Morning individuals in Saudi Arabia have higher self-regulation of eating behavior compared to evening types

ABSTRACT Chronotype is a manifestation of an individual’s circadian system, which can be presented as a preference for morningness or eveningness. Evening chronotype has been linked to unhealthy dietary habits. These habits are partially from poor self-regulation of eating behavior (SREB). SREB is a goal-directed process that helps resist food temptation and achieve desired dietary outcomes. It is unclear whether chronotype is associated with SREB. This study aims to investigate the association between chronotypes and SREB. A total of 599 adults (18–50 years) enrolled in this cross-sectional study. Participants received an online questionnaire that collected sociodemographic, anthropometrics, chronotype through the validated reduced Morningness-Eveningness Questionnaire (rMEQ) and SREB using the SREB Questionnaire. ANOVA, Chi-square and multivariate logistic regression were used. In adjusted models, morning types had 85% higher SREB compared to evening types (p = 0.003, OR = 1.85). Evening types found fried foods (p = 0.003), chips (p = 0.005) and French fries (p = 0.018) more tempting than morning types. These findings show a link between chronotype and SREB. Strategies focusing on SREB may help evening types improve dietary habits. Further studies are required to clarify the underlying mechanisms for these associations. This may have important implications for lifestyle and behavioral change policy.


Introduction
Eating and sleeping are fundamental behaviors of human life, and they are both affected by circadian rhythms. Circadian rhythms are the internal process that regulates physiologic and behavioral cycles every 24 h including the sleep-wake cycle according to the light-dark cycle (day and night) (Kawato et al. 1982). Circadian rhythms vary between individuals and this variation allows for disparities in sleep-wake preferences, ranging from extreme morning types to extreme evening types, referred to chronotype (Horne and Ostberg 1976).
Evening chronotypes have been linked with many health complications (Lotti et al. 2022), including poor glycemic control (Kelly et al. 2020), metabolic dysfunction, a higher rate of cardiovascular risk (Patterson et al. 2018), psychiatric disorders (Kivelä et al. 2018) and other common chronic diseases (Basnet et al. 2017;Lotti et al. 2022). Consequently, evening types have higher risks of morbidity and mortality compared to morning types (Knutson and von Schantz 2018). The increase in mortality risk may be due to the chronic circadian misalignment, which is a mismatch between internal biological clocks and external timing demands of work and social activity (i.e., evening type who are obligated to wake and work in the morning) (Baron and Reid 2014). Another possible reason may be attributed to the unhealthy behaviors shown in evening types compared to morning types including sedentary lifestyle, smoking (Patterson et al. 2016;Makarem et al. 2020), poor sleep quality (Silva et al. 2020) and poor dietary habits (Bodur et al. 2021;Kawasaki et al. 2021;Zuraikat et al. 2021).
The associations between chronotype, nutrients and dietary habits have been widely studied. Evening types are more prone to meal skipping (Mazri et al. 2020), lower intakes of fruits and vegetables (Noorwali et al. 2019), high sugary drinks (Li et al. 2018) high alcohol intake and difficulties controlling portion sizes (Vera et al. 2018). This leads us to question whether evening chronotypes have poor self-regulation of eating behavior, leading them to adopt unhealthy dietary patterns. Self-regulation refers to the management process of behavior, cognition, feelings, thoughts and environment to achieve goal-directed action (Kliemann et al. 2016). In recent years, variation between individuals' vulnerability to the obesogenic environment has been well observed. It is suggested that the ability of an individual to selfregulate and resist food temptation may play a role in this variation and consequently have a healthy weight and diet. Several studies have shown that the ability to self-regulate eating behavior diminishes the intentionbehavior disparity and, therefore, helps to achieve desired dietary goals (Annesi et al. 2016;Greaves et al. 2017;Teixeira et al. 2015).
These lines of evidence provide a foundation to hypothesize that evening chronotypes may be unable to self-regulate eating behaviors. In our modern societies, it is difficult to control environmental timing demands (i.e., working hours and social activities) to achieve positive dietary habits. However, self-regulatory skills may help achieve healthy dietary habits. To study this hypothesis, it is critical to study the associations between different chronotypes and the ability to resist appealing food in relation to body weight. To our knowledge, no previous study investigated the relationship between chronotype and self-regulatory eating behavior. Therefore, the aim of this study is to assess the associations between chronotype and self-regulation of eating behavior in Saudi adults.

