Effects of Sleep and Social Jetlag on the Quality of Diet of Evening Graduate Students
Article information
Abstract
Background and Objective
Chronotype, sleep, and diet are interconnected. Social demands, such as work or school, often hinder alignment with biological rhythms, potentially harming dietary habits and health. This study investigated the association between chronotype and social jetlag in undergraduate students and their impact on diet.
Methods
Sixty-three young adults (40 females), who studied at night and worked during the week, participated. Chronotype, social jetlag, sleep quality, and daytime sleepiness were assessed through validated questionnaires. A 1-week sleep and food diary evaluated sleep and dietary intake.
Results
Most students were classified as good sleepers, despite excessive daytime sleepiness. The majority had an intermediate chronotype and social jetlag exceeding 1 hour (median: 1.57 h; range: 0.05–4.02 h). On free days, participants woke later, slept longer, and had later sleep midpoints compared to workdays (p<0.001). Social jetlag was negatively correlated with chronotype (p=0.009), but not with sleep quality, efficiency, or daytime sleepiness. Negative correlations were observed between protein, lipid, and energy intake and sleep latency on day 1. On day 4, higher carbohydrate, lipid, and energy intake were associated with later bedtimes, and lipid and energy intake were negatively correlated with sleep latency. Social jetlag was also negatively correlated with weekly average intake of carbohydrates (p=0.05), protein (p=0.03), and energy (p=0.03).
Conclusions
Social demands affected biological rhythms, leading to social jetlag exceeding one hour in most participants, especially in evening chronotypes. While no clear links were found with sleep quality, observed diet–sleep correlations suggest dietary intake may influence sleep timing and latency.
INTRODUCTION
The human body has a vital biological clock called the suprachiasmatic nucleus, which regulates various biological functions that occur on a 24-hour cycle, known as circadian rhythms. Endocrine and behavioral factors both influence the circadian rhythm and are related to an individual’s sleep [1,2].
The misalignment between an individual’s natural circadian rhythm and their actual sleep pattern due to everyday responsibilities is called “social jetlag” [2–4]. The literature suggests that individuals with a social jetlag of over an hour are at a higher risk of developing obesity and metabolic conditions [5,6]; therefore, this should be avoided. Higher social jetlag is also linked to greater daytime sleepiness, increased irritability [7] and reduced work and academic outcomes [6].
Research on this topic is still relatively sparse, but it appears that social jetlag is linked to an individual’s lifestyle. Reports suggest that individuals with unhealthy habits, such as lack of physical activity, tobacco use, and poor food choices, tend to have higher social jetlag values [2,5,6]. Undergraduate students are often exposed to these inadequate habits, especially irregular sleep timing and sleep deprivation [8–10] and are the target population of many studies investigating the relationship between sleep, social jetlag, and eating behaviour [11–14].
The literature describes that both chronotype and social jetlag are linked with food intake, with growing evidence indicating that the timing and composition of meals are associated with the sleep pattern [11–15]. Higher social jetlag appears to be related with the time [13,14], type and quantity of certain food groups consumed, with a higher tendency toward inadequate dietary patterns [12].
One of the behavioural factors that influences circadian rhythms and sleep in students is the schedule imposed by schools or universities [8,10]. Additionally, individuals who both work and study may have their sleep patterns further affected by the combined demands of these social obligations. Excessive sleepiness [16], greater physical fatigue, and reduced academic performance [17] are some of the consequences of sleep deprivation among working evening students, but more studies are necessary to understand the sleep in that population.
Considering that social demands such as work and study can disrupt sleep patterns and that these changes can have negative consequences on health, the importance of better understanding sleep and its relationship with people’s diet becomes increasingly evident. Thus, the primary goal of this study was to investigate how sleep behaviors and social jetlag relate to nutrient consumption in evening undergraduate students.
METHODS
Participants and Eligibility Criteria
This is a quantitative and qualitative observational study. The study was submitted and approved by the Research Ethics Committee of the Universidade Estadual de Campinas (n. 5.888.395). Participants signed an Informed Consent Form authorizing the use of their data for the research, and personal information that could lead to the volunteer’s identification was not disclosed.
Young people of both sexes, between 18 and 35 years old, were recruited. As an inclusion criterion, volunteers were required to be evening undergraduate students, work during the week, and have free wake-up and bedtimes on weekends. Recruitment was conducted through the dissemination of the study via social media and online college groups.
Experimental Procedures
All individuals who agreed to participate received online questionnaires. The first was an anamnesis form, containing general questions about health, mobility, housing, caffeine consumption, and other factors that could influence sleep quality or lifestyle.
