20** 0.06 −0.08 −0.01

HADS – Anxiety −0.05 −0.01 −0.28** 0.10* −0.02 −0.03

HADS – Depression −0.25** −0.14** −0.31** −0.06 −0.16** −0.19**

ESS = Epworth Sleepiness Scale, HADS = Hospital Anxiety and Depression Scale, ISI = Insomnia Severity Index. * p < 0.05, **p < 0.01.

several other factors such as differences in time for sunrise, different climate, and cultural differences.

There is evidence that the circadian clock synchro- nizes to sunrise where earlier time for sunrise leads to earlier awakenings (Randler 2008) and that morn- ing CT are more common in countries with warmer climate (Tonetti et al. 2012). Additionally, cultural differences are important for the distribution of CT.

The study from Sudan where time for morning prayers was suggested as a reason for the high pre- valence of morningness CT is a typical example of the importance of cultural differences (BaHammam et al. 2011).

In line with several other studies (Randler 2013;

Randler et al. 2016; Urban et al. 2011), we found that male participants had a significantly higher eveningness preference according to rMEQ than female participants. Night activities such as com- puter gaming and non-favorable sleep habits such as nicotine, alcohol, and drug use are often more common among young men (Kuhn 2015;

Willoughby 2018) and might cause the increased eveningness CT in this group. However, there are also studies that have failed to find sex differences for total rMEQ scores (Caci et al. 2009; Jankowski 2013; Loureiro and Garcia-Marques 2015; Raoof et al. 2014).

Furthermore, absence of educational or work activities was associated with eveningness in the present study. This might be due to several rea- sons; both unemployment and sick leave can lead to socioeconomic deprivation and can result in sleep problems (Paine et al. 2014), and there is an increased incidence of psychiatric disorders in those with eveningness CT (Au and Reece 2017;

Reid et al. 2012), which can make ongoing employment difficult. Additionally, there is no pressing need to wake up at a certain time when one is unemployed.

It is known that the midpoint of sleep becomes increasingly later from 10 years of age until about 20 years of age, after which people tend toward earlier CTs again (Roenneberg et al. 2004). This phenomenon is in accordance with our finding that significantly lower rMEQ scores (evening- ness) were related to the teenage group compared to the young adults. This result is further sup- ported by Randler et al. (2017) who found that

the turning point toward morningness is 15.7 years in girls and 17.2 years in boys (Hagenauer et al. 2009).

The finding that nicotine use was associated with eveningness CT was expected since nicotine and other psychostimulants can influence sleep, shortened sleep duration, prolonged sleep latency, and lighter sleep (McNamara et al. 2014; Zhang et al. 2006). Eveningness orientation measured with rMEQ has also previously been associated with nicotine use (Adan 1994). Surprisingly, there were no significant correlation between eveningness and alcohol consumption. Other stu- dies have found a correlation between increased alcohol use and eveningness CT both in adoles- cents and in adults (Adan 1994; Prat and Adan 2011; Urban et al. 2011). In the present study, we cannot interpret if any of the participants had a hazardous alcohol use or if there were any hang- over symptoms caused by alcohol use disorder, since there was only one dichotomized question about weekly alcohol consumption. Furthermore, persons with delayed sleep patterns and late CT are not a homogenous group considering sub- stance use, as have been observed in a study on DSPD (Danielsson et al. 2016).

Very few studies have investigated the validity

of the rMEQ in relation to other measures of

circadian rhythm or sleep patterns. However, the

rMEQ has been compared to the Composite Scale

of Morningness (Caci et al. 2009) and external

validity has been tested with actigraphy monitor-

ing (Natale et al. 2006). Both these studies pre-

sented results pointing toward a sound validity of

the rMEQ. In the present study, five questions

about sleep onset and difficulties rising were used

to validate the measurement of CT with the

rMEQ. The strong correlations observed between

lower rMEQ scores (eveningness) and later sleep

times, difficulties rising in the morning, often

arriving late to work/school, and difficulties in

falling asleep earlier if needed indicate that the

rMEQ has a good validity. Future studies could

also use sleep diaries, melatonin sampling, or core

body temperature measurements for validation of

rMEQ. Furthermore, to analyze rMEQ with

a confirmatory factor analysis could be of interest

to evaluate the factorial validity.