Study design and participants
This cross-sectional study was conducted between September and December 2021 using an anonymous electronic questionnaire. Exclusion criteria: participants younger than 18 years, pregnant, breastfeeding, having chronic illnesses, using any medication or those with diagnosed psychiatric conditions.
The questionnaire was distributed in English and Arabic and shared via Google form link through social media platforms including Twitter®, WhatsApp®, Telegram® and email. The two researchers independently translated the original version of the questionnaire from English to Arabic. Prior to use in this study, the questionnaire was initially administered to 10 bilingual subjects, who completed both the Arabic version and the English version to determine the test-retest reliability. Circadian typology and SREB were independent of the version. An average time of 3 ± 2 min was needed to complete the questionnaire. The first page of the questionnaire clearly described the aims and objectives of the study, voluntary nature of participation, declaration of anonymity and confidentiality and obtained informed consent. Once the participant agreed to participate, the questionnaire can commence to the next sections.
The questionnaire consisted of 24 questions including demographic data, chronotype and self-regulation of eating behavior data, respectively. Demographic data included age, sex (male/female), educational level (never attended school, primary or intermediate school, high school, diploma or equivalent graduate, bachelor's degree, master's or doctorate), employment status (employed, not employed, self-employed, student or retired) and shiftwork (yes/no). Height and weight were also self-reported. All participants gave their consent as a prerequisite to enrolment. Ethical approval was obtained from the Standing Committee for Scientific Research Ethics, Umm Al-Qura University. Registration No. HAPO-02-K-012.

Independent variable: chronotype
The Morningness-Eveningness Questionnaire (MEQ) is considered the gold standard in measuring chronotype (Danielsson et al. 2019). The MEQ is a self-assessment questionnaire developed in 1976 by Horne and Östberg and has been intensively validated and widely used in research (Horne and Ostberg 1976). However, due to its length, a reduced 5-item version (rMEQ) was used in this study to increase participants' compliance. The rMEQ has proven to be reliable with good psychometric properties and convergent validity (Adan and Almirall 1991). The reduced questionnaire contains five questions with Likert-type responses and yields scores of 4 to 25. In questions 1 through 3, participants are asked about their preferred waking time and their subjective alertness throughout the day and night. Question 4 asks participants to indicate what time in the day they can perform optimally. Finally, Question 5 is a direct question considering their morningness or eveningness preferences. The final score of the five items ranges from 4 (extreme eveningness) to 25 (extreme morningness), resulting in 5 chronotype classifications: 'definitely morning' type, 'moderately morning' type, 'neither (intermediate)' type, 'moderately evening' type and 'definitely evening' type (Adan and Almirall 1991). Consistent with previous studies, the groups were simplified and coded as morning type, neither (intermediate) type and evening type to avoid having categories with small sample sizes (Adan et al. 2010;Culnan et al. 2013;Sun et al. 2020).

