The interplay between sleep and mood in predicting academic functioning, physical health and psychological health: A longitudinal study

https://doi.org/10.1016/j.jpsychores.2012.08.014Get rights and content

Abstract

Objectives

Existing studies on sleep and behavioral outcomes are mostly correlational. Longitudinal data is limited. The current longitudinal study assessed how sleep duration and sleep quality may be causally linked to daytime functions, including physical health (physical well‐being and daytime sleepiness), psychological health (mood and self-esteem) and academic functioning (school grades and study effort). The mediation role of mood in the relationship between sleep quality, sleep duration and these daytime functions is also assessed.

Methods

A sample of 930 Chinese students (aged 18–25) from Hong Kong/Macau completed self-reported questionnaires online across three academic semesters. Sleep behaviors are assessed by the Sleep Timing Questionnaire (for sleep duration and weekday/weekend sleep discrepancy) and the Pittsburgh Sleep Quality Index (sleep quality); physical health by the World Health Organization Quality of Life Scale—Brief Version (physical well‐being) and Epworth Sleepiness Scale (daytime sleepiness); psychological health by the Depression Anxiety Stress Scale (mood) and Rosenberg Self-esteem Scale (self-esteem) and academic functioning by grade-point-average and the College Student Expectation Questionnaire (study effort).

Results

Structural equation modeling with a bootstrap resample of 5000 showed that after controlling for demographics and participants' daytime functions at baseline, academic functions, physical and psychological health were predicted by the duration and quality of sleep. While some sleep behaviors directly predicted daytime functions, others had an indirect effect on daytime functions through negative mood, such as anxiety.

Conclusion

Sleep duration and quality have direct and indirect (via mood) effects on college students' academic function, physical and psychological health. Our findings underscore the importance of healthy sleep patterns for better adjustment in college years.

Introduction

College students face multiple challenges, such as intellectual demands and identity formation. Furthermore, their sleep behaviors have been characterized by sleep deprivation, poor sleep quality and excessive daytime sleepiness [1]. Growing evidence suggests poor sleep patterns are related to impaired academic performance [2], physical health [3] and psychological well‐being [4]. Yet, the temporal relationships among sleep and these functional outcomes are unclear.

The relationship between sleep, mood and other psychological functions has lately become a rapidly developing research area. While negative mood, such as depression and anxiety, has long been identified as harming nighttime sleep, recent findings show that the relationship between sleep and mood is bi-directional [5]. Neuropsychological evidences suggest that both quality and quantity of sleep are vital to the optimal functioning of brain activity in regulating our emotions [5]. Fredriksen and coworkers [4] provided longitudinal data to show that among adolescents, sleep loss was a significant predictor of increased depressive feeling and self-esteem. In fact, Walker and Harvey [6] argued that while sleep and mood are unquestionably linked, future studies should assess how exactly they are related and the outcomes of such interaction/relation.

Sleep and mood also affect school performance. Kelly, Kelly and Clanton [7] found a positive correlation between hours of sleep and school grades. Students with ≥ 8 h of sleep reported an average grade-point-average (GPA) of 3.24 compared to an average GPA of 2.74 for those with < 7 h of sleep. Psycho-physiological research indicated that sleep is crucial for the consolidation and reactivation of memory [8]. Sleep-deprived participants have greater difficulty than healthy controls in recalling learned materials [9]. Good sleep quality was also associated with higher learning motivation and school performance. Gomes, Tavares and de Azevedo [10] recently demonstrated that both sleep duration and quality are significant predictors of school grades among undergraduates. While other evidence suggests that mood is related to school grades and motivation [11], how mood mediates the relationship between sleep behaviors and academic functioning remains to be determined.

A healthy sleep pattern has been shown to relate to desirable health conditions. Sleep duration and quality are suggested to closely relate to daytime sleepiness which reflects one's inability to sustain attention [1]. Daytime sleepiness has been used as an indicator of the health status in both patient and healthy population. For instance, excessive daytime sleepiness is seen as a cardinal symptom in sleep apnea and it is correlated with increasing medical problems in an otherwise healthy population [12]. Apart from daytime sleepiness, sufficient sleep is also demonstrated to predict health conditions, such as blood pressure [13] and body mass index (BMI) [3]. Chang and coworkers [14] found that among cancer patients' caregivers, poor self-reported sleep quality predicted dissatisfaction with physical health. While sleep behaviors appear to correlate with physical health, limited studies compared the relative contribution of different sleep behaviors in predicting health conditions.

