We examined the relationship between chronotype and depression/ somatic symptoms in South Korean adolescents. About 38% of the total variation in depression and somatic symptoms was attributable to among-subject differences. The mean depression and somatic symptom scores slightly increased over the 3 years of follow-up.
In prospective analyses, it was found that individuals with a late chronotype were more likely to experience somatic symptoms and depression during the current year, and this tendency also predicted the presence of somatic symptoms one year later. Specifically, among females, a late chronotype was associated with experiencing depression and somatic symptoms within the same year, while among males, it predicted the occurrence of somatic symptoms one year later. Additionally, having less than 7 hours of sleep amplified the negative impact of a late chronotype on depression in the subsequent year.
In concurrent analyses, a positive relationship was observed between having a late chronotype and experiencing depression. Furthermore, the duration of sleep on weekdays was found to moderate the negative impact of a late chronotype on somatic symptoms. This effect was significant for both insufficient and excessive sleep durations. Stratification analysis revealed that regardless of gender, excessive sleep amplified the adverse effects of a late chronotype on somatic symptoms.
Chronotype influences adolescent life through various aspects such as affect, eating habits, academic performance, and other health behaviors. Specifically, individuals with a late chronotype tend to exhibit less positive affect compared to those with an early chronotype (43). Late chronotype is linked to increased evening energy intake, and a delay in chronotype is associated with a higher likelihood of regularly skipping breakfast (44). They are at risk of having a high BMI and adopting poor dietary patterns, including the consumption of unhealthy snacks, nighttime caffeine intake, and insufficient daily intake of fruits and vegetables (45). Furthermore, an early chronotype is correlated with better academic performance compared to a late chronotype(46). A late chronotype has a negative impact on school performance, which is mediated by daytime sleepiness and low learning motivation(47). Finally, students with early and intermediate chronotypes are less likely to engage in smoking, alcohol use, and physical inactivity (48). Hence, it is crucial to take into account adolescents' chronotype in various aspects of their lives.
There are several possible explanations for the relationship between chronotype and mental health. First, there may be differences in brain structure and function according to chronotype. Localized atrophy in the right hippocampus in late-chronotype individuals may result in prolonged exposure to glucocorticoids and stress (49). Late-chronotype people show enhanced emotional reactivity to negative stimuli, altered reward processing, and impaired emotional regulation (50). Second, genetic differences linked to the late chronotype are linked to mental illness. A CLOCK variant and PER3 polymorphism related to the late chronotype are associated with more depressive cognition and higher trait-anxiety scores, respectively (50). Sleep disorders such as difficulties in sleep initiation, inadequate sleep duration, and poor sleep quality resulting from the late chronotype can contribute to affective symptoms such as depression and anxiety (22). Regarding behavioral aspects, late-sleeping adolescents are less likely to be exposed to bright light, to be physically active, and to socialize with friends and family, possibly leading to a less positive mood (51). Self-blame and reduced positive reappraisal are associated with the late chronotype (52).
The interaction between chronotype and sleep duration has been investigated. Patterson et al. showed that inadequate sleep and the late chronotype are more closely related to cardiovascular risk factors (e.g., tobacco use, sedentary behavior, and less consumption of fruit and vegetables) than adequate sleep and the early chronotype (53). The late chronotype and insufficient sleep have a positive interaction effect on the levels of inflammatory cytokines (54). Indeed, chronotype and short sleep duration are independently related to high depressive symptoms (33). Insufficient sleep duration has a negative effect on mental health, and the late chronotype is linked to sleep disorders (e.g., short sleep duration). Therefore, sleep duration moderates the relationship between chronotype and somatic symptoms.