Self-regulation of eating behavior
The Self-regulation of Eating Behavior Questionnaire (SREBQ) is used to assess an individual's eating selfregulatory skills. SREBQ has been validated and demonstrated good reliability in serving its purpose (Kliemann et al. 2016). It is a five-item questionnaire with response options ranging from 1 (never) to 5 (always). To serve its purpose, the questionnaire begins with a list of 13 commonly craving foods preceded by the question: "do you find any of the following food tempting (that is, do you want to eat more of them than you think you should)?" Participants can tick all foods they find tempting. The next two questions ask if they intend to control their consumption of these tempting foods and if they have the intention to follow a healthy diet. These questions are designated to exclude participants who do not have healthy dietary intentions by any chance from the SREBQ. Therefore, only participants who have healthy dietary intentions are eligible to answer the SREBQ and their level of self-regulatory skills in maintaining their intentions was calculated. Total mean score cut-offs: < 2.8 indicates low, 2.8 to 3.6 indicates moderate and > 3.6 indicates high self-regulation of eating behavior (Kliemann et al. 2016).

Statistical analyses
Data analysis was performed using the Statistical Package for the Social Sciences (SPSS) 23rd version. Frequency and percentages were used to display categorical variables. Mean and standard deviation were used to present numerical variables. The Chi-square test was used to test for the presence of an association between categorical variables. ANOVA test was also used to test for associations. Multivariate logistic regression was used to determine the factors predicting high self-regulation of eating behavior. Confounders included in the logistic regression model were selected based on previous studies (Patterson et al. 2016(Patterson et al. , 2018. The logistic regression model included the following variables: age, sex, education, shift work, use of vitamins and chronotype. Use of vitamins was included in the model due to previous research showing effects of intake of some vitamins on eating behavior (Schroeter et al. 2013) and circadian rhythms (Djokic et al. 2019;Feeney et al. 2016;Mayer et al. 1996).
Shift work influences the sleep-wake cycle and chronotype (Hittle and Gillespie 2019); therefore, shift workers (n = 75) were excluded in the chi-square test. However, in the logistic regression, shift work was used as one of the confounders. The SREB eligibility includes the participants' intention to have a healthy diet; therefore, participants who do not have any healthy dietary intentions in the first place (n = 39) were tested for their association with chronotype separately. The level of significance was set at 0.05.

Socio-Demographic characteristics
A total of 599 participants were included in this study. Table 1 shows the sociodemographic profile of the participants. Among the participants, 324 (54.1%) were between 18 and 24 years, 151 (25.2%) were between 30 and 40 years, 91 (15.2%) were between 41 and 50 years, and 33 (5.5%) were between 50 and 60 years. Most of the participants were females (n = 474; 79.1%), and the mean age was 31 years (SD ± 10.37). Ninety-two percent of the participants were Saudis and 65% of the participants (n = 394) lived in the Western region. As for BMI, almost half of our sample had a normal BMI (n = 263; 43.9%) and about a quarter were overweight (n = 173; 28.9%) and the rest of the participants were obese (n = 113; 18.8%) ( Table 1).

Chronotype and associated factors
Participants' chronotype based on the reduced morningness-eveningness questionnaire was evening chronotype (n = 145; 24.2%) and morning chronotype (n = 186; 31.1%), whereas the majority (n = 268; 44.7%) were intermediate (i.e., neither morning nor evening chronotype). Table 2 shows the associations between BMI and SREB based on chronotype. The mean BMI for morning type was 25.55 + 6.11 kg/m 2 and evening type (26.2 + 7.74 kg/m 2 ). BMI was not significantly associated with chronotype (Table 2). Having healthy dietary intentions and SREB did not differ between chronotypes. Age, sex, education and occupation were also not significantly associated with chronotype.