Correlations of rMEQ with depressive and anxiety symptoms, sleep difficulties, and daytime sleepiness

A correlation between “eveningness” and more depressive symptomatology and sleep difficulties was observed in the present study; this finding is in accordance with previous studies (Au and Reece 2017; Hirata et al. 2009; Simor et al. 2014; Tutek et al. 2016). There is growing evidence that depres- sion and circadian disturbances are closely related and may even have a common etiology and com- mon genetic factors (Bielen et al. 2015; Merikanto et al. 2013). Furthermore, short sleep durations can increase the vulnerability to depression (Toomey et al. 2014). That said, depressive symp- toms might result in less daylight because of reduced physical and social activity which in turn leads to poorer sleep. Furthermore, sleep distur- bances and morning tiredness can be symptoms of depression, but can also be a co-occurring phe- nomenon of a sleep disorder (American Academy of Sleep Medicine 2014).

There was no significant relationship between CT and anxiety in the present study. Similar results have been observed both in adolescents and in young adults (Pasch et al. 2010; Selvi et al. 2012; Taylor et al. 2011). Furthermore, eveningness CT has not been observed as a risk factor for any anxiety dis- order (Antypa et al. 2016). Contradictorily, there are some studies that have observed an association between anxiety and eveningness (Alvaro et al.

2014; Lemoine et al. 2013; Pabst et al. 2009; Passos et al. 2017). These studies have, however, investi- gated specific groups, for example, only young girls, insomnia patients, or psychiatric patients. In the present study, we used a random sample from the population which might explain our results.

Daytime sleepiness was not significantly related to total rMEQ scores. Relationships between even- ingness and higher daytime sleepiness have been reported (Taillard et al. 1999; Tutek et al. 2016).

The difference between our own results and those of the previous studies may be explained by the fact that their samples were conducted in adults (the age ranged between 18–43 and 17–80 years, respectively) compared to our study (16–26 years).

Swedish universities and high schools do not always have mandatory presence in whole class

lectures, and there are possibilities to choose eve- ning courses for many students. Therefore, younger adults and adolescents may have the opportunity to compensate late bedtimes with ade- quate sleep in the morning. Adults, on the other hand, generally have more obligations to be awake early in the morning and thus risk to experience inadequate sleep and more daytime sleepiness as a result for adults with an eveningness CT.

Limitations

The present study relied on self-reported measures, which might result in response biases. We were not able to determine any somatic or psychiatric diag- nosis or potential treatments precisely, and these factors could affect sleep. Furthermore, we did not specifically ask the respondents about their under- standing of the questions. To use a think-aloud method might be interesting in future studies of rMEQ. Since this study is cross-sectional, we could not examine test–retest reliability. Choosing a sample with adolescents and young adults was useful since sleep disturbances and delayed sleep–

wake patterns and circadian rhythms are more com- mon in this age-group. However, choosing a specific age group may have influenced the evaluation of the psychometric properties of the rMEQ. We therefore suggest that future studies of the Swedish rMEQ examine validity and test–retest reliability in a large population and in different age groups.

Conclusion

To our knowledge, this was the first study to assess the psychometric properties of the Swedish rMEQ.

The main finding was that the Swedish version of

the rMEQ had a one-factor structure and the rMEQ

score was well correlated to other questions about

sleep–wake patterns. However, the explained var-

iance was relatively low, and the internal consis-

tency was only satisfactory. Eveningness was related

to lower age, male sex, nicotine use, sleep difficul-

ties, and depressive symptoms. Finally, these results

partly question the usefulness of the rMEQ, because

of its reliability. Nevertheless, due to the fair validity

observed in this study, we suggest that the rMEQ

could be preferred to measure CT in large epide-

miological studies or as an initial screening in

preventive or treatment studies, especially when there is a need to reduce the length of the questionnaire.

