Social Support and Daytime Sleepiness Among Chinese Medical Students: Mediating Roles of Loneliness and Problematic Smartphone Use

Introduction

Good quality sleep is a crucial indicator of human health, and sleep issues have become a widespread public health problem.¹ Research on daytime sleepiness is relatively limited, despite some efforts that have been made to understand sleep problems in diverse populations.^2–5 Daytime sleepiness is a common phenomenon and physiological state in which individuals desire to sleep during the day.⁶ Daytime sleepiness is also an indicator of numerous physical and mental health problems (eg, insufficient sleep, insomnia, and psychological distress).^7–9 Based on the significant impact of daytime sleepiness on an individual’s physical and mental health and wellbeing, there has been increasing concern about daytime sleepiness. Most studies on daytime sleepiness have focused on populations with sleep-related breathing disorders^2,10,11 or those secondary to somatic and psychiatric disorders.^12–15 Several studies have also shown a high prevalence of daytime sleepiness, poor sleep quality, and poor sleep health among medical students,^9,16 nurses,¹⁷ and the general population.¹⁸ However, these studies are only the beginning regarding the prevalence of daytime sleepiness.

Over the past decade, sleep issues have been the most detected mental health problem among Chinese college students.¹⁹ Daytime sleepiness among medical students adversely affects their cognitive function,²⁰ academic performance,²¹ professional performance in clinical practice,²² and social interactions,²³ and is a risk factor for a number of mental disorders.^24,25 According to the social-ecological model of sleep, societal, social, and individual factors affect sleep health. At the individual level, attitudes, psychology, health behaviors, and other factors influence sleep health. Home, neighborhood, work/school, socioeconomics, and other factors contribute to the social level, while globalization, technology, public policy, and others are related elements at the societal level. The individual level is embedded within the social level, which, in turn, is embedded within the societal level.²⁶ These societal and social/interpersonal factors, filtered through the social structure, ultimately influence sleep at the individual level.⁶

Social support in social/interpersonal-level factors is characterized as an individual’s perception or experience of affection, concern, value, belonging, or assistance in connection with others,²⁷ which is frequently evaluated by the structure of social networks or the function of network members.^27,28 According to the social-ecological model, social support can influence the emotions and health-related behaviors of individuals,^29,30 which could affect individual sleep health. Moreover, there is also an association between social support and daytime sleepiness.³¹ Research on the relationship between social support and daytime sleepiness among nurses and college students in recent years shows that social support is negatively related to daytime sleepiness.^17,32 Despite this evidence about the adverse effect of poor social support on daytime sleepiness, questions regarding the mediating factors that contribute to this association remain unanswered.³³ In light of the critical role of social support for physical and mental health and the significant impact of daytime sleepiness on medical students, it is essential to investigate the association between social support and daytime sleepiness to elucidate important mediating mechanisms that could serve as potential intervention targets. This might also provide a new perspective to explore the pathways in which social relationships affect sleep health.

Loneliness and problematic smartphone use are important indicators of mental health problems among medical students.³⁴ Loneliness is commonly referred to as feelings of depression, melancholy, poor spirits, and emptiness,³⁵ which is a state of unmet individual or social-emotional needs.^36,37 Medical students are in a crucial developmental period, building interpersonal relations and acquiring a sense of belonging, but loneliness has substantial adverse effects on their health and wellbeing. Previous studies have explored loneliness and daytime sleepiness, but the findings were inconsistent. Earlier research found that high levels of loneliness affect daily life and are associated with poor sleep quality, increasing daytime sleep symptoms.³⁸ While Kurina et al found that loneliness was a significant predictor of sleep fragmentation in adults, they did not find an association between loneliness and daytime sleepiness. Recently, an increasing number of studies have shown that the relationship between loneliness and sleep health is bidirectional, with increased loneliness contributing to the continuation of a vicious cycle of poor sleep quality, prolonged loneliness, and excessive daytime sleepiness. Individuals exposed to this vicious cycle may continue to suffer from high loneliness, poor sleep quality, and excessive daytime sleepiness.^39,40 Social support, a significant indicator of social relationships, is regarded as the most critical component that can effectively alleviate and predict the developmental trajectory of loneliness.⁴¹ However, there is a lack of research on mediating the relationship between loneliness on social support and daytime sleepiness.