Self-regulation of Eating Behavior Questionnaire (SREBQ)
Regarding eligibility to answer the questionnaire, the vast majority have healthy dietary intentions (n = 560; 93.5%), and, therefore, their SREBQ scores were valid, while (n = 39; 6.5%) did not have healthy dietary intentions. Based on the participants' SREBQ score, the participants had low (n = 78; 13.9%), moderate (n = 361; 64.5%) and high (n = 12; 21.6%) self-regulation of eating behavior. Participants' responses to the Self-regulation of Eating Behavior Questionnaire (SREBQ) are in Table 3. Figure 1 demonstrates the participants' responses to the question "Do You Find any of the Following Food Tempting?" (That is, do you want to eat more of them than you think you should), (n = 61; 10.2%) reported that they do not find any food tempting. The most reported tempting foods were chocolate (n = 284; 47.4%), French fries (n = 234; 39.1%), pastries and bakeries (n = 229; 38.2%), pizza (n = 217; 36.2%) and ice cream (n = 200; 33.4%). Table 4 demonstrates the association between foods that participants find to be tempting and chronotype. Fried foods (p = 0.003), chips (p = 0.005) and French fries (p = 0.018) were all significantly associated with Shift workers were excluded. ANOVA and Chi-square tests were used. Self-regulation of eating behavior cut-offs: low < 2.8, medium 2.8 to 3.6 and high > 3.6. chronotype. Where it was observed that participants with evening chronotype had a notably higher rate of finding these foods tempting and difficult to resist compared to participants with morning or intermediate chronotype. Moreover, participants with the morning chronotype had a notable lowest rate of finding these foods tempting compared to other chronotypes. Table 4 demonstrates the association between foods the participants find to be tempting according to sex. Females were observed to find the following food tempting significantly higher than males did: chips (p = 0.001), cake (p = 0.018), ice cream (p = 0.003), popcorn (p = 0.008), bread/toast (p = 0.002) and pastries/bakeries (p < 0.001), while males were observed to find soft drinks tempting significantly higher than females did (p = 0.008). Table S1 demonstrates the association between foods the participants find to be tempting according to BMI categories. There was no significant association between BMI categories and any of the food the participants found to be tempting. Table S2 presents the association between foods the participants find to be tempting and occupation. There was a significant association between occupation and the following food in finding them to be tempting: soft drinks (p = 0.01), chips (p < 0.001), sweets (p = 0.018), French fries (p = 0.024), ice cream (p = 0.011) and pastries/bakeries (p = 0.017). It was observed that the highest group who found soft drinks and French fries to be tempting was those who were unemployed. It was also observed that the highest group who found chips, sweets and pastries/bakeries to be tempting was those who were students. Moreover, retired participants were the group with the highest rate of finding ice cream to be tempting.

Chronotype, sex, BMI and SREB
Multivariate logistic regression was conducted to determine the factors predicting high self-regulation  Figure 1. Participants' responses to the question "do you find any of the following food tempting (That is, do you want to eat more of them than you think you should)?" of eating behavior (Table 5). Morning chronotype was the only factor significantly predicting higher self-regulation of eating behavior (p = 0.003, OR = 1.85 (CI 1.05, 3.25) meaning that they are 85% more likely to have high self-regulation of eating behavior compared to those with evening chronotype.