In addition to anamnesis, they also answered some sleep questionnaires. The Morningness-Eveningness Questionnaire (MEQ) was used to assess chronotype [18], and the version validated with the Brazilian population was applied [19]. Participants were classified as afternoon people (16 to 41), indifferent people (42 to 58) or morning people (59 to 86) [18].
The Munich Chronotype Questionnaire (MCTQ), which asks about participants’ sleep on work/study days and on free days, was applied to evaluate social jetlag [20]. The social jetlag was calculated using the formula that considers the difference between the midpoint of sleep (MS) on free days and workdays (ΔMS): ΔMS=MSF–MSW (MSF=MS on free days; MSW=MS on work or college days) [21]. The social jetlag was calculated in hours, with values ranging from 0 to 12 hours [21].
Participants also answered the validated Brazilian version of the Epworth Sleepiness Scale (ESS) [22], which assesses the degree of daytime sleepiness through the probability of a person fall asleep during daily activities. The score ranges from 0 to 24 points, and scores greater than 10 indicate excessive daytime sleepiness [22].
Finally, to assess sleep quality, volunteers responded to the Pittsburgh Sleep Quality Index (PSQI) in the version validated for the Brazilian population [23]. The score ranges from 0 to 21, and a score above five classifies the person as a “bad sleeper” [23].
After completing these initial questionnaires, the volunteers were instructed to fill out an online sleep diary for 7 consecutive days, in which they recorded data of bedtime, wake-up time, as well as subjective sleep quality and the number of perceived awakenings [24].
During this week, they also responded to a food diary, in which they reported what they ate and drank during the day, in detail, including time and quantity. The description was made in an online form, which was sent daily by the research team, and notes could be made right after meals or at the end of the day, to reduce memory errors. Although the literature recommendation is to apply a food diary for 3 days, including the weekend [25], the need to apply it for a full week is justified to investigate the correlation between diet and sleep. The collected data was used to calculate nutrients and energy ingestion, and data were analyzed using the Dietbox software® (Dietbox).
Statistical Analysis
Descriptive statistics were calculated for all variables. Quantitative data are presented as mean and standard deviation and qualitative data as absolute and relative frequency. The Kolmogorov-Smirnov test and Levene test were used to analyse data normally and homogeneity, respectively. To evaluate the difference in the general sample characteristics (age, body mass, height, and body mass index [BMI]) between the sexes, the Mann-Whitney test was applied.
To compare the data from the sleep diary and food diary between the seven days of application, the Friedman test was used, with the Conover post hoc test to identify significant differences between the different groups. To analyse the correlation between social jetlag and chronotype, daytime sleepiness, and PSQI data, the Spearman correlation test was applied. This same test was also applied to analyse the correlation between food consumption and sleep diary data on the 7 evaluated days, as well as to evaluate the correlation between the average macronutrient intake on the seven days evaluated and sleep data (jetlag social, chronotype, daytime sleepiness, and sleep quality [PSQI]). The differences of baseline characteristics and sleep data (age, sex, chronotype, body mass, height, BMI, social jetlag, sleep quality, sleepiness, and chronotype) were compared between participants who completed the dietary diaries (n=16) and those who did not (n=47) using Mann-Whitney U test. The JASP program (JASP Team, version 0.17.1.10) was used for all analyses and the significance level was set at p<0.05.
RESULTS
Initially, 90 volunteers agreed to participate in the study; however, 27 were excluded for not meeting the inclusion criteria. The final sample comprised 63 undergraduate students (40 females), all of whom were enrolled in evening academic programs, with classes starting between 6:30 p.m. and 7:00 p.m. and ending between 9:00 p.m. and 11:00 p.m.
Most participants commuted to college by public transportation (primarily by bus), and 33 individuals (51.56%) reported commuting directly from work to campus. Most students lived with their families (n=43, 67.19%), and a similar proportion (n=43, 67.19%) engaged in regular physical activity, predominantly resistance training.
The average reported weight, height, BMI, and other sleep characteristics are presented in Table 1. A significant difference in body mass and height was observed between sexes, with males presenting higher values than females (p<0.001). Despite these differences, no significant difference in BMI was found between sexes. Table 1 also presents data on sleep characteristics. No significant differences were observed between males and females in chronotype, social jetlag, daytime sleepiness, or sleep quality (p>0.05).