Long and short sleep durations have negative effects on mental health. Specifically, in the stratification analysis, all the interaction effects observed were associated with long sleep rather than short sleep. According to Grandner et al. (55), weekday sleep duration has a u-shaped relationship with sleep complaints (e.g., difficulty falling asleep, waking during night, daytime sleepiness), which means that people who sleep < 7 h or > 9 h have lower sleep quality than those who sleep for 8 h. In addition, very short and long sleep durations are related to major depressive disorder (56). Therefore, both long and short sleep duration are related to low sleep quality and possibly to poor mental health.
Late chronotype was positively correlated with concurrent depression and prospectively associated with somatic symptoms. This contrasts with previous reports that indicated a concurrent and prospective relationship between late chronotype and depression(17, 21, 52). Nevertheless, the reliability of the results is strengthened by the larger sample size in this study compared to previous ones. Conversely, there has been limited exploration of the relationship between chronotype and somatic symptoms in existing literature. Typically, somatic symptoms are closely associated with anxiety and depression. Consequently, further research is warranted to delve into the relationship between chronotype and somatic symptoms, along with investigating the underlying mechanisms.
The stratification analysis provided valuable insights. In the gender-based stratification analysis, a significant relationship between late chronotype and depression was observed only in females. This aligns with another study, which attributes such differences to variations in melatonin activity between men and women, as well as differences in sex hormones(57). Additionally, the observed gender-specific association between late chronotype and depression in females may be attributed to factors such as lower sleep quality(58) and depressogenic cognitive responses, which are more prevalent in women than men(59).
To alleviate the mental health problems that late chronotype adolescents may face, various interventions have been proposed. These include light interventions, delaying school starting times, and rearranging examination schedules(60). Light intervention involves increasing light exposure in the morning and reducing it at night, although its effectiveness has not been extensively proven in previous research studies(60). Modifying school starting times appears to be a crucial intervention, considering the strong association between early school schedules and sleep deprivation, as well as daytime sleepiness, even for individuals with early chronotypes(61). A study randomly assigned 753 students to start school in the morning (07:45), afternoon (12:40), or evening (17:20). As a result, early chronotypes showed improved performance in the morning class, late chronotypes performed better in the evening class, while no differences were observed in the afternoon class (62).
Finally, if adjusting school starting times is not feasible, an alternative approach could be to delay examination times to minimize the discrepancy in academic performance between chronotypes. In contrast to morning examinations (8:15 − 12:15), where early chronotypes achieved significantly higher grades than late chronotypes, early afternoon examinations (12:45 − 15:00) showed no group differences(63). This indicates that, for both school starting time and examination time, the afternoon appears to be the most suitable for all chronotypes. However, it is essential to tailor this scheduling approach based on school grades, considering that chronotypes tend to shift to later preferences as students age. Although students may be better aligned with afternoon schedules in the early school years, evening schedules might be the best fit during later periods(64).
The strength of this study lies in its ability to identify both concurrent and prospective relationships between chronotype and mental health using highly representative longitudinal panel data. Notably, the less-known association between chronotype and somatic symptoms was discovered. Furthermore, stratification analysis revealed distinct variations in the relationship based on gender. The identification of a significant interaction effect of weekday sleep duration provided a potential mechanism for this relationship. These findings offer valuable insights applicable to interventions for adolescent mental health care. Importantly, the robustness of the results persisted even after adjusting for potential confounding factors such as smartphone usage and breakfast consumption.
This study had several limitations. First, the mechanisms underlying the relationship between chronotype and somatic symptoms are unclear; these could be similar to those for depression and anxiety, which have comparable symptoms. In addition, self-reported data have a risk of recall bias. There could also be reporting bias for some covariates such as delinquency and smartphone dependency. Selection bias should be considered because nonresponders and those lost to follow-up could have different characteristics than subjects who completed the study. If the number of late-chronotype individuals were underestimated, there is a risk of bias towards the null hypothesis. Moreover, the data used to determine chronotype were not obtained from verified measures such as the MEQ or MCTQ. However, we used data on awake/asleep hours, which are used to define chronotypes. Finally, because the study was based on the data of Korean adolescents, the generalizability of the findings to the general population is unclear.