Discussion
To our knowledge, this is the first study to examine the association between chronotype and SREB in Saudi adults. In line with our hypothesis, we found that circadian preference towards morningness significantly predicts high self-regulation of eating behavior compared to  Multivariate Logistic Regression analyses conducted. Self-regulation of eating behavior (SREB), where low SREB indicates a score ≤ 3.6 and high SREB indicates a score > 3.6. *Significant at level 0.05.
eveningness. Moreover, evening types found several unhealthy foods tempting and difficult to resist, specifically, chips, fried foods and French fries, whereas morning types had the lowest rate of finding these foods tempting among all chronotypes. No significant evidence was found between chronotype, BMI and other factors in our study. Morning types had significantly higher eating selfregulatory skills by 85% compared to evening types in our study. This might be consistent with the differences found in personality traits between chronotypes; morning types tend to be more conscientious and have higher self-control (Digdon and Howell 2008;Tonetti et al. 2009).
A recent systematic review showed that morning chronotypes had healthy and protective dietary habits, such as eating early and eating predominantly fresh/ minimally processed foods. On the other hand, late chronotypes most likely ate late at night, often skipped breakfast and consumed processed foods (Teixeira et al. 2022). Furthermore, 49% of the studies in the systematic review found that late chronotypes had higher associations with obesity compared to morning types. One of the underlying reasons for these associations may be related to self-regulation eating skills. Some studies concluded that poor SREB is one of the strongest predictors of overweight and obesity (Balani et al. 2019;Leong et al. 2012). This may be explained by the high SREB which has been associated with higher intakes of fruits and vegetables and a healthy diet (Kliemann et al. 2018).
In our study, the mean BMI was relatively higher among evening types; however, no significant association was detected between chronotype and BMI and the association between foods the participants find to be tempting according to BMI categories was not significant. This may be due to the assessment of anthropometrics which were self-reported and may result in over/ under reporting. This is consistent with previous findings suggesting no association between chronotype and BMI. Several cross-sectional studies, including a largescale Finnish study (n = 4421) (Maukonen et al. 2016) and a Norwegian population-based study (n = 6413) (Johnsen et al. 2013), found no association between BMI and chronotype. A possible explanation is shown by a review of 36 studies that concluded that evening types are mostly engaged in negative dietary behaviors; however, both chronotypes consume almost the same total energy intake (Mazri et al. 2020). In contrast, a follow-up study (n = 54) found that evening types experienced a significant BMI increment after 8 weeks of follow-up, although there was no association detected at the baseline (Culnan et al. 2013). A 7-year follow-up study also found evening types tend to increase in their anthropometrics with time, particularly among eveningtyped women (Maukonen et al. 2019). This contradiction might be due to the cross-sectional nature capturing weight parameter at one time point.
The present study showed that evening types tend to find some foods more tempting and difficult to resist compared to morning types, specifically chips, French Fries and fried foods. These results are in line with Kanerva et al. (2012) who found that eveningness was associated with the tendency to ingest more fat and other negative dietary choices (Kanerva et al. 2012). One possible explanation is food timing. Later, intake of all three meals has been consistently observed among evening chronotypes (Mazri et al. 2020;Vera et al. 2018). Epidemiological studies have found that consuming breakfast early is associated with high fiber intake, whereas at night, people tend to consume fatty foods, particularly foods with high saturated fat (Khare and Inman 2006;Mazri et al. 2020;Roßbach et al. 2018). Another possible explanation for fried food temptation in evening chronotypes is the norms and preconceived ideas of certain foods associated with the time of the day that they should be consumed in (e.g., breakfast cereal in the morning vs. French fries at night). Furthermore, the availability and accessibility of fatty foods during latenight hours (opening hours of fast-food restaurants) in Saudi Arabia may be another reason for the craving of fried foods in evening types. A final possible explanation for the increased temptation for fried food in evening types is that healthier food choices in the morning are easier than in the evening because self-regulation tends to decrease later in the day (Boland et al. 2013). More representative studies are needed to assess the associations between chronotype, SREB, BMI and the underlying reasons for food temptation in evening types.
In this study, females were observed to find the following food tempting significantly higher than males did: chips, cake, ice cream, popcorn, bread/toast and pastries/bakeries, while males were observed to find soft drinks tempting significantly higher than females did. These results are supported by previous research which showed that emotional eating had a stronger negative association with intuitive eating in women than in men (Smith et al. 2020). Another possible explanation is that women have more frequent cravings for chocolate and a greater difficulty to resist food cravings than men, and the beneficial effects of exercise on appetite control were observed in men and not in women (Drenowatz et al. 2017). Although not assessed in this study, women tend to have higher food cravings of chocolate, pastries, snacks and desserts during the premenstrual period (De Souza et al. 2018;Gorczyca et al. 2016;Hintze et al. 2017).
Our further analysis (Table S2) showed that that the highest group who found soft drinks and French fries to be tempting was those who were unemployed. It was also observed that the highest group who found chips, sweets and pastries/bakeries to be tempting was those who were students. Moreover, retired participants were the group with the highest rate of finding ice cream to be tempting. These differences may be due to the variation in income levels, awareness and knowledge of healthy food and preparation skills. Students may have limited time in meal preparation due to studying and stress that may impact eating behaviors (Deliens et al. 2014;Kato et al. 2019). Future studies are needed in representative samples to study the underlying mechanisms of this association.