Problematic smartphone use is a common health behavior problem among college students, described as a cognitive, emotional, and behavioral pattern of excessive use of smartphones that negatively impacts the individual’s daily life.⁴² Theoretically, the compensatory model of internet use explains this relationship.^43–45 In reality, perceived stress and difficulties, and the lack of social support may prompt individuals to use an escape-coping strategy to satisfy their needs or relieve their negative emotions through a convenient online world, which can increase their problematic smartphone use behavior and lead to overindulgence. Social support has been found to be a protective factor for individual development and health.⁴⁶ Among the consequences of problematic smartphone use, most research has focused on sleep disorders.⁴⁵ Previous findings suggest that adolescents with lower perceived social support have problematic smartphone use behaviors and significantly higher severity of depressive symptoms and daytime sleepiness.⁴⁷ The findings may support the hypothesis that problematic smartphone use might mediate the association between social support and daytime sleepiness. Therefore, the roles of loneliness and problematic smartphone use in the relationship between social support and daytime sleepiness should be explored more broadly and thoroughly.

Furthermore, loneliness may influence individual decisions to use the internet for gratification, which, in turn, may affect problematic internet use.⁴⁸ According to the Interaction of Person-Affect-Cognition-Execution (I-PACE) model for problematic and addictive behaviors, adolescents with high levels of loneliness might use online social relationships to satisfy their social-emotional needs.⁴⁹ A longitudinal association also exists between loneliness and problematic smartphone use.^50,51 A study of Chinese college students found that those with lower levels of loneliness were less likely to experience problematic smartphone use after 6 and 18 months.⁵¹ Recent research on adolescents also revealed that loneliness mediates the association between perceived social support from family and friends and problematic smartphone use.⁴⁶ These findings might provide support for the hypothesis that loneliness and problematic smartphone use might sequentially mediate the association between social support and daytime sleepiness. Therefore, further research is needed to determine the causal relationships between these variables with the aim of exploring the pathways in which social relationships affect sleep health.⁴¹

We aimed to explore the interrelationship between social support, loneliness, problematic smartphone use, and daytime sleepiness among medical students, as well as to test whether loneliness and problematic smartphone use play a sequential mediation role in the relationship between social support and daytime sleepiness. Based on the literature review, a theoretical hypothesis model was established, as shown in Figure 1. We hypothesized that: (1) loneliness would play a mediating role in the relationship between social support and daytime sleepiness among medical students; (2) problematic smartphone use would play a mediating role in the relationship between social support and daytime sleepiness among medical students; and (3) loneliness and problematic smartphone use would serially mediate the relationship between social support and daytime sleepiness among medical students.

Figure 1 Hypothesized model for the present study. Notes: Social support, loneliness, and problematic smartphone use were assumed to have direct impacts on daytime sleepiness. Social support was also suggested to affect daytime sleepiness via loneliness only, or via problematic smartphone use only, or via loneliness and problematic smartphone use.

Materials and Methods

Participants

Between November 3 and 14, 2022, we conducted a series of self-reported questionnaires with undergraduate medical students recruited from Chinese Medical University. We determined the estimated sample size to be 383 through the following formula: , using a power of 80%, a confidence level of 95%, and a margin of error of 5%.^52,53 Accounting for potential invalid data (ie, up to 15% invalid), the minimum number of participants was determined to be 450. We distributed a total of 466 questionnaires. All participants entered the study voluntarily with any incentive and could withdraw at any time, and all participants provided informed consent before completing the measures. Exclusion criteria for all participants included self-reported sedative medication or recreational drug use; self-reported suffering from central disorders of hypersomnolence, sleep-related breathing disorders, circadian rhythm sleep-wake disorders, and excessive sleepiness secondary to other somatic diseases and mental disorders. A total of 455 students completed the questionnaires, 53 invalid responses were excluded, and 402 (86.27%) were effectively recovered. The sample consisted of 175 male and 227 female students aged 19–25 years (mean: 22.488 ± 1.828 years), including 43 first-year, 59 second-year, 98 third-year, 133 fourth-year, and 69 fifth-year undergraduate medical students. The Human Research Ethics Committee of China Medical University approved this study (ref no. 2023126).

Measures

Social Support

The Social Support Rate Scale (SSRS) developed by Xiao for the Chinese population^54,55 was used to comprehensively examine the social support of medical students. The SSRS contains 10 items and is divided into three dimensions (objective support, subjective support, and use of social support). Higher total scores reflect a higher level of social support. Cronbach’s α was 0.781 in the present study.