Anthropometric and sleep characterization
In the PSQI questionnaire, the median classified the sample with good sleep. Most of the students were classified as “good sleepers” (n=36, 56%), with the other 28 (44%) having poor sleep quality (or “bad sleepers”). The average sleep efficiency (PSQI) indicated that, in general, participants assess that they have efficient sleep, not staying in bed for a long time before starting to sleep.
Although most students had good sleep, the median of the sample classified participants with excessive daytime sleepiness (>10 score) (median of 12; with results between 8.00 and 15.00), with 36 (56%) classified with excessive daytime sleepiness.
Regarding the MEQ questionnaire, the majority were classified as the intermediate chronotype (n=33, 51.56%), followed by morningness (n=21, 32.81%) and afternoon (n=10, 15.62%) chronotype.
Results from the MCTQ are described in Table 2. Considering that no significant differences were found between males and females, variations in sleep midpoint, bedtimes, wake-up times, and sleep duration between workdays and free days were analysed regardless of sex. Sleeping and waking times on work-days are different from those on free days. On free days, participants woke up later (p<0.001), slept longer (p<0.001), and had a MS later than on workdays (p<0.001). This difference between the MS on work and free days reflects a median social jetlag of 1.57 (0.65; 2.35), with social jetlag values varying between 0.05 and 4.02 in the sample. Most students (59%) exhibited a social jetlag greater than 1 hour.
Munich questionnaire results
There was a weak negative correlation between social jetlag and chronotype (MEQ questionnaire, p=0.009 and r=−0.33), indicating that the more afternoon the chronotype, the higher the individual’s social jetlag and the circadian misalignment (Fig. 1). There was no correlation between social jetlag and daytime sleepiness (p=0.52 and r=0.08), sleep quality score (p=0.52 and r=0.08) and with sleep efficiency according to the PSQI (p=0.66 and r= 0.06).
Correlation between social jetlag and chronotype score (A), sleepiness score (B), PSQI score (C), and sleep efficiency according to PSQI (D). *p≤0.05. MEQ, Morningness-Eveningness Questionnaire; ESS, Epworth Sleepiness Scale; PSQI, Pittsburgh Sleep Quality Index; EFIC-PSQI, sleep efficiency according to the PSQI.
In addition to the questionnaires, participants also completed sleep and food diaries for seven consecutive days. Considering the complexity and frequency of answers, only 16 participants completed this phase. Baseline characteristics and sleep data (age, sex, chronotype, body mass, height, BMI, social jetlag, sleep quality, sleepiness, and chronotype) did not differ significantly (all p>0.05) between participants who completed the dietary diaries (n=16) and those who did not (n=47).
Data from the sleep diary are described in Table 3. There was no significant difference in bedtime between the 7 days (p=0.24), but the wake-up time was significantly different between the week (p<0.01). On Saturday (6th day), the sample woke up later than in the other 6 days (p<0.01). The total sleep time (TST) was significantly different between the week (p=0.05), with TST on Saturday being longer than on Monday (p=0.02), Tuesday (p= 0.004), Wednesday (p=0.01), and Sunday (p=0.02). There was no significant difference in sleep latency (p=0.32) and in the number of awakenings (p=0.79) between the seven days.
Sleep parameters according to sleep diary (n=16)
Concerning data from food diaries (Table 4), it is noted that although the energy intake (kcal) appears to be higher on weekends, especially on Saturdays, compared to weekdays, there was no significant difference between the 7 days (p=0.24). While the consumption of macronutrients also appears to have been higher on weekends, there was no significant difference between the consumption of carbohydrates (p=0.56), protein (p=0.91), and lipids (p=0.43) among the 7 evaluated days.
Nutritional intake during the week (n=16)
There was a moderate negative correlation between protein (p<0.001 and r=−0.85), lipid (p=0.006; r=−0.66), and energy ingestion (p=0.04; r=−0.52) and sleep latency on day 1, indicating that greater nutrient consumption was associated with shorter sleep latency (Fig. 2). On the 4th day, a positive correlation was also observed between bedtime and carbohydrate (p=0.01; r= 0.60), lipid (p=0.004 and r=0.68), and energy intake (p=0.002 and r=0.72), in addition to a negative correlation between sleep latency and lipid (p=0.05 and r=−0.50) and energy intake (p=0.04 and r=−0.53), suggesting that those with a higher lipid and energy intake sleep later and has a lower sleep latency (Fig. 3). There was no significant correlation between macronutrients/energy intake and data from the sleep diary on the other days evaluated.
Correlation between sleep latency and carbohydrate (A), protein (B), lipid (C), and energy (D) intake on day 1. *p≤0.05.