Implications
Based on our results, evening types have poorer SREB compared to morning types. Furthermore, evening types find fried food more tempting and difficult to resist compared to morning types. Consequently, supported by previous studies, evening-type individuals may be more prone to unhealthy dietary choices. These unhealthy dietary choices eventually result in negative health outcomes, regardless of obesity (Jayedi et al. 2020). Therefore, registered dietitians and health care practitioners may assess chronotype, chrono-nutrition (circadian timing of food intake) (Veronda et al. 2020) and SREB and provide personalized interventions focusing on eating self-regulatory skills that may impact cardiometabolic health such as improvements of glucose and lipid metabolism and blood pressure consequently this may decrease the risk of type 2 diabetes and cardiovascular disease (Almoosawi et al. 2019;Lotti et al. 2022). The current obesogenic environment is a major public health problem in need of practical solutions. Dietary self-regulation could help individuals cope with this environment, and fortunately, SREB are durable skills that can be enhanced through training (Kliemann et al. 2016). Interventions promoting dietary self-regulation training, particularly among those who are more vulnerable (e.g., evening types, shift workers, university students and individuals with eating disorders) may help those individuals to resist food temptation and sustain healthy dietary habits despite environmental cues. Interventions could also include identifying potential barriers for enhancing SREB in evening types (e.g., only fast-food options are available at late night) and creating action plans to resolve these barriers (e.g., making healthy food options available and easily accessible at any time). These interventions help in understanding the relationship between food environment, policy and SREB that has been highlighted in a review (Schwartz et al. 2017). Multiple environmental influences on SREB such as pricing strategies, food marketing and social norms are important to study in future studies. These environmental influences may be controlled by policies because they are the most distal influence on SREB of individuals (Schwartz et al. 2017).

Strengths and limitations
Our main strength is the novelty of this study; to our knowledge, this is the first study to assess the associations between chronotype and SREB in Saudi adults. Furthermore, this study includes the use of rMEQ in assessing chronotype, which has been validated and widely used in the field. We also used a relatively new and validated tool to assess the self-regulation of eating behavior (Kliemann et al. 2016). In addition, we were able to consider multiple potential confounders in our analyses, including age, sex, education, vitamin intake and shift work.
Limitations: First, the cross-sectional design cannot reveal causality. Second, participants were recruited through social media and were mostly college graduates and females who are not a representative sample. In addition, recruitment through social media may result in selection bias with only those having access to social media platforms participating. Data was collected based on online self-reports; thus, possible reporting bias may affect the data to some extent. Furthermore, healthconscious people are more likely to participate in health surveys and that may have affected the present findings (Jousilahti et al. 2005). This may limit the generalizability of the study findings to other populations. Finally, besides BMI, this study was unable to explore other aspects of adiposity in chronotypes such as waist circumference, waist-to-hip ratio and body fat percentage.

Conclusion
Our study found that chronotype predicted SREB with morning types having higher SREB compared to evening types in Saudi adults. Evening types found fried foods more tempting and difficult to resist compared to morning types. Further investigations and prospective studies are required to assess the relationship between chronotype and SREB in representative samples and other populations. The findings of this study did not find an association between chronotype and BMI. The relationship between chronotype, SREB and obesity requires further studying. Finally, adopting strategies (such as meal planning, food intake monitoring, healthy snacks, reducing stress and anxiety and exercising regularly) to enhance eating self-regulation may be beneficial for maintaining long-term healthful dietary goals, especially in evening types. Further studies are needed to examine the effects of such strategies.