Declaration of Interest

This was not an industry-supported study. There are no conflicts of interest to declare for any of the authors.

Funding

This work was supported by the Stiftelsen Professor Bror Gadelius Minnesfond.

References

Adan A. 1994. Chronotype and personality factors in the daily consumption of alcohol and psychostimulants. Addiction. 89 (4):455 –62. doi: 10.1111/j.1360-0443.1994.tb00926.x. Cited in PubMed; PMID: 8025504.

Adan A, Almirall H. 1991. Horne & Östberg Morningness- Eveningness Questionnaire: a reduced scale. Personal Individ Differ. 12(3):241 –53.

Adan A, Natale V. 2002. Gender differences in morningness-eveningness preference. Chronobiol Int. 19 (4):709 –20. doi: 10.1081/CBI-120005390. Cited in PubMed PMID: 12182498.

Alvaro PK, Roberts RM, Harris JK. 2014. The independent relationships between insomnia, depression, subtypes of anxiety, and chronotype during adolescence. Sleep Med.

15(8):934 –41. doi: 10.1016/j.sleep.2014.03.019. Cited in PubMed: PMID: 24958244.

American Academy of Sleep Medicine. 2014. International classification of sleep disorders. 3rd ed. Darien (IL):

Diagnostic and coding manual.

Antypa N, Vogelzangs N, Meesters Y, Schoevers R, Penninx BW. 2016. Chronotype assosiations with depres- sion and anxiety disorders in a large cohort study. Depress Anxiety. 33(1):75 –83. doi: 10.1002/da.22422. Cited in PubMed; PMID: 26367018.

Au J, Reece J. 2017. The relationship between chronotype and depressive symptoms: A meta-analysis. J Affect Disord. 15 (218):93 –104. doi: 10.1016/j.jad.2017.04.021. Cited in PubMed; PMID: 28463712.

BaHammam A, Almestehi W, Albatli A, AlShaya S. 2011.

Distribution of chronotypes in a large sample of young adult saudis. Ann Saudi Med. 31(2):183 –86. doi: 10.4103/

0256-4947.78207. Cited in PubMed; PMID: 21422657.

Bastien CH, Vallieres A, Morin CM. 2001. Validation of the Insomnia Severity iIdex as an outcome measure for insom- nia research. Sleep Med. 2(4):297 –307. doi: 10.1016/S1389- 9457(00)00065-4. Cited in PubMed PMID: 11438246.

Bielen J, Melada A, Markeli ć I. 2015. Depression and circa- dian typology. Psychiatr Danub. 27(2):190 –92. Cited in PubMed; PMID: 26057316.

Biswas A, Adan A, Haldar P, Majumder P, Natale V, Randler C, Tonetti L, Sahu S. 2014. Exploration of transcultural proper- ties of the reduced version of the Morningness–Eveningness Questionnaire (rMEQ) using adaptive neuro-fuzzy inference system. Biol Rhythm Res. 45(6):955–68.

Borisenkov MF, Perminova EV, Kosova AL. 2010. Chronotype, sleep length, and school achievement of 11- to 23-year-old students in northern European Russia. Chronobiol Int. 27 (6):1259–70. doi:10.3109/07420528.2010.487624. Cited in PubMed; PMID: 20653453.

Caci H, Deschaux O, Adan A, Natale V. 2009. Comparing three morningness scales: age and gender effects, structure and cut-off criteria. Sleep Med. 10(2):240–45. doi:10.1016/

j.sleep.2008.01.007. Cited in PubMed PMID: 18387342.

Carciofo R, Du F, Song N, Qi Y, Zhang K. 2012. Age-related chronotype differences in Chinese, and reliability assessment of a reduced version of the Chinese morningness-eveningness questionnaire. Sleep Biol Rhythms. 10(4):310–18.

Chelminski I, Petros TV, Plaud JJ, Ferraro FR. 2000.

Psychometric properties of the reduced Horne and Ostberg questionnaire. Personal Individ Differ. 29(3):469–78.

Cohen J. 1988. Statistical power analysis for the behavioral sciences. Hillside (NJ): L Erlbaum Associats.