Correlation between sleep parameters and nutritional intake on day 4. Correlation between bedtime and nutritional intake (A–D), and sleep latency and nutritional intake (E–H) on day 4. *p≤0.05.
To evaluate the relationship between social jetlag, sleep quality and efficiency (PSQI), daytime sleepiness, and chronotype with diet, the average nutritional intake over seven days was calculated. There was a negative correlation between social jetlag and the mean of carbohydrate intake (p=0.05 and r=−0.49), protein (p=0.03 and r=−0.57), and energy intake (p=0.03 and r= −0.55), suggesting that higher intake of these macronutrients and energy was associated with lower social jetlag (Fig. 4). When dietary intake was analysed separately for weekdays and weekends, significant negative correlations with social jetlag were found only for weekday consumption. Social jetlag was significantly correlated with mean of carbohydrate (p=0.01 and r=−0.62), protein (p=0.02 and r=−0.72), and energy (p=0.02 and r=−0.59) intake during weekdays, whereas no significant correlations were observed with weekend averages (all p>0.05). There was no significant correlation between nutritional intake and chronotype, daytime sleepiness, sleep quality and efficiency (p>0.05).
Correlation between social jetlag and carbohydrate (A), protein (B), lipid (C), and energy (D) intake. *p≤0.05.
DISCUSSION
The objective of this study was to evaluate the correlation between chronotype and social jetlag among undergraduate students, as well as to investigate potential associations between nutrient intake and sleep patterns. Our results revealed a weak negative correlation between social jetlag and chronotype, suggesting that individuals with a more evening-oriented chronotype tend to experience higher levels of social jetlag and circadian misalignment. Regarding the relationship between nutrition and sleep, we found significant isolated correlations on 2 days; however, overall, no consistent or significant association emerged between energy/macronutrient intake and sleep diary data across the week. This limited association may reflect sample size constraints or variability in participants’ daily behaviors.
The study highlighted distinct differences in sleep-wake patterns between workdays and free days. On free days, participants typically woke up later, had longer sleep durations, and experienced later midpoints of sleep. These findings reflect the presence of social jetlag, with some participants experiencing over four hours of discrepancy between their biological and social clocks. According to prior literature, social jetlag greater than one hour is associated with elevated risks of obesity and metabolic disorders [5]. Notably, 59% of our sample fell into this risk category, warranting attention to their long-term health implications.
Moreover, social jetlag has been negatively associated with physical activity levels [26], acting as a barrier to regular exercise and potentially fostering sedentary lifestyles that are detrimental to health. Most of our participants displayed an intermediate chronotype (51.56%), a finding consistent with population-level data [20]. Extreme chronotypes, while less common, often face more challenges adapting to socially imposed schedules, leading to more pronounced sleep disturbances.
The observed negative correlation between social jetlag and chronotype may stem from the tendency of evening types to delay sleep onset, only to be woken prematurely by alarm clocks on workdays. Consequently, they compensate with longer sleep durations on free days. This pattern aligns with Roenneberg et al. [2] and Caliandro et al. [6], who report that over 70% of working or studying individuals exhibit at least one hour of social jetlag. Industrialized societies, with rigid social schedules, exacerbate this misalignment between biological and societal times.
Our PSQI findings indicated that the sample had generally good sleep quality, with no participants classified as having a sleep disorder. This may partly be attributed to our exclusion criteria, which filtered out individuals with known sleep disorders. Sleep efficiency was high (90.09%), exceeding the recommended threshold (>85%) [27], possibly linked to regular physical activity reported by 67.19% of the sample.
One limitation of this study was the relatively small number of participants who completed both sleep and food diaries, likely due to the time-intensive nature of daily reporting. This small sample size limits the reliability of the correlations and reduces the generalizability of our findings. However, baseline comparisons indicated no significant differences between completers and non-completers, suggesting limited risk of attrition bias. Therefore, the results should be interpreted with caution and considered preliminary. Nevertheless, the use of diaries enhances data accuracy compared to retrospective self-report questionnaires.
Analysis of the sleep diary data revealed consistent bedtimes but significantly different wake times, with Saturday showing the latest wake times and longest TST. This aligns with previous findings suggesting that lack of alarm clock use on weekends allows for longer, more restorative sleep [28].