Daytime Sleepiness

Daytime sleepiness was assessed by the Epworth Sleepiness Scale (ESS),⁵⁶ containing eight different scenarios commonly encountered in daily life (eg, “Sitting and reading”), in which participants estimate their likelihood of falling asleep or dozing off. Each item is scored on a 4-point Likert scale (0 = never, 3 = high chance). The sum of the eight-item scores yields a total score that ranged from 0 to 24. A total score between 0 and 10 is considered normal, and a score greater than 10 is considered excessive daytime sleepiness.^56,57 In this study, Cronbach’s α of ESS was 0.739.

Loneliness

Loneliness was measured using the Chinese version of the UCLA loneliness scale⁵⁸ developed by Russell.⁵⁹ This measure consists of 20 items (eg, “Do you often feel lonely?”), including 11 positive and nine negative scoring items. All items are scored on a 5-point Likert scale (1 = never, 5 = always), with higher total scores reflecting higher levels of loneliness. Cronbach’s α in the present study was 0.908.

Problematic Smartphone Use

The current study utilized the Mobile Phone Addiction Index (MPAI) to evaluate problematic smartphone use,⁶⁰ of which is widely employed in the Chinese college student population.^61,62 The MPAI consists of 17 items that examine four domains, including “inability to control craving”, “feeling anxious and lost”, “withdrawal/escape”, and “productivity loss” (eg, “Your friends and family complained about your use of the mobile phone”). All items are scored on a 5-point Likert scale (1 = not at all, 5 = always). Higher total scores indicate higher levels of problematic smartphone use. In this study, MPAI (Cronbach’s α = 0.910) showed good internal consistency.

Data Analysis

Statistical Package for Social Sciences (SPSS 28.0, SPSS Inc., Chicago, IL) and Hayes’s PROCESS Macro were utilized to analyze our data. First, demographic characteristics were analyzed using descriptive statistics, and the scores for each measure were expressed as means and standard deviations. Counting data was reported as percentages (%). Examining the histograms and analyzing the skewness and kurtosis values of the variables in the study revealed that the data were approximately normally distributed. We examined the relationship between social support, daytime sleepiness, loneliness, and problematic smartphone use by Pearson correlation analysis. After adjusting the demographic characteristics of the participants, a multiple linear regression model was developed to determine the association between social support, daytime sleepiness, loneliness, and problematic smartphone use. Second, we evaluated the sequentially mediating effects of loneliness and problematic smartphone use in the association between social support and daytime sleepiness. Hayes’s PROCESS Macro (http://www.afhayes.com) was specially developed to assess the complex models, including mediators and moderators.⁶³ To test the hypothesized model (Figure 1), we conducted mediation model analyses using PROCESS Macro (Models 6).⁶⁴ In the mediation model, social support was the independent variable; daytime sleepiness was the dependent variable; and loneliness and problematic smartphone use were the first and second mediators, respectively. Moreover, age, gender, and grade were controlled for in the mediation model because prior evidence has shown that they are important confounders in daytime sleepiness.^26,65,66 All study variables were standardized before the mediation model analyses. Based on 5000 bootstrap samples, the confidence intervals (CIs) determined whether the effects were statistically significant.⁶⁴ The 95% bias-corrected CIs did not contain zero, indicating that the effect was significant. P-values < 0.05 were considered statistically significant.

Results

Common Method Bias

Considering that the data for the current study were collected from participant self-reports, there may have been a common method bias. Based on the research on confidentiality and the reverse scoring of some items, we examined the data for common method bias employing Harman’s single-factor test. Without rotation, the results indicated 10 factors with a characteristic root greater than 1, and the variance shown by the first factor was 22.141% (< 40%). Therefore, the results demonstrated that there was no significant common method bias in the present study.

Descriptive Statistics and Correlational Analysis

Table 1 shows the means, standard deviations (SDs), and correlation matrix for several crucial research variables. There was a negative correlation between social support and daytime sleepiness (r = −0.360, p < 0.01), loneliness (r = −0.164, p < 0.01), and problematic smartphone use (r = −0.224, p < 0.01). Loneliness was positively correlated with daytime sleepiness (r = 0.448, p < 0.01) and problematic smartphone use (r = 0.391, p < 0.01). There was a positive correlation between problematic smartphone use and daytime sleepiness (r = 0.380, p < 0.01).