Danielsson K, Markström A, Broman J-E, von Knorring L, Jansson-Fröjmark M. 2016. Delayed sleep phase disorder in a Swedish cohort of adolescents and young adults:

prevalence and associated factors. Chronobiol Int. 33 (10):1331–39. doi:10.1080/07420528.2016.1217002. Cited in PubMed; PMID: 27537980.

Di Milia L, Adan A, Natale V, Randler C. 2013. Reviewing the psychometric properties of contemporary circadian typology measures. Chronobiol Int. 30(10):1261–71.

doi:10.3109/07420528.2013.817415. Cited in PubMed;

PMID: 24001393.

Hagenauer MH, Perryman JI, Lee TM, Carskadon MA. 2009.

Adolescent changes in the homeostatic and circadian reg- ulation of sleep. Dev Neurosci. 31(4):276–84. doi:10.1159/

000216538. Cited in Pubmed; PMID: 19546564.

Hair JF Jr., Anderson RE, Tatham RL, Black WC. 1998.

Multivariate data analysis. 5th ed. Upper Saddle River (New Jersey): Prentice.

Hirata FC, Lima MC, de Bruin VM, Nóbrega PR, Wenceslau GP, de Bruin PF. 2009. Depression in medical school: the influ- ence of morningness-eveningness. Chronobiol Int. 24 (5):939–46. doi:10.1080/07420520701657730. Cited in PubMed; PMID: 17994347.

Horne JA, Ostberg O. 1976. A self-assessment questionnaire to determine morningness-eveningness in human circa- dian rhythms. Int J Chronobiol. 4(2):97–110. Cited in Pubmed; PMID: 1027738.

Jankowski KS. 2013. Polish version of the reduced Morningness–Eveningness Questionnaire. Biol Rhythm Res.

44:427–33.

Johns MW. 1991. A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep. 14 (6):540–45. doi:10.1093/sleep/14.6.540. Cited in PubMed;

PMID: 1798888.

Kim S, Dueker GL, Hasher L, Goldstein D. 2002. Children’s time of day preference: age, gender and ethnic differences.

Personal Individ Differ. 33(7):1083–90.

Kuhn C. 2015. Emergence of sex differences in the development of substance use and abuse during adolescence. Pharmacol Ther. 153:55–78. doi:10.1016/j.pharmthera.2015.06.003.

Cited in PubMed; PMID: 26049025.

Lee JH, Kim SJ, Lee SY, Jang KH, Kim IS, Duffy JF. 2014.

Reliability and validity of the Korean version of morning- ness-eveningness questionnaire in adults aged 20–39 years.

Chronobiol Int. 31(4):479–86. doi:10.3109/07420528.

2013.867864. Cited in PubMed; PMID: 24467306.

Lemoine P, Zawieja P, Ohayon MM. 2013. Associations between morningness/eveningness and psychopathology:

an epidemiological survey in three in-patient psychiatric clinics. J Psychiatr Res. 47(8):1095–98. doi:10.1016/j.jpsy- chires.2013.04.001. Cited in PubMed; PMID: 23628386.

Loureiro F, Garcia-Marques T. 2015. Morning or evening person? Which type are you? Self-assessment of chronotype. Personal Individ Differ. 86:168–71.

McNamara JP, Wang J, Holiday DB, Warren JY, Paradoa M, Balkhi AM, Fernandez-Baca J, McCrae CS. 2014. Sleep dis- turbances associated with cigarette smoking. Psychol Health Med. 19(4):410–19. doi:10.1080/13548506.2013.832782.

Cited in PubMed; PMID: 24040938.

Merikanto I, Lahti T, Kronholm E, Peltonen M, Laatikainen T, Vartiainen E, Salomaa V, Partonen T.

2013. Evening types are prone to depression. Chronobiol Int. 30(5):719–25. doi:10.3109/07420528.2013.784770.

Cited in PubMed; PMID: 23688117.

Natale V, Esposito MJ, Martoni M, Fabbri M. 2006. Validity of the reduced version of the Morningness-Eveningness Questionnaire. Sleep Biol Rhythms. 4(1):72–74.