With respect to dietary intake, no significant weekly variation in energy or macronutrient consumption was observed. However, a moderate negative correlation was found between protein intake and sleep latency on day 1, suggesting that higher protein consumption may promote faster sleep onset. This could be due to tryptophan’s role in serotonin and melatonin synthesis [29]. On day 4, higher intake of carbohydrates, lipids and energy correlated with later bedtimes and shorter sleep latency, suggesting that dietary patterns may influence circadian rhythms and sleep initiation. These results partially diverge from the literature, which links high fat intake with poorer sleep efficiency and high carbohydrate intake with better sleep quality [29,30].
Interestingly, our hypothesis that greater social jetlag would be associated with increased energy intake was not supported. Contrary to expectations and previous findings [31,32], a negative correlation was found with energy, carbohydrate and protein. Although this was not the expected finding, Yoshizaki and Togo [33] also reported a negative correlation between social jetlag and energy intake, whereas other studies did not find significant associations [34,35]. One hypothesis is that this discrepancy may be attributed to sample size and the limitations of dietary assessment instruments, as underreporting or overreporting by participants may occur. Another possible explanation is the presence of university restaurants (UR), which offer balanced, nutritionist-planned meals during weekdays, potentially regulating students’ dietary habits and mitigating excess caloric consumption [36]. Although we did not assess which participants ate at the UR during the evaluation period, the UR was available to all students. Previous studies have reported that undergraduates who eat at the UR consume more fruits and vegetables during the week compared to weekend due to the availability of these food groups in the restaurant [36,37], which could impact nutritional intake. It is important to emphasize that this hypothesis remains speculative, and future studies should include an evaluation of the place where meals were consumed.
The present study did not observe significant correlations between chronotype and nutritional intake. Although such associations were expected, Yoshizaki and Togo [33] likewise did not identify relationships with energy, carbohydrate, or lipid intake. Instead, they reported negative correlations with protein intake and diet quality, as well as with the consumption of specific food groups, suggesting that greater social jetlag was associated with less healthy eating habits [33]. The absence of correlations in the present study may have been influenced by several factors, such as socioeconomic and cultural conditions, individual preferences, or even food availability. However, to better understand these aspects, a qualitative analysis of dietary patterns would be necessary. Future studies examining associations between chronotype and diet should therefore address not only quantitative but also qualitative aspects of food intake.
Although some correlations between dietary intake and sleep parameters were observed on specific days, these findings were not corrected for multiple comparisons and should be considered exploratory. They may reflect preliminary trends rather than robust associations, and further studies with larger samples are needed to confirm these results.
As with any self-report study, the accuracy of participant responses may be influenced by recall bias. Additionally, the limited sample size restricts generalizability. Despite these limitations, the study’s strengths include the use of validated instruments and daily data collection on sleep and nutrition, offering a more nuanced understanding of individual patterns.
In conclusion, social commitments appear to significantly influence biological rhythms, with free days allowing participants to align sleep patterns more closely with their natural chronotypes. The majority of participants experienced social jetlag exceeding one hour, posing potential health risks. Evening chronotypes were associated with greater social jetlag. No consistent relationships emerged between chronotype, social jetlag, and sleep quality. However, some associations between macronutrient intake and sleep parameters suggest that diet may influence sleep, although further investigation is needed. Given that only a small proportion of participants completed both the sleep and food diaries, these findings should be interpreted with caution and considered preliminary. Nevertheless, they underscore the importance of aligning social demands with biological needs and call for future studies to explore the interplay between circadian rhythms, sleep, and dietary patterns in greater depth.
Notes
Availability of Data and Material
The datasets generated or analyzed during the study are available from the corresponding author on reasonable request.
Author Contributions
Conceptualization: all authors. Data curation: Giovanna Antonella Martineli Rodrigues, Natália Vilela Silva Daniel. Formal analysis: all authors. Funding acquisition: Giovanna Antonella Martineli Rodrigues, Andrea Maculano Esteves. Investigation: Giovanna Antonella Martineli Rodrigues. Methodology: all authors. Project administration: Natália Vilela Silva Daniel, Andrea Maculano Esteves. Supervision: Andrea Maculano Esteves. Validation: all authors. Visualization: all authors. Writing—original draft: Giovanna Antonella Martineli Rodrigues, Natália Vilela Silva Daniel. Writing—review & editing: all authors.
Conflicts of Interest
The authors have no potential conflicts of interest to disclose.
Funding Statement
This study was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo - FAPESP (grant nº 2022/14006-0).
Acknowledgements
The authors would like to thank the participants in the research, the Funding agency, Centro de Pesquisa em Ciências do Esporte (CEPECE) and Laboratório do Sono e Exercício Físico (LASEF).