Although increasing evidence suggests that sleep behaviors are closely linked with mood and functional outcomes (including academic functioning, physical and psychological health), causal relationships cannot be established without longitudinal or experimental evidence [2]. The current study aims primarily to explore the temporal relationships from sleep behaviors to the aforementioned functional outcomes. We also intend to explore if sleep affects daytime functions through inducing negative mood. In other words, we plan to test the mediating role of mood between sleep behaviors and daytime functions. While some sleep behaviors were shown to have differential roles in predicting daytime functions in previous cross-sectional studies, we aim to compare the relative contribution of interdependent sleep behaviors in predicting the outcome measures with a longitudinal structural equation modeling (SEM) approach. With the use of SEM, all regression pathways can be tested at once and comparisons of strengths of pathways can be made accordingly.

Section snippets

Participants

Chinese undergraduates, aged 18–25 from 16 universities and colleges in Hong Kong and Macau were recruited through campus flyers, emails and online platforms. Of the 1195 participants who completed the measurements in Time 1, 1006 (84.2%) continued in Time 2 and 930 (77.8% of Time 1) in Time 3.

Procedures

The current investigation was a sub-study of a large-scale longitudinal research program on the formation and transformation of beliefs, lifestyle, and well‐being in Chinese. Ethics approval was obtained

Results

Details of the demographic information are summarized in Table 2. The final sample, with mean age of 21.7 (SD = 2.22), was composed of second-year students (45.2%), with 10.3% and 44.5% in their first- and third-year or above, respectively. There were 307 males and 619 females. Table 3 summarized the correlational data and showed that there were less than 5% data missing among all measured variables. The Little's MCAR test suggested that the data missing were completely at random, χ2(1584) = 

Discussion

The current study primarily aimed to investigate the directional relationships between different sleep behaviors, mood and daytime functions, including physical health (daytime sleepiness and physical well‐being), psychological health (mood and self-esteem) and academic functioning (school grades and study effort). Our findings showed that not all, but some specific sleep behaviors may directly or indirectly (through inducing negative mood) predict the aforementioned daytime functions

Conclusion

Taken together, the current study strengthens the claim that specific domains of sleep behaviors can directly or indirectly (through mood) predict physical and psychological health and academic functioning. The current findings call attention to the need for colleges to raise students' awareness of the relationships between their sleep, mood, and academic performance and to provide sleep hygiene education. Besides testing the predictive/mediating roles of different sleep behaviors in affecting

Conflict of interest statement

This is not an industry‐supported study. The authors have indicated no financial conflicts of interest.

Acknowledgments

We thank Ms. Jasmine Lam for her assistance in data collection.

References (33)

  • R. Stickgold

    Sleep-dependent memory consolidation

    Nature

    (2005)
  • A. Karni et al.

    Dependence on REM sleep of overnight improvement of a perceptual skill

    Science

    (1994)
  • A.A. Gomes et al.

    Sleep and academic performance in undergraduates: a multi-measure, multi-predictor approach

    Chronobiol Int

    (2011)
  • E.O. Bixler et al.

    Excessive daytime sleepiness in a general population sample: the role of sleep apnea, age, obesity, diabetes, and depression

    J Clin Endocrinol Metab

    (2005)
  • M. Bursztyn et al.

    Siesta and ambulatory blood pressure monitoring. Comparability of the afternoon nap and night sleep

    Am J Hypertens

    (1994)
  • E.W.-C. Chang et al.

    Quality of sleep and quality of life in caregivers of breast cancer patient

    Psychooncology

    (2007)
  • Cited by (171)

    • The relationships among sleep problems, anxiety, memory complaints and compulsive checking behaviours

      2022, Journal of Obsessive-Compulsive and Related Disorders
      Citation Excerpt :

      A total score of PSQI>5 indicates poor global sleep quality. Apart from using the PSQI global score, we also examined specific dimensions, including subjective sleep quality, sleep disturbances and daytime dysfunction, as other dimensions are either better captured by sleep diary and actigraphy (e.g. sleep latency, sleep duration and sleep efficiency) or uncommon among college students (e.g. use of sleep medication) (Wong et al., 2013). Similar to previous studies (Gelaye et al., 2014; Manzar et al., 2018; Qiu, Gelaye, Zhong, Enquobahri, Frederick, & Williamse, 2016), we found a rather low internal consistency of the PSQI (Cronbach's alpha, a = 0.538) in the present study, suggesting that it does not fit into a unidimensional structure.

    View all citing articles on Scopus
    1

    These authors contributed equally to this work.

    View full text