O’Connor BP. 2000. SPSS and SAS programs for determining the number of components using parallel analysis and Velicer’s MAP test. Behav Res Methods Instrum Comput.

32(3):396–402. doi:10.3758/BF03200807. Cited in PubMed;

PMID: 11029811.

Ostberg O. 1973. Interindividual differences in circadian fatique patterns of shift workers. Br J Ind Med. 30 (4):341–51. Cited in PubMed PMID:4753717.

Pabst SR, Negriff S, Dorn LD, Susman EJ, Huang B. 2009.

Depression and anxiety in adolescent females: the impact of sleep preference and body mass index. J Adolesc Health.

44(6):554–60. doi:10.1016/j.jadohealth.2008.11.012. Cited in PubMed; PMID: 19465319.

Paine SJ, Fink J, Gander PH, Warman GR. 2014. Identifying advanced and delayed sleep phase disorders in the general population: a national survey of the New Zealand adults.

Chronobiol Int. 31(5):627–36. doi:10.3109/07420528.

2014.885036. Cited in PubMed; PMID: 24548144.

Pasch KE, Laska MN, Lytle LA, Moe SG. 2010. Adolescent sleep, risk behaviors, and depressive symptoms: are they linked? Am J Health Behav. 34(2):237–48. doi:10.5993/

AJHB.34.2.11. Cited in PubMed; PMID: 19814603.

Passos GS, Santana MG, Poyares D, D’Aurea CV, Teixeira AA, Tufik S, de Mello MT. 2017. Chronotype and anxiety are associated in patients with chronic pri- mary insomnia. Rev Bras Psiquiatr. 39(2):183–86.

doi:10.1590/1516-4446-2016-2007. Cited in PubMed;

PMID: 28076650.

Prat G, Adan A. 2011. Influence of circadian typology on drug consumption, hazardous alcohol use, and hangover symptoms. Chronobiol Int. 28(3):248–57. doi:10.3109/

07420528.2011.553018. Cited in PubMed; PMID: 21452920.

Rahafar A, Sadeghi JM, Sadeghpour A, Heidari Z, Kasaeian A. 2015. Psychometric properties of the persian version of the reduced Morningness-Eveningness Questionnaire: further evidence. Sleep Biol Rhythms. 13 (2):112–16.

Randler C. 2008. Differences in sleep and circadian preference between Eastern and Western German adolescents.

Chronobiol Int. 25(4):565–75. doi:10.1080/07420520802 257794. Cited in PubMed; PMID: 18622816.

Randler C. 2013. German version of the reduced Morningness?Eveningness Questionnaire (rMEQ). Biol Rhythm Res. 44(5):730–36.

Randler C, Freyth-Weber K, Rahafar A, Florez Jurado A, Kriegs JO. 2016. Morningness-eveningness in a large sam- ple of German adolescents and adults. Heliyon. 2(11):

e00200. doi:10.1016/j.heliyon.2016.e00200. Cited in PubMed; PMID: 27957554.

Raoof AM, Asaad YA, Al-Hadithi TS. 2014. Distribution of chronotypes among a sample of Iraqi Kurdish medical students. Sultan Qaboos Univ Med J. 14(3):356–60. Cited in PubMed; PMID: 25097771.

Raykov T. 1997. Estimation of composite reliability for con- generic measures. Appl Psychol Meas. 21(2):173–84.

Reid KJ, Jaksa AA, Eisengart JB, Baron KG, Lu B, Kane P, Kang J, Zee PC. 2012. Systematic evaluation of axis-I DSM diagnoses in delayed sleep phase disorder and evening-type circadian preference. Sleep Med. 13 (9):1171–77. doi:10.1016/j.sleep.2012.06.024. Cited in PubMed; PMID: 22910327.

Roenneberg T, Kuehnle T, Pramstaller PP, Ricken J, Havel M, Guth A, Merrow M. 2004. A marker for the end of adolescence. Curr Biol. 14(24):R1038–9.