Table 1 Means, Standard Deviations, and Correlations of the Main Study Variables (N = 402)

Multiple Linear Regression

The collinearity analysis of social support, daytime sleepiness, loneliness, and problematic smartphone use demonstrated that the tolerance values ranged from 0.81 to 0.97 and that variance inflation factor (VIF) values were around 1.00, with daytime sleepiness serving as the dependent variable. Daytime sleepiness was used as the dependent variable, and hierarchical regression was utilized. After controlling demographic variables (age, gender, and grade), the variables of social support, loneliness, and problematic smartphone use were gradually added to form four models, as displayed in Table 2. All crucial research variables showed a linear regression relationship (β = −0.266, 0.331, and 0.188, respectively; all p-values < 0.001).

Table 2 Multiple Linear Regression of Daytime Sleepiness in Medical Students

Testing for Sequential Mediation Effect

A bias-corrected percentile bootstrap test was used to analyze the sequential mediating effects of loneliness and problematic smartphone use on social support and daytime sleepiness. Age, gender, and grade were taken as covariates. The pathway analysis of social support, daytime sleepiness, loneliness, and problematic smartphone use revealed that all direct associations (social support to loneliness, problematic smartphone use, and daytime sleepiness; loneliness to problematic smartphone use and daytime sleepiness; problematic smartphone use to daytime sleepiness) were significant (Figure 2 and Table 3). There were significant indirect effects observed in the relationship between social support and daytime sleepiness via loneliness only (β = −0.037; 95% CI = −0.062, −0.015), problematic smartphone use only (β = −0.022; 95% CI = −0.041, −0.008), or loneliness and problematic smartphone use sequentially (β = −0.008; 95% CI = −0.016, −0.002). These results indicate that loneliness and problematic smartphone use were mediators of the association between social support and daytime sleepiness (Table 3).

Table 3 Bootstrapping Results of the Sequential Mediation Model on Daytime Sleepiness

Figure 2 Results of the sequential mediation model describing the mediator roles of loneliness and problematic smartphone use in the association between social support and daytime sleepiness. Notes: Mediated effects of daytime sleepiness on the associations between social support, loneliness, and problematic smartphone use. All models used 5000 bootstrapping resamples and controlled for age, gender, and grade of participants. All coefficients and effects are unstandardized coefficients/effects. All path coefficients were significant. **p < 0.01, ***p < 0.001.

Discussion

Loneliness and problematic smartphone use might play an independent or sequential mediating role between social support and daytime sleepiness. To our knowledge, this is the first study investigating the relationships between social support, loneliness, problematic smartphone use, and the severity of daytime sleepiness symptoms among medical students.

Based on the study population, our primary findings are as follows: First, the direct effect measured in the current study indicated that lower levels of social support were directly associated with higher levels of daytime sleepiness among medical students. Similar results were found recently, indicating that the social support level of female shift nurses was significantly and negatively associated with daytime sleepiness and insomnia,¹⁷ and that the level of social support among college students was a predictor of daytime sleepiness.³² Moreover, our findings subscribed to the social-ecological theoretical model. Low levels of social support (social-level risk factors) have the potential to heighten individual levels of stress and anxiety.⁶⁷ When confronted with the various obstacles and adversities of life, individuals may experience increased levels of anxiety and a diminished ability to effectively manage these challenges. This can be attributed to an absence of social support and a limited interpersonal network of dependable individuals with whom they can confide and seek solace.⁶⁸ Consequently, these circumstances may impede the process of falling asleep and aggravate individual insufficient sleep or sleep-related problems.²⁶

Second, we examined the effects of loneliness and problematic smartphone use on the association between social support and daytime sleepiness symptoms. Our main findings were that loneliness and problematic smartphone use had serial mediating roles in the relationship between social support and daytime sleepiness, and social support was associated with daytime sleepiness symptoms by increasing loneliness levels and problematic smartphone use behaviors, thus supporting our hypothesis. These findings are consistent with a previous study conducted among 1056 Chinese adolescents, which reported that loneliness mediated the relationship between social support and problematic smartphone use.⁴⁶ Also, the results of a study conducted among Turkish university students found that loneliness predicted problematic smartphone use.⁶⁹ In the present study, social support had significant indirect effects on problematic smartphone use through loneliness. Medical students with lower levels of social support may suffer from social isolation, and they are more likely to experience loneliness.^34,70 This also supports the I-PACE model,⁴⁹ namely that medical students with higher levels of loneliness are more prone to utilizing online social interactions to meet their social-emotional and social relationship needs, and are more at risk of exhibiting problematic smartphone use behaviors, emotional problems, and poor sleep health.^71,72