Roveda E, Vitale J, Montaruli A, Galasso L, Carandente F, Caumo A. 2017. Predicting the actigraphy-based acrophase using the Morningness-Eveningness Questionnaire (MEQ) in college students of North Italy. Chronobiol Int. 34 (5):551–62. doi:10.1080/07420528.2016.1276928. Cited in PubMed; PMID: 28276851.

Selvi Y, Aydin A, Gulec M, Boysan M, Besiroglu L, Ozdemir PG, Kilic S. 2012. Comparison of dream anxiety and subjective sleep quality between chronotypes. Sleep Biol Rhythms. 10(1):14–22.

Simor P, Zavecz Z, Pálosi V, Török C, Köteles F. 2014. The

influence of sleep complaints on the association between

chronotype and negative emotionality in young adults.

Chronobiol Int. 32(1):1–10. doi:10.3109/07420528.2014.

935786. Cited in PubMed; PMID: 25003651.

Ståleby M, Steingrimsson S, Ioannou M. 2016. Swedish trans- lation and reliability of the Morningness-Eveningness Questionnaire. Eur Psychiatry. 33(Supplement):S524.

Taillard J, Philip P, Bioulac B. 1999. Morningness/evening- ness and the need for sleep. J Sleep Res. 8(4):291–95. Cited in PubMed; PMID: 10646169.

Taylor DJ, Clay KC, Bramoweth AD, Sethi K, Roane BM. 2011.

Circadian phase preference in college students: relationships with psychological functioning and academics. Chronobiol Int. 28(6):541–47. doi:10.3109/07420528.2011.580870. Cited in PubMed; PMID: 21797783.

Tonetti L, Sahu S, Natale V. 2012. Cross-national survey of winter and summer patterns of mood seasonality:

a comparison between Italy and India. Compr Psychiatry.

53(6):837–42. doi:10.1016/j.comppsych.2011.11.010. Cited in PubMed; PMID: 22225787.

Toomey R, Panizzon MS, Kremen WS, Franz CE, Lyons MJ.

2014. A twin-study of genetic contributions to morning- ness?eveningness and depression. Chronobiol Int. 32 (3):303–09. doi:10.3109/07420528.2014.971366. Cited in PubMed; PMID: 25347156.

Tutek J, Emert SE, Dautovich ND, Lichstein KL. 2016.

Association between chronotype and nonrestorative sleep in a college population. Chronobiol Int. 33(9):1293–304.

doi:10.1080/07420528.2016.1212870. Cited in PubMed;

PMID: 27715320.

Urban R, Magyarodi T, Rigo A. 2011. Morningness- eveningness, chronotypes and health-impairing behaviors in adolescents. Chronobiol Int. 28(3):238–47. doi:10.3109/

07420528.2010.549599. Cited in Pubmed; PMID: 21452919.

Weaver DR. 2002. Coordination of circadian timing mammals. Nature. 418(6901):935–41. doi:10.1038/nat- ure00965. Cited in PubMed; PMID: 12198538.

Willoughby T. 2018. A short-term longitudinal study of internet and computer game use by adolescent boys and girls: prevalence, frequency of use, and psychosocial predictors. J Behav Addict. 7(2):454–65. doi:10.1556/

2006.7.2018.28. Cited in PubMed; PMID: 18194017.

Young MW. Time travels: A 40-year journey from Drosophila’s clock mutants to human circadian disorders (nobel lecture). Angew Chem Int Ed Engl. 2018 Jul 13.

doi:10.1002/anie.201803337. Epub ahead of print.

Zhang L, Samet J, Caffo B, Punjabi NM. 2006. Cigarette smoking and nocturnal sleep architecture. Am J Epidemiol. 164(6):529–37. doi:10.1093/aje/kwj231. Cited in PubMed: PMID: 16829553.

Zigmond AS, Snaith RP. 1983. The hospital anxiety and depression scale. Acta Psychiatr Scand. 67(6):361–70.

doi:10.1111/j.1600-0447.1983.tb09716.x. Cited in PubMed;

PMID: 6880820.