Third, our results suggested the mediating role of loneliness in the relationship between social support and daytime sleepiness. Specifically, lower levels of social support are associated with higher levels of loneliness, which in turn indirectly affect the development of daytime sleepiness symptoms. This may be considering loneliness is normally accompanied by emotional stress and anxiety in individuals, both of which can increase emotional insecurity at night, making it more difficult to fall asleep and promoting increased daytime fatigue and daytime sleepiness.⁷³ Similarly, Grey et al found that exposure to lower levels of social support were associated with higher levels of loneliness and depression, which ultimately led to more severe sleep problems.⁷⁴ Furthermore, we explored the role of problematic smartphone use in the relationship between social support and daytime sleepiness. Problematic smartphone use mediated the relationship between social support and daytime sleepiness symptoms. A recent meta-analysis has found similar results, which reported that excessive smartphone use among nursing students was associated with lower levels of perceived social support sleep quality.⁷⁵ Meanwhile, a study in China investigated that the detection rate of smartphone addiction among 2741 medical students was as high as 52.8%, and medical students with smartphone addiction had higher levels of insomnia and subhealth.⁷⁶ Additionally, our findings support the compensatory model of internet use.^43–45 Medical students with low levels of social support are prone to escape and gratify their needs or alleviate their negative emotions via the online world, which may increase their problematic smartphone use behaviors and impact their physical and mental health. Therefore, social support may also have indirect effects on daytime sleepiness through medical student loneliness or problematic smartphone use only.

Sleep health among medical students is an important issue. To our knowledge, this is the first study to investigate the mediating effects of loneliness and problematic smartphone use in the relationship between social support and daytime sleepiness. This makes it possible to explore the pathways in which social relationships affect sleep health through a sequential mediation model between these variables. Meanwhile, a socio-ecological model of sleep provides a framework to consider the dynamic interplay among individual, social, and societal factors.¹ Interventions based on social/interpersonal-level factors (ie, social support) and individual level factors (ie, loneliness) can improve outcomes (ie, problematic smartphone use and daytime sleepiness). Family and group supportive therapies, as well as cognitive behavioral therapy, have been demonstrated to be beneficial.^77–79 The social and environmental determinants of sleep health are commonly beyond the scope of individual behavior.¹ Therefore, it is crucial to think about multilevel prevention strategies to improve sleep health when conducting intervention research.⁸⁰

This study also had several limitations. First, we could not provide evidence of causality due to the cross-sectional design, which does not account for temporal effects. Future studies should use a longitudinal design to establish and validate causal relationships. Second, all variables used in the current study were evaluated based on self-reports from the participants. Future studies utilizing multiple methods to collect data should be conducted, such as objective monitoring to measure daytime sleepiness levels and diagnostic assessment by a clinical specialist to exclude participants with sleep disorders. Third, all the participants in this study were medical students from a college in northeastern China. Consequently, the findings may not be applicable to other professional populations and cannot be generalized to other countries. Future research should recruit a larger, more representative sample to validate our findings. Finally, measures of sleep duration and blue light-emitting device usage before bedtime were not obtained, which have been suggested to influence the onset of daytime sleepiness.^81,82 Therefore, there is a residual confounding risk from unmeasured covariates. In future studies, it is recommended to include sleep duration and blue light-emitting device usage before bedtime as covariates.

Conclusions

The present study provides evidence of factors mediating the relationship between daytime sleepiness and social support among medical students. Specifically, we describe two mediating factors in the aforementioned relationship. For medical students, social/interpersonal-level factors had a significant impact on their sleep health. Additionally, social support affected daytime sleepiness, and this relationship was mediated by loneliness and problematic smartphone use. Further research is essential to confirm the effects of these factors on daytime sleepiness. Furthermore, our study provided evidence that loneliness and problematic smartphone use could be used as interventional entry points to improve daytime sleepiness among medical students. Interventions targeting loneliness and health-related behaviors among medical students may mitigate the effects from low levels of social support, and further improve their daytime functions.